{
  "schema_version": 4,
  "study_role": "existing_references",
  "formal_base": {
    "method": null,
    "reason": "既存方式の比較。現自作ベースを測定・選定基準にしない"
  },
  "comparison_reference": "world_harvest",
  "selected_methods": [
    "world_harvest",
    "facodec",
    "hifiglot",
    "golf",
    "sifigan",
    "llsm2",
    "magphase"
  ],
  "optional_omissions": [
    "dac"
  ],
  "method_labels": {
    "identity": "原音",
    "world_harvest": "WORLD（Harvest）",
    "facodec": "FACodec",
    "hifiglot": "HiFi-Glot",
    "golf": "GOLF-ff",
    "sifigan": "SiFi-GAN",
    "llsm2": "LLSM2",
    "magphase": "MagPhase"
  },
  "native_sample_rates": {
    "world_harvest": 16000,
    "facodec": 16000,
    "hifiglot": 44100,
    "golf": 24000,
    "sifigan": 24000,
    "llsm2": 24000,
    "magphase": 48000
  },
  "index_methods": [
    "identity",
    "world_harvest",
    "facodec",
    "hifiglot",
    "golf",
    "sifigan",
    "llsm2",
    "magphase"
  ],
  "generated_at": "2026-10-12T11:39:13.503994+00:00",
  "scope": "2音声の実行確認・品質採否には使用不可",
  "planned_cases": 2,
  "contract": {
    "schema_version": 1,
    "methods": [
      "world_harvest",
      "facodec",
      "hifiglot",
      "golf",
      "sifigan",
      "llsm2",
      "magphase"
    ],
    "inputs": [
      {
        "id": "smoke_arctic",
        "sample_rate": 44100,
        "samples": 147734,
        "pcm_sha256": "bd719b730a7005fc41842c648e42de53638f4d1a1f791be8da1a831d38988742"
      },
      {
        "id": "smoke_ready",
        "sample_rate": 44100,
        "samples": 220500,
        "pcm_sha256": "035587c5437df50083bc61d5cbb58f9b30924e025a3a5b4adcc2d515f61fc7cd"
      }
    ],
    "scope": "smoke",
    "seed": 0,
    "threads": 1,
    "device": "cpu",
    "mos_requested": false,
    "source_hashes": {
      "uv.lock": "c8b2bc28ebb1523bf8cfe0f4627d06bd5d03e0ba6a2bb5dc0905be77a3dc8823",
      "vocoderbench/extended_metrics.py": "ab3537d2c8e5d568a0cd2fa8aa36309de09c3bce7cf2f849214331a22fe3624f",
      "vocoderbench/metrics.py": "b8767feb49494058bd9ec08fd95ae2fb50ce7897fcdd6cebe319ff8a0ac51f22",
      "vocoderbench/mos.py": "fb632aa5440245525289962372db4f3eb007cd2b3b493b2d05ecccb878e85f03",
      "vocoderbench/reference.py": "72cebd1ee7c185ac3a6cdf451451b0ca21ba21cb0f52ffe29e324c1fdc56ae55",
      "vocoderbench/references/__init__.py": "d96022954c3cd493c7e6d7baf35e8bd33eafea192a9f9154c8470b7937d05732",
      "vocoderbench/references/adapters_dsp.py": "ab5a0cf63085524842a0614d2ebc63adbe11521852f53bf9d561a131c37ab285",
      "vocoderbench/references/adapters_neural.py": "cfe845d99cf0257556da99e5a1f950c9d9ac5ac490fb7d8e8cfb9b46f2c03a08",
      "vocoderbench/references/adapters_source.py": "28b70e6680a3c8090fa5d16f696eae5c28d6b55bee5eee9e90ef4b80735f2b99",
      "vocoderbench/references/catalog.py": "ac325b15312caa6e888033a909cb7487b49de288d7f88e5af5891cb95fe6797c",
      "vocoderbench/references/common.py": "d56c6b113aa1d6da8bc4a126103ca14b42caa3d73a01c009e26e0d1f71e7c7c7",
      "vocoderbench/references/evaluate.py": "16cb052b8e7c10c6e0b6d91f5d810a77c9376ae6664395f9dac23c72b2c425bc",
      "vocoderbench/references/inputs.py": "623d7d8ffeba294d66a7a000ddf608e3cb48f7b018f2a4ebd39e236394127f6b",
      "vocoderbench/references/integration.py": "d64e036964d4b6ef782d69b927d5d3b1ca62aedca22e2979255d0eea861a1930",
      "vocoderbench/references/native.py": "40c1efba52953f5b87f1d372e30fd08cdb1ae9e45a7bf69aecd9c8cfab2b7365",
      "vocoderbench/references/prepare.py": "82092abb7f8e950b0c09b08568580627e6696796d054a716824e038ed80699d4",
      "vocoderbench/references/report.py": "1adf53aa3e28224e3be3ad96441603d7204f9dd3ccfafde210359758d750ba9b",
      "vocoderbench/references/requirements.txt": "52fe25d537c36ec24ca7ba6a00e6856cfb540c3d776d0af928ac03ecd06042d0",
      "vocoderbench/references/sources.lock.json": "a9545775f1381f6d56e390286384993ab8f1b6c3b3010058d5fc54bccef494c1",
      "vocoderbench/references/worker.py": "bcaa7902ec63856da3ca7a32a71f9dd4658d25277c8eee72bdbcc4664b977df2",
      "vocoderbench/temporal_metrics.py": "ce21b6a4767071a7b98ea0cdbfe54bdba15b596bf8d6b6127281a97095e7be91",
      "vocoderbench/world.lock.json": "44b8bc3adb62802ed060834d9e06ed7f5f3cc4d28583b8a44066c5959a42462b",
      "vocoderbench/world.py": "d14558eed772e07031a9183f57facc55acb1c27deccb3dd4fa098ffcbe0cde3a"
    },
    "prepared_methods": {
      "world_harvest": {
        "schema_version": 1,
        "method": "world_harvest",
        "environment": {
          "python": "3.10.22 (main, Oct  3 2026, 01:32:06) [Clang 22.1.3 ]",
          "platform": "Linux-6.17.0-1022-azure-x86_64-with-glibc2.39",
          "dependencies": "absl-py==2.5.1\naiohappyeyeballs==2.7.1\naiohttp==3.14.5\naiosignal==1.4.0\nantlr4-python3-runtime==4.9.3\napprise==2.0.1\nargbind==0.3.9\nasttokens==3.0.2\nasync-timeout==5.0.1\nattrs==26.1.0\naudioread==3.1.0\nbeautifulsoup4==4.15.0\ncertifi==2026.7.22\ncffi==2.1.1\ncharset-normalizer==3.5.2\nclick==8.5.0\ncloudpickle==3.1.2\ncontourpy==1.3.2\ncycler==0.12.1\ncython==3.3.0\ndecorator==5.3.1\ndescript-audiotools==0.7.2\ndiffsptk==4.0.1\ndocstring-parser==0.18.0\neinops==0.8.2\neinx==0.4.3\nexceptiongroup==1.3.1\nexecuting==2.3.0\nffmpy==1.0.0\nfilelock==3.32.3\nfire==0.7.1\nflatten-dict==0.5.0\nfonttools==4.65.0\nfrozendict==2.4.7\nfrozenlist==1.8.0\nfsspec==2024.12.0\nfuture==1.0.0\ngdown==6.4.2\ngrpcio==1.84.0\nh5py==3.16.0\nhf-xet==1.7.0\nhuggingface-hub==0.36.2\nhydra-core==1.3.7\nidna==3.20\nimportlib-resources==7.1.0\nipython==8.39.0\njedi==0.20.0\njinja2==3.1.6\njoblib==1.6.0\njsonargparse==4.52.0\njulius==0.2.8\nkazane==1.0.4\nkiwisolver==1.5.1\nlazy-loader==0.6\nlibrosa==0.10.2.post1\nlightning==2.1.3\nlightning-utilities==0.15.3\nllvmlite==0.50.0\nloguru==0.7.3\nmarkdown==3.10.3\nmarkdown-it-py==4.2.0\nmarkdown2==2.5.5\nmarkupsafe==3.0.3\nmatplotlib==3.10.9\nmatplotlib-inline==0.2.2\nmdurl==0.1.2\nmpmath==1.3.0\nmsgpack==1.2.3\nmultidict==7.1.0\nnetworkx==3.4.2\nnnaudio==0.3.4\nnumba==0.68.0\nnumpy==1.26.4\noauthlib==4.0.0\nomegaconf==2.3.1\npackaging==24.2\nparso==0.8.7\npenn==0.0.14\npexpect==4.9.0\npillow==12.3.0\nplatformdirs==4.12.4\npooch==1.9.0\npraat-parselmouth==0.4.7\nprompt-toolkit==3.0.53\npropcache==0.5.4\nprotobuf==3.19.6\npsutil==7.2.2\nptyprocess==0.7.0\npure-eval==0.2.4\npycparser==3.11\npygments==2.21.0\npyloudnorm==0.2.0\npylstraight==0.1.3\npyparsing==3.3.3\npysocks==1.7.1\npysptk==1.0.1\npystoi==0.4.1\npython-dateutil==2.9.0.post0\npytorch-lightning==2.6.6\npyworld==0.3.5\npyyaml==6.0.3\nrandomname==0.2.1\nrequests==2.34.2\nrequests-oauthlib==2.0.0\nrich==15.0.0\nsafetensors==0.8.0\nscikit-learn==1.7.2\nscipy==1.13.1\nsetuptools==80.10.2\nsix==1.17.0\nsoundfile==0.14.0\nsoupsieve==2.10\nsoxr==1.1.0\nstack-data==0.6.3\nsympy==1.13.1\ntensorboard==2.20.0\ntensorboard-data-server==0.7.2\ntensorboardx==2.6\ntermcolor==3.3.0\nthreadpoolctl==3.7.0\ntorch==2.5.1+cpu\ntorch-einops-utils==0.1.36\ntorch-fftconv==1.0.0\ntorch-stoi==0.2.3\ntorchaudio==2.5.1+cpu\ntorchcomp==0.2.1\ntorchlpc==0.6\ntorchmetrics==1.9.0\ntorchutil==0.1.2\ntqdm==4.70.1\ntraitlets==5.16.1\ntyping-extensions==4.16.0\nurllib3==2.8.0\nvector-quantize-pytorch==1.31.6\nwcwidth==0.9.2\nwerkzeug==3.1.9\nyapecs==0.5.0\nyarl==1.25.1\n"
        },
        "lock_sha256": "a9545775f1381f6d56e390286384993ab8f1b6c3b3010058d5fc54bccef494c1",
        "requirements_sha256": "52fe25d537c36ec24ca7ba6a00e6856cfb540c3d776d0af928ac03ecd06042d0",
        "files": {
          "../../vocoderbench/.cache/world/c794856b537758a3d300757db2e5e31685a822c2/libworld.so": "aa36c83c669323a941ed326a802e5aba630d0765742679e9c19f8411c3753760"
        },
        "sources": {},
        "note": "上流コードと重みをキャッシュへ取得。第三者コードを本リポジトリに複製しない。"
      },
      "facodec": {
        "schema_version": 1,
        "method": "facodec",
        "environment": {
          "python": "3.10.22 (main, Oct  3 2026, 01:32:06) [Clang 22.1.3 ]",
          "platform": "Linux-6.17.0-1022-azure-x86_64-with-glibc2.39",
          "dependencies": "absl-py==2.5.1\naiohappyeyeballs==2.7.1\naiohttp==3.14.5\naiosignal==1.4.0\nantlr4-python3-runtime==4.9.3\napprise==2.0.1\nargbind==0.3.9\nasttokens==3.0.2\nasync-timeout==5.0.1\nattrs==26.1.0\naudioread==3.1.0\nbeautifulsoup4==4.15.0\ncertifi==2026.7.22\ncffi==2.1.1\ncharset-normalizer==3.5.2\nclick==8.5.0\ncloudpickle==3.1.2\ncontourpy==1.3.2\ncycler==0.12.1\ncython==3.3.0\ndecorator==5.3.1\ndescript-audiotools==0.7.2\ndiffsptk==4.0.1\ndocstring-parser==0.18.0\neinops==0.8.2\neinx==0.4.3\nexceptiongroup==1.3.1\nexecuting==2.3.0\nffmpy==1.0.0\nfilelock==3.32.3\nfire==0.7.1\nflatten-dict==0.5.0\nfonttools==4.65.0\nfrozendict==2.4.7\nfrozenlist==1.8.0\nfsspec==2024.12.0\nfuture==1.0.0\ngdown==6.4.2\ngrpcio==1.84.0\nh5py==3.16.0\nhf-xet==1.7.0\nhuggingface-hub==0.36.2\nhydra-core==1.3.7\nidna==3.20\nimportlib-resources==7.1.0\nipython==8.39.0\njedi==0.20.0\njinja2==3.1.6\njoblib==1.6.0\njsonargparse==4.52.0\njulius==0.2.8\nkazane==1.0.4\nkiwisolver==1.5.1\nlazy-loader==0.6\nlibrosa==0.10.2.post1\nlightning==2.1.3\nlightning-utilities==0.15.3\nllvmlite==0.50.0\nloguru==0.7.3\nmarkdown==3.10.3\nmarkdown-it-py==4.2.0\nmarkdown2==2.5.5\nmarkupsafe==3.0.3\nmatplotlib==3.10.9\nmatplotlib-inline==0.2.2\nmdurl==0.1.2\nmpmath==1.3.0\nmsgpack==1.2.3\nmultidict==7.1.0\nnetworkx==3.4.2\nnnaudio==0.3.4\nnumba==0.68.0\nnumpy==1.26.4\noauthlib==4.0.0\nomegaconf==2.3.1\npackaging==24.2\nparso==0.8.7\npenn==0.0.14\npexpect==4.9.0\npillow==12.3.0\nplatformdirs==4.12.4\npooch==1.9.0\npraat-parselmouth==0.4.7\nprompt-toolkit==3.0.53\npropcache==0.5.4\nprotobuf==3.19.6\npsutil==7.2.2\nptyprocess==0.7.0\npure-eval==0.2.4\npycparser==3.11\npygments==2.21.0\npyloudnorm==0.2.0\npylstraight==0.1.3\npyparsing==3.3.3\npysocks==1.7.1\npysptk==1.0.1\npystoi==0.4.1\npython-dateutil==2.9.0.post0\npytorch-lightning==2.6.6\npyworld==0.3.5\npyyaml==6.0.3\nrandomname==0.2.1\nrequests==2.34.2\nrequests-oauthlib==2.0.0\nrich==15.0.0\nsafetensors==0.8.0\nscikit-learn==1.7.2\nscipy==1.13.1\nsetuptools==80.10.2\nsix==1.17.0\nsoundfile==0.14.0\nsoupsieve==2.10\nsoxr==1.1.0\nstack-data==0.6.3\nsympy==1.13.1\ntensorboard==2.20.0\ntensorboard-data-server==0.7.2\ntensorboardx==2.6\ntermcolor==3.3.0\nthreadpoolctl==3.7.0\ntorch==2.5.1+cpu\ntorch-einops-utils==0.1.36\ntorch-fftconv==1.0.0\ntorch-stoi==0.2.3\ntorchaudio==2.5.1+cpu\ntorchcomp==0.2.1\ntorchlpc==0.6\ntorchmetrics==1.9.0\ntorchutil==0.1.2\ntqdm==4.70.1\ntraitlets==5.16.1\ntyping-extensions==4.16.0\nurllib3==2.8.0\nvector-quantize-pytorch==1.31.6\nwcwidth==0.9.2\nwerkzeug==3.1.9\nyapecs==0.5.0\nyarl==1.25.1\n"
        },
        "lock_sha256": "a9545775f1381f6d56e390286384993ab8f1b6c3b3010058d5fc54bccef494c1",
        "requirements_sha256": "52fe25d537c36ec24ca7ba6a00e6856cfb540c3d776d0af928ac03ecd06042d0",
        "files": {
          "models/naturalspeech3_facodec/ns3_facodec_encoder.bin": "c8fa7b8d79ccd915e6c3b19cd3c9d5f4cba3fbeb794a2e598d7e31db8efee47e",
          "models/naturalspeech3_facodec/ns3_facodec_decoder.bin": "4b8f2c9a691385c1a61542585116f52263a029758dcb4a3a955ff2c190afd536",
          "src/facodec/ns3_codec/__init__.py": "6b86c557f111bb6c0c1c43fe77bf34ac16ad973411d652fda63449f112cb24c5",
          "src/facodec/ns3_codec/alias_free_torch/__init__.py": "7e8bdb5ab07a79de67d1490600964f60b0a995c6691217f4fcfb76e175a0b79a",
          "src/facodec/ns3_codec/alias_free_torch/act.py": "2add4414a8dec8b5ffe55e55dc26c17aeba31224b5a5fb3959cc82804454b558",
          "src/facodec/ns3_codec/alias_free_torch/filter.py": "674e5fafffd4bc46b58a7de3590b3bc6c2228930e70b50f4ceea014367438dce",
          "src/facodec/ns3_codec/alias_free_torch/resample.py": "9341fe6756d3c8222ecbd4155366ee12aefab8c98ab0cd9badc34a0c6d7eb755",
          "src/facodec/ns3_codec/facodec.py": "086962bbe3b0913491bf1d5874b9005afc8f3a6b7f2707f385e5711f0df84140",
          "src/facodec/ns3_codec/gradient_reversal.py": "8d727bbb012ee5275ad4c823ac747a86db88a69faf393b04b9a46c7408c85ebf",
          "src/facodec/ns3_codec/melspec.py": "6a14da2a4d531f1e8fdb3f83398bfb80c379f5fa41162a28742d6b9dd31b0c3f",
          "src/facodec/ns3_codec/quantize/__init__.py": "69964f5c0ecf93a00b2b4767cc2952cad19e7eebbc7bbd154214c01fc34c2943",
          "src/facodec/ns3_codec/quantize/fvq.py": "40a1efbbcedd5b09fff3db1d91d831114dba49a63d4f58215892d3fe5dbf1c10",
          "src/facodec/ns3_codec/quantize/rvq.py": "e761c3b3f6130e91546830ff4308111ebd23a9e9a7c7443204a358399d98cfd4",
          "src/facodec/ns3_codec/transformer.py": "1cf20b2262cd3dd754b4884f9a134ced9b0ce416332382c00e75b41ef6401f12",
          "src/facodec/setup.py": "1cc676afa26edc2e6d2d0babf2864f787f8940e7d510f74e537f21b45bd1b0e9",
          "src/facodec/test.py": "a4e50e05663a335615c07a93e0586d36a64eaa7fe8cccd317ac95caea882d90a"
        },
        "sources": {
          "facodec": {
            "repository": "https://github.com/lifeiteng/naturalspeech3_facodec.git",
            "revision": "aac16b7268b1b96e70b9e3526b13e216984e80d0"
          }
        },
        "note": "上流コードと重みをキャッシュへ取得。第三者コードを本リポジトリに複製しない。"
      },
      "hifiglot": {
        "schema_version": 1,
        "method": "hifiglot",
        "environment": {
          "python": "3.10.22 (main, Oct  3 2026, 01:32:06) [Clang 22.1.3 ]",
          "platform": "Linux-6.17.0-1022-azure-x86_64-with-glibc2.39",
          "dependencies": "absl-py==2.5.1\naiohappyeyeballs==2.7.1\naiohttp==3.14.5\naiosignal==1.4.0\nantlr4-python3-runtime==4.9.3\napprise==2.0.1\nargbind==0.3.9\nasttokens==3.0.2\nasync-timeout==5.0.1\nattrs==26.1.0\naudioread==3.1.0\nbeautifulsoup4==4.15.0\ncertifi==2026.7.22\ncffi==2.1.1\ncharset-normalizer==3.5.2\nclick==8.5.0\ncloudpickle==3.1.2\ncontourpy==1.3.2\ncycler==0.12.1\ncython==3.3.0\ndecorator==5.3.1\ndescript-audiotools==0.7.2\ndiffsptk==4.0.1\ndocstring-parser==0.18.0\neinops==0.8.2\neinx==0.4.3\nexceptiongroup==1.3.1\nexecuting==2.3.0\nffmpy==1.0.0\nfilelock==3.32.3\nfire==0.7.1\nflatten-dict==0.5.0\nfonttools==4.65.0\nfrozendict==2.4.7\nfrozenlist==1.8.0\nfsspec==2024.12.0\nfuture==1.0.0\ngdown==6.4.2\ngrpcio==1.84.0\nh5py==3.16.0\nhf-xet==1.7.0\nhuggingface-hub==0.36.2\nhydra-core==1.3.7\nidna==3.20\nimportlib-resources==7.1.0\nipython==8.39.0\njedi==0.20.0\njinja2==3.1.6\njoblib==1.6.0\njsonargparse==4.52.0\njulius==0.2.8\nkazane==1.0.4\nkiwisolver==1.5.1\nlazy-loader==0.6\nlibrosa==0.10.2.post1\nlightning==2.1.3\nlightning-utilities==0.15.3\nllvmlite==0.50.0\nloguru==0.7.3\nmarkdown==3.10.3\nmarkdown-it-py==4.2.0\nmarkdown2==2.5.5\nmarkupsafe==3.0.3\nmatplotlib==3.10.9\nmatplotlib-inline==0.2.2\nmdurl==0.1.2\nmpmath==1.3.0\nmsgpack==1.2.3\nmultidict==7.1.0\nnetworkx==3.4.2\nnnaudio==0.3.4\nnumba==0.68.0\nnumpy==1.26.4\noauthlib==4.0.0\nomegaconf==2.3.1\npackaging==24.2\nparso==0.8.7\npenn==0.0.14\npexpect==4.9.0\npillow==12.3.0\nplatformdirs==4.12.4\npooch==1.9.0\npraat-parselmouth==0.4.7\nprompt-toolkit==3.0.53\npropcache==0.5.4\nprotobuf==3.19.6\npsutil==7.2.2\nptyprocess==0.7.0\npure-eval==0.2.4\npycparser==3.11\npygments==2.21.0\npyloudnorm==0.2.0\npylstraight==0.1.3\npyparsing==3.3.3\npysocks==1.7.1\npysptk==1.0.1\npystoi==0.4.1\npython-dateutil==2.9.0.post0\npytorch-lightning==2.6.6\npyworld==0.3.5\npyyaml==6.0.3\nrandomname==0.2.1\nrequests==2.34.2\nrequests-oauthlib==2.0.0\nrich==15.0.0\nsafetensors==0.8.0\nscikit-learn==1.7.2\nscipy==1.13.1\nsetuptools==80.10.2\nsix==1.17.0\nsoundfile==0.14.0\nsoupsieve==2.10\nsoxr==1.1.0\nstack-data==0.6.3\nsympy==1.13.1\ntensorboard==2.20.0\ntensorboard-data-server==0.7.2\ntensorboardx==2.6\ntermcolor==3.3.0\nthreadpoolctl==3.7.0\ntorch==2.5.1+cpu\ntorch-einops-utils==0.1.36\ntorch-fftconv==1.0.0\ntorch-stoi==0.2.3\ntorchaudio==2.5.1+cpu\ntorchcomp==0.2.1\ntorchlpc==0.6\ntorchmetrics==1.9.0\ntorchutil==0.1.2\ntqdm==4.70.1\ntraitlets==5.16.1\ntyping-extensions==4.16.0\nurllib3==2.8.0\nvector-quantize-pytorch==1.31.6\nwcwidth==0.9.2\nwerkzeug==3.1.9\nyapecs==0.5.0\nyarl==1.25.1\n"
        },
        "lock_sha256": "a9545775f1381f6d56e390286384993ab8f1b6c3b3010058d5fc54bccef494c1",
        "requirements_sha256": "52fe25d537c36ec24ca7ba6a00e6856cfb540c3d776d0af928ac03ecd06042d0",
        "files": {
          "models/HiFi-Glot/hifiglot/config.json": "d8af1f8f4207d3d56a7959ebff54d0645b9557256d518002869bdfeefa691c87",
          "models/HiFi-Glot/hifiglot/epoch-0175_step-1070000/model.safetensors": "8955e6f43593099b912c25e2e5f9abd2120950e2927b34c5aaebbe3df9eb14c4",
          "src/hifiglot/configs/baseline/config_feature_map.json": "43187c3e97d1a45be976fa43b7d2fe21630e5a01f1670b0859b2bcd3449f4cbd",
          "src/hifiglot/configs/baseline/config_hifigan.json": "d8af1f8f4207d3d56a7959ebff54d0645b9557256d518002869bdfeefa691c87",
          "src/hifiglot/configs/hifiglot/config_feature_map.json": "d293e790bd06fd217bd24f9a80251cd5a7ad08c209bcf9a3e0076267096d9f87",
          "src/hifiglot/configs/hifiglot/config_hifigan.json": "d8af1f8f4207d3d56a7959ebff54d0645b9557256d518002869bdfeefa691c87",
          "src/hifiglot/dataset/filelist/AIShell.json": "6ef6e4bec7077e063a24854613f23732d405df2b688cf8398b156cb25e6c8910",
          "src/hifiglot/dataset/train.json": "d44f55a104301b14fe8f914ee035f5472d96ba8fd6079b7613fd73c75e7a4f06",
          "src/hifiglot/inference_baseline.py": "438ddf835f2a75001a0d7f154aef966c96a86cc3dd5b8e52902a8a3fc1c04fba",
          "src/hifiglot/inference_hifiglot.py": "fa317b8301d26cc1f5d88f4c72f50d2366a7bf31113549ee935fad18a5940d99",
          "src/hifiglot/src/neural_formant_synthesis/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/hifiglot/src/neural_formant_synthesis/dataset.py": "2006f4a492e3af8750b4d2540c3446521c76240e2d8904b79f7c2ed4b896cea1",
          "src/hifiglot/src/neural_formant_synthesis/feature_extraction.py": "1c4a4870802bcd37a8ea0b5d5b58aee8aa46cf3a9e261c0ed954a6acfc603973",
          "src/hifiglot/src/neural_formant_synthesis/functions.py": "2bd16d715b16e606c1d2f682611eba332d485344f4de603a499cf0ebc1bf292e",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/model/feedforward/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/model/feedforward/activations.py": "bb5a2695d386174f1d97f42dd924f4f034b22ecb915267a24f2f5b26a94524e9",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/model/feedforward/convolution.py": "64933a35dbc1ba6abefc2ec55b62615efe703d4df7ecc4ab7556e8bab292772b",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/model/feedforward/convolution_layer.py": "b0e1d8f20284dcb7a1441725443fff3be7deaee897ee81d79e04ac5311f1443d",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/model/feedforward/convolution_stack.py": "01a717165dc3681ac95506a73924c353e1a5d89bb11da286426c5ec5756089af",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/model/feedforward/wavenet.py": "dd094bc413310531fd1f39c36a3f0f363996cbda540f70492a30f3886224aefd",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/allpole.py": "a5ccef9b233b3b00e6eb4ebfbb6056321aab2c5eef20454e9d36e49ee8b5080e",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/biquad.py": "92cc1be986c988b33bff148f87374efe3e251db7c1f05477a7da3c45d430a7f0",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/emphasis.py": "07d217c53c8df9d2c60ccca688265849975c39bd316af8e218903b985e519f62",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/levinson.py": "c9d6955346e72f6f31fd86432a14aa6d4bbc9c52ee21e3df7d35a9fabc8821be",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/lfilter.py": "7b5c2a8f7adbb9fbf29b52ff726c7f12523af0ec3ece332529a2f60bd58b994b",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/lpc.py": "35dc64c1ec4ff7dd5f475318f0573b4084ea4d99e868363634c374f34e427b05",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/lsf.py": "b8f363001862c59c43726e82c4963adc4a9e6023a9cf3214f87b1d6c43cde923",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/melspec.py": "4d4308c727e78403f162a3c4841ca3c65a3399259a1b302375d81add16206155",
          "src/hifiglot/src/neural_formant_synthesis/glotnet/sigproc/oscillator.py": "98d7908bf1481f5d703629087c706478d9e71bc5beec47cd4adf4aea8ca516be",
          "src/hifiglot/src/neural_formant_synthesis/models.py": "5413d55019c4936e19d71b409cc87dd9a5daf9c563b3903dad6893861681f8c1",
          "src/hifiglot/src/neural_formant_synthesis/sigproc/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/hifiglot/src/neural_formant_synthesis/sigproc/levinson.py": "310e4e5b2059affc3e3d10fa71f02e7a3e7796565260e13d4035dc8ec041a349",
          "src/hifiglot/src/neural_formant_synthesis/sigproc/lpc.py": "b3f90c50610ac5ad2ebc0de26c747954776be53002bf177498f92fdf03956e51",
          "src/hifiglot/src/neural_formant_synthesis/third_party/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/hifiglot/src/neural_formant_synthesis/third_party/hifi_gan/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/hifiglot/src/neural_formant_synthesis/third_party/hifi_gan/env.py": "f5212e8402468ba2b68f79fac906ae48b0762625db368232bbff3bc31fa2b42b",
          "src/hifiglot/src/neural_formant_synthesis/third_party/hifi_gan/models.py": "9d7c786bd53e691c1d3551250108ef7ed5a9a87fb7da4c0854a259d0bdcf1a7f",
          "src/hifiglot/src/neural_formant_synthesis/third_party/hifi_gan/utils.py": "b496698a1c6d9e0fdf7f84c4231746f491ef477631089503da195c3477d7a0a0",
          "src/hifiglot/train_e2e_baseline.py": "98ec299765142efe6f1ce80d62b2ec754acda656a2ef6a83b4d8960f854ae87d",
          "src/hifiglot/train_e2e_hifiglot.py": "af5e4096acf0880028d5bec8bb2df9b37b2eedc13d26ea4ca925810405c81a79"
        },
        "sources": {
          "hifiglot": {
            "repository": "https://github.com/PupuAI/HiFi-Glot.git",
            "revision": "78951625e8f0a113f401fef7032e98d58cfa7dac"
          }
        },
        "note": "上流コードと重みをキャッシュへ取得。第三者コードを本リポジトリに複製しない。"
      },
      "golf": {
        "schema_version": 1,
        "method": "golf",
        "environment": {
          "python": "3.10.22 (main, Oct  3 2026, 01:32:06) [Clang 22.1.3 ]",
          "platform": "Linux-6.17.0-1022-azure-x86_64-with-glibc2.39",
          "dependencies": "absl-py==2.5.1\naiohappyeyeballs==2.7.1\naiohttp==3.14.5\naiosignal==1.4.0\nantlr4-python3-runtime==4.9.3\napprise==2.0.1\nargbind==0.3.9\nasttokens==3.0.2\nasync-timeout==5.0.1\nattrs==26.1.0\naudioread==3.1.0\nbeautifulsoup4==4.15.0\ncertifi==2026.7.22\ncffi==2.1.1\ncharset-normalizer==3.5.2\nclick==8.5.0\ncloudpickle==3.1.2\ncontourpy==1.3.2\ncycler==0.12.1\ncython==3.3.0\ndecorator==5.3.1\ndescript-audiotools==0.7.2\ndiffsptk==4.0.1\ndocstring-parser==0.18.0\neinops==0.8.2\neinx==0.4.3\nexceptiongroup==1.3.1\nexecuting==2.3.0\nffmpy==1.0.0\nfilelock==3.32.3\nfire==0.7.1\nflatten-dict==0.5.0\nfonttools==4.65.0\nfrozendict==2.4.7\nfrozenlist==1.8.0\nfsspec==2024.12.0\nfuture==1.0.0\ngdown==6.4.2\ngrpcio==1.84.0\nh5py==3.16.0\nhf-xet==1.7.0\nhuggingface-hub==0.36.2\nhydra-core==1.3.7\nidna==3.20\nimportlib-resources==7.1.0\nipython==8.39.0\njedi==0.20.0\njinja2==3.1.6\njoblib==1.6.0\njsonargparse==4.52.0\njulius==0.2.8\nkazane==1.0.4\nkiwisolver==1.5.1\nlazy-loader==0.6\nlibrosa==0.10.2.post1\nlightning==2.1.3\nlightning-utilities==0.15.3\nllvmlite==0.50.0\nloguru==0.7.3\nmarkdown==3.10.3\nmarkdown-it-py==4.2.0\nmarkdown2==2.5.5\nmarkupsafe==3.0.3\nmatplotlib==3.10.9\nmatplotlib-inline==0.2.2\nmdurl==0.1.2\nmpmath==1.3.0\nmsgpack==1.2.3\nmultidict==7.1.0\nnetworkx==3.4.2\nnnaudio==0.3.4\nnumba==0.68.0\nnumpy==1.26.4\noauthlib==4.0.0\nomegaconf==2.3.1\npackaging==24.2\nparso==0.8.7\npenn==0.0.14\npexpect==4.9.0\npillow==12.3.0\nplatformdirs==4.12.4\npooch==1.9.0\npraat-parselmouth==0.4.7\nprompt-toolkit==3.0.53\npropcache==0.5.4\nprotobuf==3.19.6\npsutil==7.2.2\nptyprocess==0.7.0\npure-eval==0.2.4\npycparser==3.11\npygments==2.21.0\npyloudnorm==0.2.0\npylstraight==0.1.3\npyparsing==3.3.3\npysocks==1.7.1\npysptk==1.0.1\npystoi==0.4.1\npython-dateutil==2.9.0.post0\npytorch-lightning==2.6.6\npyworld==0.3.5\npyyaml==6.0.3\nrandomname==0.2.1\nrequests==2.34.2\nrequests-oauthlib==2.0.0\nrich==15.0.0\nsafetensors==0.8.0\nscikit-learn==1.7.2\nscipy==1.13.1\nsetuptools==80.10.2\nsix==1.17.0\nsoundfile==0.14.0\nsoupsieve==2.10\nsoxr==1.1.0\nstack-data==0.6.3\nsympy==1.13.1\ntensorboard==2.20.0\ntensorboard-data-server==0.7.2\ntensorboardx==2.6\ntermcolor==3.3.0\nthreadpoolctl==3.7.0\ntorch==2.5.1+cpu\ntorch-einops-utils==0.1.36\ntorch-fftconv==1.0.0\ntorch-stoi==0.2.3\ntorchaudio==2.5.1+cpu\ntorchcomp==0.2.1\ntorchlpc==0.6\ntorchmetrics==1.9.0\ntorchutil==0.1.2\ntqdm==4.70.1\ntraitlets==5.16.1\ntyping-extensions==4.16.0\nurllib3==2.8.0\nvector-quantize-pytorch==1.31.6\nwcwidth==0.9.2\nwerkzeug==3.1.9\nyapecs==0.5.0\nyarl==1.25.1\n"
        },
        "lock_sha256": "a9545775f1381f6d56e390286384993ab8f1b6c3b3010058d5fc54bccef494c1",
        "requirements_sha256": "52fe25d537c36ec24ca7ba6a00e6856cfb540c3d776d0af928ac03ecd06042d0",
        "files": {
          "src/golf/ckpts/interspeech24/golf-ff/checkpoints/epoch=539-step=996840-val_loss=3.073.ckpt": "1eba39620c64d1c59b0a6acb4149200c4791f8e22ec0435bf13bfe5fd86468ab",
          "src/golf/autoencode.py": "2a2f12303a23d09fd44d94bd45054709439c4d109d7f5b28c57d1957b9a06025",
          "src/golf/biquads.py": "adec860a081743399fcba8852d6f7292c26d6716ba6990d047bbe829c8854da0",
          "src/golf/cfg/ae/decoder/ddsp.yaml": "9e8eb0a9b276db7bb8fc72fba4abb34758c708c76e2632eeb5b93bed7fc26961",
          "src/golf/cfg/ae/decoder/golf-precise.yaml": "327e2a1c08e689185e58f6c138697470a4b52fb1254a0bc3d8568dfdb4e2c449",
          "src/golf/cfg/ae/decoder/golf-v1.yaml": "e1222ec194a949fc6658de1eff9c2a972777f97954703368d0cd768bef64a60c",
          "src/golf/cfg/ae/decoder/golf.yaml": "a0f4938cafabf211af968a3a39ad2f0588d6831d619924c10893a39fe55e3d60",
          "src/golf/cfg/ae/decoder/mlsa-taylor.yaml": "7d965b156e559c9e59f3bdf1ee0bda68ca30c7aa29d1a04a162cf2bb845e17da",
          "src/golf/cfg/ae/decoder/mlsa.yaml": "11b3aeaf4ad4e5b0763f5447bda495c60da606c1cba8cc6724da8e0157d9a98a",
          "src/golf/cfg/ae/decoder/nhv.yaml": "e0c9b88eb3f3c1847e79763a91d35d0de336160874c922383f7b9b9c2e5d36bf",
          "src/golf/cfg/ae/decoder/world.yaml": "96354d851ba54bca2da610bf070be6bad34957ab287a42e2e353ea00a777118f",
          "src/golf/cfg/ae/pyworld.yaml": "3030b4fab5766c9cea1678eb61ecfeaac94f6ceeeb999c80d0d384221eb00103",
          "src/golf/cfg/ae/vctk.yaml": "d60df8ffc9b089985f900e650e47c1647d7565c5f1b10fcd6b677012dd5d43c2",
          "src/golf/ckpts/interspeech24/ddsp/config.yaml": "0191313884d522ba81fec1f5283c05785995beeae7d3ddc0aa1e15ebfcbec992",
          "src/golf/ckpts/interspeech24/golf-ff/config.yaml": "a80c16801d77ac2b7541621fe0b68839fb0894988c3725d4b5f6fbb5f57c566d",
          "src/golf/ckpts/interspeech24/golf-ss/config.yaml": "a73e5acc91035b294703cbbeaf03c458a6a31533841732c1853ee25592cef498",
          "src/golf/ckpts/interspeech24/golf-v1/config.yaml": "b115807cb4641e5c189d345f2fff6923ca354de44a0e0b52656dd0176e2d964a",
          "src/golf/ckpts/interspeech24/mlsa/config.yaml": "09894c5ff25733047ee6f2e9d3adc73e5dd9c7cb26d7e6bc05f980e9ec871bc8",
          "src/golf/ckpts/interspeech24/nhv/config.yaml": "a892ce1923fce68ab98d0f3e4875bbeff00f8e6be2d30f223d29baa1a8963930",
          "src/golf/ckpts/interspeech24/world/config.yaml": "5254514fe16427e33d722a30a8280858f3bcfd3c1fa822b4b49dd1968805127c",
          "src/golf/ckpts/ismir23/ddsp_f1/config.yaml": "f6ea30c95a2648e0c699d9e0165fac116bd46fd6c73a419f2f347224321aee22",
          "src/golf/ckpts/ismir23/ddsp_m1/config.yaml": "424d3a7608b88d32b2276de288adf2775c32c8ffacfac9b3babfdea82be7423f",
          "src/golf/ckpts/ismir23/glottal_d_f1/config.yaml": "2b085ec1ad9a544c582267d2621c2ecd908969d0557805f81b8a37c817141824",
          "src/golf/ckpts/ismir23/glottal_d_m1/config.yaml": "496ecf006e2ac4601f1d7b4db539c28583f186a014b4d44cefb0dcf7a560a3bb",
          "src/golf/ckpts/ismir23/pulse_f1/config.yaml": "9206730b004983bef907598f4f1c465fda3f72337df2ba4d1b50ca50732ef7a8",
          "src/golf/ckpts/ismir23/pulse_m1/config.yaml": "62da88b5d6fdfa25e330608dbdd34f77b78a923a4e224c711c146aa400acf20a",
          "src/golf/ckpts/ismir23/sawsing_f1/config.yaml": "adc7d851f4d5dc4f3aa66e386484a4490770108801962cea7e8409447192f223",
          "src/golf/ckpts/ismir23/sawsing_m1/config.yaml": "e70a279a8fa5c3b674773ea87658c9d165ef1d0c8fc5521b55367089e6014ddf",
          "src/golf/convert2v2.py": "7725e1d2751f49f9dc48b425c2016d5cd35e63c93022f149e05d490b3a2c30b7",
          "src/golf/datasets/__init__.py": "32517fc25fc15e7bbb19506c56da5587e5a94035749dd86471a68d41ebec10a2",
          "src/golf/datasets/autoregressive_wav.py": "9ee9a500c467d5dc8a8f1b86868e15552128a7ef3f36edc3e258e4ecc83ab63b",
          "src/golf/datasets/fixed_sr_random_wav.py": "9b27a411235103c1c6301c3088c426e8fc89a3a980c09ae23613896f91f4af08",
          "src/golf/datasets/mir1k.py": "5658b61b63a2df7b1b13a14a073332dc68f3ec5909b602a1b110dd3fb4708d3c",
          "src/golf/datasets/mpop600.py": "da9bd7ecab718b33b65f6c103d55022dd42156e4d20723c1cdde09c945da008f",
          "src/golf/datasets/random_wav.py": "b7818dec321c9bcd6f35fd06327a3fc7ee90745dda4cb447eb0c6fc2e548a8e2",
          "src/golf/datasets/speaker.py": "77ff1ea3468d1778657f5e194d08a7503e121bb802b6d5d6a9de154b6eb25dc6",
          "src/golf/datasets/wav.py": "81187f5e3a20fd00fded9d98f408ecbd4adea14beea01418d4ca039f0c073629",
          "src/golf/eval_pesq.py": "a45fdf7e30ce8693c32871a9c77dc74a97abe59583a2b52bc793886ed074c3e9",
          "src/golf/fad.py": "b94b4e9491e710e23f853edff1c18c372303daed65c37aca0fb6c3ee8a27dec4",
          "src/golf/harm_and_noise.py": "04d4a00566e936f7fcd19f24a9cfccc07990531b71fbd630833b90504efdaead",
          "src/golf/loss/spec.py": "5c1273c069342c4883e299fd3641f0178b327e90ec6d80d390d0ad4accdda722",
          "src/golf/ltng/ae.py": "ee35f49f9d13d6772fd74cf4b150dad2ef10c0b7cd8237551e87a1d675195fb5",
          "src/golf/ltng/cli.py": "6c42cfc3f43c54607572ede66cfa6bb2229cc125c81910bb405cd95cf9c853cb",
          "src/golf/ltng/data.py": "16e85b554969b06d2d7b7549f81b80abe09e5574fdf1859cb1dc4ae6f610aadc",
          "src/golf/ltng/lpcnet.py": "0a17284f0426cc65e80e0670f35ed22cf2ecb5423a8d5759ee93f08c6106f606",
          "src/golf/ltng/vocoder.py": "ed7e3a3074803d89588a25b47b3799084940471839c3446389b7de2f427a100f",
          "src/golf/ltng/world_ae.py": "70e2849a6fa26d4daf44e4958125572bf3aac6e4b5486ab8e8899919b61e12a7",
          "src/golf/main.py": "500b8944c1331a49ef4ce72e103fa8d1e9b965d93df9147b35432cd7724757dd",
          "src/golf/models/audiotensor/__init__.py": "fab3af586887478666f1e1454b8d0a46cfe805dfada05641e0e636b3540401d4",
          "src/golf/models/audiotensor/audiotensor.py": "1ac461dd57554457f325939814ec1f6120d9b84cbd1e576ae450a11f5a7527c2",
          "src/golf/models/crepe.py": "6118fb1b3f43cbadffb6db0502d103f9ad714ec9e0654780dc76a988c2fb34d6",
          "src/golf/models/ctrl.py": "35ca67a52299b1452c741ee197ac3449d2363d02f2ee5aad54da3c8906c6a11c",
          "src/golf/models/enc.py": "288c3729e88f64a16b38018bf19ba5ab9906edbc58e5b43909b68a7064fc2d47",
          "src/golf/models/filters.py": "6414a4ef443cf147d4d30fd3a5eead58bef1d9c399fbed512e31eec9b4701bb7",
          "src/golf/models/hpn.py": "62aca27a29790f6f158bbf66f8bb319ba80424edc0b669469c1a7734346885d5",
          "src/golf/models/lpc.py": "6c72b47e460ca610ad8d2ff2fbf62f9aecef95ff54325b75bdf181086cdaf839",
          "src/golf/models/lpcnet.py": "f992f99d5fe6bb000f08cc391457a8899b7d9fabec28f2935585d2caa1dc5dd4",
          "src/golf/models/lru/__init__.py": "900fbf7b8089409a4358dbf34522d82f7fa4e38d06c17738269d8f09a5d3ced3",
          "src/golf/models/lru/lru.py": "295fadd12f67eb8e900952a217ebb8f9fe6d86e17c2cf37a5943a66d0bb90e07",
          "src/golf/models/lru/parallel_scan.py": "259ebf28f6496ee5fa22174a39b3e3e16bd35ec4bae53ba2fa3f0b28a503d843",
          "src/golf/models/lru/recurrence.py": "5350766b079db0fb8deb3d117ca31777f109930ca9b4a464572ebc4d10c9707e",
          "src/golf/models/lru/test.py": "eb03bde88a02f3b0291a548fcc1d6590af7522c273724e907ec97f5f2337bb6a",
          "src/golf/models/mel.py": "0ea64cf409a11a5490a73e8691e351b2a1d9aec8649ccb418f64f1db0523871b",
          "src/golf/models/noise.py": "1449921f428a194eb6efbce2495875532c388b1b6db7cfe220230b7ea3475a67",
          "src/golf/models/sf.py": "a72c4398df77042f0e1061e4014fa1460df3ee608852491e74d8c9658277f17d",
          "src/golf/models/synth.py": "25dfc2ee890e477be13af55419e0671ead5a7b199b2bfb7b3b074d294e926c10",
          "src/golf/models/tspn.py": "ab26963db50f47f05f9aef55eae8ff98b83511764b12bc4b6598afcab929deb5",
          "src/golf/models/unet.py": "c645b21f802328f366a4d0bb0f3b8f7e854294158cd8a47e2473605124c50c2d",
          "src/golf/models/utils.py": "e11d97b27d42504d31974a1404fc6bde58a71af89b5024e7c91f10f5fd9ea1e6",
          "src/golf/scripts/resample_dir.py": "a7154048c5ea19c0138dfc8077a5fc45f2f009b11061e049ca0fc7a727350974",
          "src/golf/scripts/wav2f0.py": "f86aedc473ba6ceecf856798b3163c9f8cf773f727214ac71abe8328e4538467",
          "src/golf/test_rtf.py": "a7bef08d57a5942b4fc67cabf21ce46b2b749ec53983906121ebd93e15c32ea6",
          "src/golf/tests/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/golf/tests/test_glottal.py": "3ac3a12e0d875d748ff6207b5d88b9b65b5abad8bdfa029d57abef20d71f3cb3",
          "src/golf/tests/test_time_tensor.py": "253bf48388fd00fa0f97d3cf3d869263b99adbf344e33d5f960601a1e56e5449"
        },
        "sources": {
          "golf": {
            "repository": "https://github.com/iamycy/golf.git",
            "revision": "dfe4f4628b8d59d05c9988ae9341a544b1ff30c4"
          }
        },
        "note": "上流コードと重みをキャッシュへ取得。第三者コードを本リポジトリに複製しない。"
      },
      "sifigan": {
        "schema_version": 1,
        "method": "sifigan",
        "environment": {
          "python": "3.10.22 (main, Oct  3 2026, 01:32:06) [Clang 22.1.3 ]",
          "platform": "Linux-6.17.0-1022-azure-x86_64-with-glibc2.39",
          "dependencies": "absl-py==2.5.1\naiohappyeyeballs==2.7.1\naiohttp==3.14.5\naiosignal==1.4.0\nantlr4-python3-runtime==4.9.3\napprise==2.0.1\nargbind==0.3.9\nasttokens==3.0.2\nasync-timeout==5.0.1\nattrs==26.1.0\naudioread==3.1.0\nbeautifulsoup4==4.15.0\ncertifi==2026.7.22\ncffi==2.1.1\ncharset-normalizer==3.5.2\nclick==8.5.0\ncloudpickle==3.1.2\ncontourpy==1.3.2\ncycler==0.12.1\ncython==3.3.0\ndecorator==5.3.1\ndescript-audiotools==0.7.2\ndiffsptk==4.0.1\ndocstring-parser==0.18.0\neinops==0.8.2\neinx==0.4.3\nexceptiongroup==1.3.1\nexecuting==2.3.0\nffmpy==1.0.0\nfilelock==3.32.3\nfire==0.7.1\nflatten-dict==0.5.0\nfonttools==4.65.0\nfrozendict==2.4.7\nfrozenlist==1.8.0\nfsspec==2024.12.0\nfuture==1.0.0\ngdown==6.4.2\ngrpcio==1.84.0\nh5py==3.16.0\nhf-xet==1.7.0\nhuggingface-hub==0.36.2\nhydra-core==1.3.7\nidna==3.20\nimportlib-resources==7.1.0\nipython==8.39.0\njedi==0.20.0\njinja2==3.1.6\njoblib==1.6.0\njsonargparse==4.52.0\njulius==0.2.8\nkazane==1.0.4\nkiwisolver==1.5.1\nlazy-loader==0.6\nlibrosa==0.10.2.post1\nlightning==2.1.3\nlightning-utilities==0.15.3\nllvmlite==0.50.0\nloguru==0.7.3\nmarkdown==3.10.3\nmarkdown-it-py==4.2.0\nmarkdown2==2.5.5\nmarkupsafe==3.0.3\nmatplotlib==3.10.9\nmatplotlib-inline==0.2.2\nmdurl==0.1.2\nmpmath==1.3.0\nmsgpack==1.2.3\nmultidict==7.1.0\nnetworkx==3.4.2\nnnaudio==0.3.4\nnumba==0.68.0\nnumpy==1.26.4\noauthlib==4.0.0\nomegaconf==2.3.1\npackaging==24.2\nparso==0.8.7\npenn==0.0.14\npexpect==4.9.0\npillow==12.3.0\nplatformdirs==4.12.4\npooch==1.9.0\npraat-parselmouth==0.4.7\nprompt-toolkit==3.0.53\npropcache==0.5.4\nprotobuf==3.19.6\npsutil==7.2.2\nptyprocess==0.7.0\npure-eval==0.2.4\npycparser==3.11\npygments==2.21.0\npyloudnorm==0.2.0\npylstraight==0.1.3\npyparsing==3.3.3\npysocks==1.7.1\npysptk==1.0.1\npystoi==0.4.1\npython-dateutil==2.9.0.post0\npytorch-lightning==2.6.6\npyworld==0.3.5\npyyaml==6.0.3\nrandomname==0.2.1\nrequests==2.34.2\nrequests-oauthlib==2.0.0\nrich==15.0.0\nsafetensors==0.8.0\nscikit-learn==1.7.2\nscipy==1.13.1\nsetuptools==80.10.2\nsix==1.17.0\nsoundfile==0.14.0\nsoupsieve==2.10\nsoxr==1.1.0\nstack-data==0.6.3\nsympy==1.13.1\ntensorboard==2.20.0\ntensorboard-data-server==0.7.2\ntensorboardx==2.6\ntermcolor==3.3.0\nthreadpoolctl==3.7.0\ntorch==2.5.1+cpu\ntorch-einops-utils==0.1.36\ntorch-fftconv==1.0.0\ntorch-stoi==0.2.3\ntorchaudio==2.5.1+cpu\ntorchcomp==0.2.1\ntorchlpc==0.6\ntorchmetrics==1.9.0\ntorchutil==0.1.2\ntqdm==4.70.1\ntraitlets==5.16.1\ntyping-extensions==4.16.0\nurllib3==2.8.0\nvector-quantize-pytorch==1.31.6\nwcwidth==0.9.2\nwerkzeug==3.1.9\nyapecs==0.5.0\nyarl==1.25.1\n"
        },
        "lock_sha256": "a9545775f1381f6d56e390286384993ab8f1b6c3b3010058d5fc54bccef494c1",
        "requirements_sha256": "52fe25d537c36ec24ca7ba6a00e6856cfb540c3d776d0af928ac03ecd06042d0",
        "files": {
          "models/sifigan.tar.gz": "5f9ceaafd8d2fd14a683b6cf2e208e8da149fc59d17e34d4027ba02573eeedfe",
          "models/model+stats/sifigan_libritts-r-clean+nus-48e_checkpoint-1000000steps.pkl": "2d673e3cee10b4e88dd69b344443b38e0ef0d48f15f1ee5b417c66c0544e5df6",
          "src/sifigan/setup.py": "955660e15fab96d91c0c195ecec4d7da25790dd3ff9e42fe9480935db9d0a1f6",
          "src/sifigan/sifigan/__init__.py": "f32c41fc1efa7937f87ee36aeedcec4fbb8aeb4b0f1dca5e33d1321a396a6cc5",
          "src/sifigan/sifigan/bin/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/sifigan/sifigan/bin/anasyn.py": "3d12a912478076e73bd558ec4b28da4cdd7e27be4cd514b511d67ac28b88d490",
          "src/sifigan/sifigan/bin/compute_statistics.py": "ffb76f39a1930dd3222d9ef6758f36ee914fafed0b8d2a929f1204e9ab9ace39",
          "src/sifigan/sifigan/bin/config/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/sifigan/sifigan/bin/config/anasyn.yaml": "0cae3f3f1258392019b08c0fcad782eaad45ed4c1d2f4b3689906b7f3bc655eb",
          "src/sifigan/sifigan/bin/config/compute_statistics.yaml": "c6d2c147339f08984c226d05760748d08054e45194056bc08799b2c08027fe2b",
          "src/sifigan/sifigan/bin/config/data/libritts-r-clean+nus-48e.yaml": "2a21098aabf5ec7c217d1aede1dbd594095babc6c1bb0684828b352406cebd3e",
          "src/sifigan/sifigan/bin/config/data/namine_ritsu.yaml": "333443ecbea006f6687123474ea5107264b2ac9815528b12747fab67835ba2db",
          "src/sifigan/sifigan/bin/config/decode.yaml": "904304f93df7e35de6c1bcd73c0aa088610c9ee14b4c875ed5864e4a6aa90220",
          "src/sifigan/sifigan/bin/config/discriminator/hifigan.yaml": "7e2b0892c5329ad2f578d0a2286099176003cd9fe05d5de9a09cc1449b6539db",
          "src/sifigan/sifigan/bin/config/discriminator/univnet.yaml": "aee026484dc245ffa67ecdac9516125549bebb4c5784db284d5a257788760b9e",
          "src/sifigan/sifigan/bin/config/extract_features.yaml": "fefe173d53b0740b0f9f2925adf9b7b17ced1d11dce72451e0eb888ce1670056",
          "src/sifigan/sifigan/bin/config/generator/hifigan.yaml": "18956fc64c7d9691439038708bdd1879a50e2c15d9a535e22eff893f7482e576",
          "src/sifigan/sifigan/bin/config/generator/sifigan.direct.yaml": "a85a2b28cc899370f89185a9833f073081a9aa61a6edb8e27fcb8eaabbbeb803",
          "src/sifigan/sifigan/bin/config/generator/sifigan.yaml": "16219976449841293da7fa90f2623a47b3d1de8aae0487b15fc2d500e21891b9",
          "src/sifigan/sifigan/bin/config/param_count.yaml": "65c71910367e44f436adc4282f25b453f090e6fe5bce236187b4e1e510a30e16",
          "src/sifigan/sifigan/bin/config/train/hifigan.yaml": "05aeb83f4595c1acedd6d42c73d349807808bcc80641f270274f8122470e7356",
          "src/sifigan/sifigan/bin/config/train/sifigan.yaml": "357f35d8131b9260f328f8670a5bf0ab71d7ac2bb9499067b836f34cbffd7253",
          "src/sifigan/sifigan/bin/config/train/sifigan_1000k.yaml": "a7e0aab88afdc9aed56da82cbc5c2446988326291e39b24a246d32b0a02b3e0f",
          "src/sifigan/sifigan/bin/config/train.yaml": "0b1fd45eb554954c42b46189cf9cee8ab120d717575f2baa10ac418a331f9aa6",
          "src/sifigan/sifigan/bin/decode.py": "75b3f5bdf31d3d294d460cacf8cfa289e1638c7a8c84b23cbedba18203ab53ab",
          "src/sifigan/sifigan/bin/extract_features.py": "d93d46084f5ccdeeb7154b865e763440ec4683a6d88fab6362f4ead3705a553a",
          "src/sifigan/sifigan/bin/param_count.py": "40c28ae974a9359f244c638cf50b7d063eb7bc579fb64abed1c6c3107dc20274",
          "src/sifigan/sifigan/bin/train.py": "e3340565559c6e00343a4cfb5f966435c3b4a86bf1ebb7fc3b800c8068445e7a",
          "src/sifigan/sifigan/datasets/__init__.py": "f3ecd1bd5729339ee21e2058351f71fd6ab2963bf0d22122381b5a5b03dcd4be",
          "src/sifigan/sifigan/datasets/audio_feat_dataset.py": "efe5a8ce9f41ea4a7b2320cf77e41fc5262f658cda866aed5fde2a1d268762b0",
          "src/sifigan/sifigan/layers/__init__.py": "5ce2179c6c21bff9c91f22e5d4d1a0d625b7112208b08fc29fce810805dc0959",
          "src/sifigan/sifigan/layers/cheaptrick.py": "c0e6655c0530146147eb516267d7e163212c9dc2841fcca19e5222c875552d86",
          "src/sifigan/sifigan/layers/residual_block.py": "d52cc8d6a434e49e0306e7da578de9e4d1dd390962250665a100337dfa27ded9",
          "src/sifigan/sifigan/layers/snake.py": "b3ef5f036e96d50813fe72d459d829516dd394d8ad1fa01562ac8d0d6f7eb563",
          "src/sifigan/sifigan/losses/__init__.py": "54ccd9347f5f1ab1c67b3cf44b2ee6087883d7acc86af919efedf83bd16e3cd8",
          "src/sifigan/sifigan/losses/adv.py": "79269d3b8c438ea8ff29dcfe5ee1877dcbd68f67de59d107a1e07b26b7553488",
          "src/sifigan/sifigan/losses/mel.py": "20a73ebde9716386e61ae5ca9e6552b70985113129332475e3092ee9b8cff6ac",
          "src/sifigan/sifigan/losses/reg.py": "6c14086c8dd74a03f12f0452d2eac0a62557c38e0fa9efb763afe8ad6a2fc2ad",
          "src/sifigan/sifigan/models/__init__.py": "e495ee47a097d992bd332b85d70f84f6adfc96d402e278c158365567798ea49a",
          "src/sifigan/sifigan/models/discriminator.py": "7e8caa445914a8ba91d7122eee8d8d91d0d6cec041004af01e1f7345ad0afb1d",
          "src/sifigan/sifigan/models/generator.py": "9d7b361049587d26c21f1101278892536efe7b36503ba31e7705490a507070ba",
          "src/sifigan/sifigan/utils/__init__.py": "09b9e8bc81ab3565c80b755892e300b786e9a14712159378a809126eaba4bc81",
          "src/sifigan/sifigan/utils/features.py": "cf8a7d0a56e54fc073fba6698c11b6ee2e35114797c2a99fca0b149c0199a609",
          "src/sifigan/sifigan/utils/index.py": "be3134981743d4738b4ffc42956df0040d5dbe5d66f5ce323ff442a5119ecf37",
          "src/sifigan/sifigan/utils/utils.py": "eef463c90a5b3a45ac8e1199abe2295339dbf991ef9d67ee3a48c53f6e47125f",
          "models/model+stats/libritts_r_clean+nus-48e_train_no_dev.joblib": "c2ae18f7be617c3d11fb1f5ceff6f1fdb57dc35bc9737bee51416c0886dc4708"
        },
        "sources": {
          "sifigan": {
            "repository": "https://github.com/chomeyama/SiFiGAN.git",
            "revision": "cdb20daeb4da4a126077819091706ee90522c7a8"
          }
        },
        "note": "上流コードと重みをキャッシュへ取得。第三者コードを本リポジトリに複製しない。"
      },
      "llsm2": {
        "schema_version": 1,
        "method": "llsm2",
        "environment": {
          "python": "3.10.22 (main, Oct  3 2026, 01:32:06) [Clang 22.1.3 ]",
          "platform": "Linux-6.17.0-1022-azure-x86_64-with-glibc2.39",
          "dependencies": "absl-py==2.5.1\naiohappyeyeballs==2.7.1\naiohttp==3.14.5\naiosignal==1.4.0\nantlr4-python3-runtime==4.9.3\napprise==2.0.1\nargbind==0.3.9\nasttokens==3.0.2\nasync-timeout==5.0.1\nattrs==26.1.0\naudioread==3.1.0\nbeautifulsoup4==4.15.0\ncertifi==2026.7.22\ncffi==2.1.1\ncharset-normalizer==3.5.2\nclick==8.5.0\ncloudpickle==3.1.2\ncontourpy==1.3.2\ncycler==0.12.1\ncython==3.3.0\ndecorator==5.3.1\ndescript-audiotools==0.7.2\ndiffsptk==4.0.1\ndocstring-parser==0.18.0\neinops==0.8.2\neinx==0.4.3\nexceptiongroup==1.3.1\nexecuting==2.3.0\nffmpy==1.0.0\nfilelock==3.32.3\nfire==0.7.1\nflatten-dict==0.5.0\nfonttools==4.65.0\nfrozendict==2.4.7\nfrozenlist==1.8.0\nfsspec==2024.12.0\nfuture==1.0.0\ngdown==6.4.2\ngrpcio==1.84.0\nh5py==3.16.0\nhf-xet==1.7.0\nhuggingface-hub==0.36.2\nhydra-core==1.3.7\nidna==3.20\nimportlib-resources==7.1.0\nipython==8.39.0\njedi==0.20.0\njinja2==3.1.6\njoblib==1.6.0\njsonargparse==4.52.0\njulius==0.2.8\nkazane==1.0.4\nkiwisolver==1.5.1\nlazy-loader==0.6\nlibrosa==0.10.2.post1\nlightning==2.1.3\nlightning-utilities==0.15.3\nllvmlite==0.50.0\nloguru==0.7.3\nmarkdown==3.10.3\nmarkdown-it-py==4.2.0\nmarkdown2==2.5.5\nmarkupsafe==3.0.3\nmatplotlib==3.10.9\nmatplotlib-inline==0.2.2\nmdurl==0.1.2\nmpmath==1.3.0\nmsgpack==1.2.3\nmultidict==7.1.0\nnetworkx==3.4.2\nnnaudio==0.3.4\nnumba==0.68.0\nnumpy==1.26.4\noauthlib==4.0.0\nomegaconf==2.3.1\npackaging==24.2\nparso==0.8.7\npenn==0.0.14\npexpect==4.9.0\npillow==12.3.0\nplatformdirs==4.12.4\npooch==1.9.0\npraat-parselmouth==0.4.7\nprompt-toolkit==3.0.53\npropcache==0.5.4\nprotobuf==3.19.6\npsutil==7.2.2\nptyprocess==0.7.0\npure-eval==0.2.4\npycparser==3.11\npygments==2.21.0\npyloudnorm==0.2.0\npylstraight==0.1.3\npyparsing==3.3.3\npysocks==1.7.1\npysptk==1.0.1\npystoi==0.4.1\npython-dateutil==2.9.0.post0\npytorch-lightning==2.6.6\npyworld==0.3.5\npyyaml==6.0.3\nrandomname==0.2.1\nrequests==2.34.2\nrequests-oauthlib==2.0.0\nrich==15.0.0\nsafetensors==0.8.0\nscikit-learn==1.7.2\nscipy==1.13.1\nsetuptools==80.10.2\nsix==1.17.0\nsoundfile==0.14.0\nsoupsieve==2.10\nsoxr==1.1.0\nstack-data==0.6.3\nsympy==1.13.1\ntensorboard==2.20.0\ntensorboard-data-server==0.7.2\ntensorboardx==2.6\ntermcolor==3.3.0\nthreadpoolctl==3.7.0\ntorch==2.5.1+cpu\ntorch-einops-utils==0.1.36\ntorch-fftconv==1.0.0\ntorch-stoi==0.2.3\ntorchaudio==2.5.1+cpu\ntorchcomp==0.2.1\ntorchlpc==0.6\ntorchmetrics==1.9.0\ntorchutil==0.1.2\ntqdm==4.70.1\ntraitlets==5.16.1\ntyping-extensions==4.16.0\nurllib3==2.8.0\nvector-quantize-pytorch==1.31.6\nwcwidth==0.9.2\nwerkzeug==3.1.9\nyapecs==0.5.0\nyarl==1.25.1\n"
        },
        "lock_sha256": "a9545775f1381f6d56e390286384993ab8f1b6c3b3010058d5fc54bccef494c1",
        "requirements_sha256": "52fe25d537c36ec24ca7ba6a00e6856cfb540c3d776d0af928ac03ecd06042d0",
        "files": {
          "src/llsm/buffer.h": "cbbc9b886bd7f20176e6f5798bf357fd18b54f1b9596a96facd07bbd0c9734d7",
          "src/llsm/coder.c": "870cbbff5d99547ad8def1bac8615319b113d84eea78c4db81b151204c887aea",
          "src/llsm/constants.h": "e17e4f921753f9608e79a118e06f5a433ed6579e69344aa852f2b042423c4c8d",
          "src/llsm/container.c": "0064ed944ea38439e9cbad3cb142f61ea3bed09804d819f0b4c9d7116b515ff3",
          "src/llsm/dsputils.c": "00bfef0a8b55a7619ccc88fc3c852da393e908316b3efec3f760465eb7cdb698",
          "src/llsm/dsputils.h": "8994e39a248784cc3e68c79b1ad231a61e020cfd246332f0a8a921753ec88a08",
          "src/llsm/filter-coef.h": "bc7e3c6f2164c8cd4effa881e0fd24deea67d8283efaac6f6f2e0b63b7d60ca0",
          "src/llsm/frame.c": "431139f767172a8f1707ac12ca6e11a6bbf328ed120016d0de18f1df09e3a205",
          "src/llsm/layer0.c": "40e44ae44e7b37a2d66e50c2320794f0fbacf9c4df78034194d9a48c299b2dee",
          "src/llsm/layer1.c": "e8a5c2df95e3c240427f1300886e8b0d129051f7d0e6dc7e2f66651a40bc0d1e",
          "src/llsm/llsm.h": "ebbe38a332b94f4d5dcf9fa0170cfbd76190bb9df049364e95990ea122079bed",
          "src/llsm/llsmrt.c": "53a6db74429167f4ae7af6b34c205737181c971f3d19c643065a01695f742136",
          "src/llsm/llsmrt.h": "82ed9d3dbc970ad766ffc1d5f40801b6f0233add8a096f4b7a4250ee1d932f08",
          "src/llsm/llsmutils.c": "853ef4972b3eb9f01ab8afa919455de652fa89a2b64e3ab3e166dd1e5f3bc3ba",
          "src/llsm/llsmutils.h": "f8b002d50e21bc4d52af2bfffeb724a00c9a6d53f8943268e947e9b095e00820",
          "src/llsm/test/demo-stretch.c": "7cd8ead474c73cbdf9281d9ddbbece49e4385d50b0d8b1d4461e48b50a6562fe",
          "src/llsm/test/test-coder.c": "753a8036ddfec1dbe990e0a28b635dfdabe23bb887bd43c46f6415f2073791ce",
          "src/llsm/test/test-dsputils.c": "d5282947347a2d91f5b686b74d646993fc1c85fa23c105fb50ec006ed3bda655",
          "src/llsm/test/test-harmonic.c": "b0ee2f7e19175da2bdb673cd0c14046e902749792501975fef73a630956ea294",
          "src/llsm/test/test-layer0-anasynth.c": "d55fbeb221460afdd7733804af5afc562f5509215457a0a7f890669c7d6cecb3",
          "src/llsm/test/test-layer0-edgecase.c": "2ea95422798bdd4f4762f821459581ffbd2ee657f306311e35e1243ab4245507",
          "src/llsm/test/test-layer1-anasynth.c": "0c0a86b673d8910edf9258c7889bfa75f9ec4e58947c0cada0efc8df082d85d3",
          "src/llsm/test/test-llsmrt.c": "ef3ef6d92cc49e69f362231b94090af6f613bd724d5547817be41e9bbf9d07bc",
          "src/llsm/test/test-pbpeffects.c": "cf00074e01b9fcd132bc618763baea5fe1ce92c22fcbdea60c90b1536e11e7d6",
          "src/llsm/test/test-structs.c": "ff8bc187136ac6ab1a70523b6c24f45399287241d1a470e236bd9eda7de12c63",
          "src/llsm/test/verify-utils.h": "541e7cbad0c9bdc5f2541014b2cd46d97d0204d6907277676a2d6f9dc06d6625",
          "src/ciglet/ciglet.c": "41d3f5b84dbe32f447675756007cac6ba6fe836cfbf9b6d122b393d47fd77b46",
          "src/ciglet/ciglet.h": "1290f7e14e6d0fc2f1849f31f75ef610ae7dd2abdf1e1f470ac93b58bf56331e",
          "src/ciglet/external/fast_median.c": "08e803cea8cec1e0d8a36a4df4224650cf667260277829c98f2eaa184f89c596",
          "src/ciglet/external/fastapprox-all.h": "568b4b09def0f5539b3487ed01a9f616955fb8f11b4d90c1c7ffc0f34c9fc543",
          "src/ciglet/external/fftsg_h.c": "bfb6457551b5d071e865bff6b7998635040e9fd6fea9d45891553e3ac9013678",
          "src/ciglet/external/wavfile.c": "c791eaa06d3b626224a2c97356f41177fb8637064c4856d912443431afeedaa4",
          "src/ciglet/test/test-fft.c": "36c41142f5e28e896806132c3f29487f335ef59f1048af737ac3268f9f6668a5",
          "src/ciglet/test/test.c": "ce19053bca09affb155bce3b8c38f3f46acfbcaf22922f509a34a85239a5c8fd",
          "native/build.json": "586febf2ca6f8a1ecb61c1defe0a3a878ca7bb1b7613cd2362abdef34874c2a8",
          "native/include/ciglet/ciglet.h": "1290f7e14e6d0fc2f1849f31f75ef610ae7dd2abdf1e1f470ac93b58bf56331e",
          "native/include/ciglet/external/fastapprox-all.h": "568b4b09def0f5539b3487ed01a9f616955fb8f11b4d90c1c7ffc0f34c9fc543",
          "native/libllsm_speed.so": "78a6417acc33fc1ccc7a79aced495a4eac6499f92e77cb77aa4568de368408f1",
          "native/wrapper.c": "6a2abdf3b87f4af1dc402dbaec4340ad9907a1f35f27222641f7f6dda40724b9"
        },
        "sources": {
          "llsm": {
            "repository": "https://github.com/Sleepwalking/libllsm2.git",
            "revision": "d18cdf02b31819670e3bc23ead933ad9467da06d"
          },
          "ciglet": {
            "repository": "https://github.com/Sleepwalking/ciglet.git",
            "revision": "895ba9b1c0eabee83d3544208bbc82420efa3206"
          }
        },
        "note": "上流コードと重みをキャッシュへ取得。第三者コードを本リポジトリに複製しない。"
      },
      "magphase": {
        "schema_version": 1,
        "method": "magphase",
        "environment": {
          "python": "3.10.22 (main, Oct  3 2026, 01:32:06) [Clang 22.1.3 ]",
          "platform": "Linux-6.17.0-1022-azure-x86_64-with-glibc2.39",
          "dependencies": "absl-py==2.5.1\naiohappyeyeballs==2.7.1\naiohttp==3.14.5\naiosignal==1.4.0\nantlr4-python3-runtime==4.9.3\napprise==2.0.1\nargbind==0.3.9\nasttokens==3.0.2\nasync-timeout==5.0.1\nattrs==26.1.0\naudioread==3.1.0\nbeautifulsoup4==4.15.0\ncertifi==2026.7.22\ncffi==2.1.1\ncharset-normalizer==3.5.2\nclick==8.5.0\ncloudpickle==3.1.2\ncontourpy==1.3.2\ncycler==0.12.1\ncython==3.3.0\ndecorator==5.3.1\ndescript-audiotools==0.7.2\ndiffsptk==4.0.1\ndocstring-parser==0.18.0\neinops==0.8.2\neinx==0.4.3\nexceptiongroup==1.3.1\nexecuting==2.3.0\nffmpy==1.0.0\nfilelock==3.32.3\nfire==0.7.1\nflatten-dict==0.5.0\nfonttools==4.65.0\nfrozendict==2.4.7\nfrozenlist==1.8.0\nfsspec==2024.12.0\nfuture==1.0.0\ngdown==6.4.2\ngrpcio==1.84.0\nh5py==3.16.0\nhf-xet==1.7.0\nhuggingface-hub==0.36.2\nhydra-core==1.3.7\nidna==3.20\nimportlib-resources==7.1.0\nipython==8.39.0\njedi==0.20.0\njinja2==3.1.6\njoblib==1.6.0\njsonargparse==4.52.0\njulius==0.2.8\nkazane==1.0.4\nkiwisolver==1.5.1\nlazy-loader==0.6\nlibrosa==0.10.2.post1\nlightning==2.1.3\nlightning-utilities==0.15.3\nllvmlite==0.50.0\nloguru==0.7.3\nmarkdown==3.10.3\nmarkdown-it-py==4.2.0\nmarkdown2==2.5.5\nmarkupsafe==3.0.3\nmatplotlib==3.10.9\nmatplotlib-inline==0.2.2\nmdurl==0.1.2\nmpmath==1.3.0\nmsgpack==1.2.3\nmultidict==7.1.0\nnetworkx==3.4.2\nnnaudio==0.3.4\nnumba==0.68.0\nnumpy==1.26.4\noauthlib==4.0.0\nomegaconf==2.3.1\npackaging==24.2\nparso==0.8.7\npenn==0.0.14\npexpect==4.9.0\npillow==12.3.0\nplatformdirs==4.12.4\npooch==1.9.0\npraat-parselmouth==0.4.7\nprompt-toolkit==3.0.53\npropcache==0.5.4\nprotobuf==3.19.6\npsutil==7.2.2\nptyprocess==0.7.0\npure-eval==0.2.4\npycparser==3.11\npygments==2.21.0\npyloudnorm==0.2.0\npylstraight==0.1.3\npyparsing==3.3.3\npysocks==1.7.1\npysptk==1.0.1\npystoi==0.4.1\npython-dateutil==2.9.0.post0\npytorch-lightning==2.6.6\npyworld==0.3.5\npyyaml==6.0.3\nrandomname==0.2.1\nrequests==2.34.2\nrequests-oauthlib==2.0.0\nrich==15.0.0\nsafetensors==0.8.0\nscikit-learn==1.7.2\nscipy==1.13.1\nsetuptools==80.10.2\nsix==1.17.0\nsoundfile==0.14.0\nsoupsieve==2.10\nsoxr==1.1.0\nstack-data==0.6.3\nsympy==1.13.1\ntensorboard==2.20.0\ntensorboard-data-server==0.7.2\ntensorboardx==2.6\ntermcolor==3.3.0\nthreadpoolctl==3.7.0\ntorch==2.5.1+cpu\ntorch-einops-utils==0.1.36\ntorch-fftconv==1.0.0\ntorch-stoi==0.2.3\ntorchaudio==2.5.1+cpu\ntorchcomp==0.2.1\ntorchlpc==0.6\ntorchmetrics==1.9.0\ntorchutil==0.1.2\ntqdm==4.70.1\ntraitlets==5.16.1\ntyping-extensions==4.16.0\nurllib3==2.8.0\nvector-quantize-pytorch==1.31.6\nwcwidth==0.9.2\nwerkzeug==3.1.9\nyapecs==0.5.0\nyarl==1.25.1\n"
        },
        "lock_sha256": "a9545775f1381f6d56e390286384993ab8f1b6c3b3010058d5fc54bccef494c1",
        "requirements_sha256": "52fe25d537c36ec24ca7ba6a00e6856cfb540c3d776d0af928ac03ecd06042d0",
        "files": {
          "src/magphase/demos/demo_copy_synthesis_lossless.py": "d3aa2abd2f7facff5dc821e5db9851b75216dc507f607bd8a77bf27df736ce68",
          "src/magphase/demos/demo_copy_synthesis_low_dim.py": "9aad8c21dd0cfce4f390008d8da3204df6dec154ad0de849e3243167d8bae25d",
          "src/magphase/development/compare_mags.py": "382f9a6ada1cd00edec29da28acda9de990e364e27af59a9f0afa48b4c97a6cd",
          "src/magphase/development/run_merlin_experiment.py": "870158543e6e6244c363ad1bbeda31184feb46f2ead86d244e205480530c4c9d",
          "src/magphase/scripts/batch_convert_label_state_aligned_to_variable_frame_rate.py": "6705bdb715d6c87caf7da424458c4ba0d726764dd23eab5fc598487dc8bcda4d",
          "src/magphase/scripts/batch_feature_extraction_for_tts.py": "e5e901d8c8807e48d002ecb8751d3beb7d3fbb856db9b223f9ff9b9f66ffff34",
          "src/magphase/scripts/batch_waveform_generation.py": "c0df72751851ab044bb980b006df6f0e5c000bc849d444fc80a275e1e19faaa3",
          "src/magphase/src/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
          "src/magphase/src/libaudio.py": "40fb05431d31fc762129a9b92f70a8f60200e564105b49f4f1e0c2b37fd43690",
          "src/magphase/src/libplot.py": "348db4785d0d6d33565a3c9192ec02cf2a3c028900cd998e1d842c6723da6b11",
          "src/magphase/src/libutils.py": "305810cc60870fa8bb6241f0bf7046477c8015c82d31a23c8e2a33b72ac69ee4",
          "src/magphase/src/magphase.py": "4dd367e608f58fde1b2699dd57f8d33092fb8626ac493697d04561e556dd92bc",
          "src/reaper/build/CMakeFiles/3.31.6/CompilerIdC/CMakeCCompilerId.c": "60d0dc2f177f0317302cef41d454988d9d49759ad3e011365bd6f6c544dfd126",
          "src/reaper/build/CMakeFiles/CMakeConfigureLog.yaml": "6d380c0c06efceaaceef623722693ada8f99ff77b0de7d43c24f2c1afcf8fc4c",
          "src/reaper/core/file_resource.h": "46b11710b474e34b15a5b5abbf648016e49f79d8825501712e2d1b6d06359c78",
          "src/reaper/core/float_matrix-inl.h": "1a2c36ff15eb5bbc20ea0d918855c2b61edeac44d99900885c76967198432117",
          "src/reaper/core/float_matrix.h": "9027d8d94753f74a2edb422acbe73cb881bf30affa8b6f74b612543c140679b0",
          "src/reaper/core/track.h": "072c9c901156fa2d7a21bf688235c53bc25cf6d40c42e9bb81dda47d93cb5f81",
          "src/reaper/epoch_tracker/epoch_tracker.h": "074f3ee3bde6fd346e28b9b4f8dbb5281870d3424d58c99baef73b6e315af80f",
          "src/reaper/epoch_tracker/fd_filter.h": "4b6e6bdb34305e2ba726a9735b2850bdff62de8b223ca13824cc56c55086f167",
          "src/reaper/epoch_tracker/fft.h": "0a6d8a9e3b45f1c9ce3356fff157a5cd543ccd39f0f90099afc19654f754174b",
          "src/reaper/epoch_tracker/lpc_analyzer.h": "a2009fad56bf64dd8cff42c75e0724b9df5c5fdf2ab92486c48b91acb2f5b945",
          "src/reaper/wave/codec_api-inl.h": "4fa4a9c669e3330c1b2cc46d31333c0a0a5e5d9bd7ab752159ad903b84d1e163",
          "src/reaper/wave/codec_api.h": "85d3e07e807e8bf6e47ab13226b4151e670095accd04b4226ddcf8d2c8c32915",
          "src/reaper/wave/codec_riff.h": "e7ea52ab3b1ffabd7b099ea2a18d051a9787b2ca9ba40fc180b44af1bfe45d70",
          "src/reaper/wave/wave.h": "0f96bece1e3313dd0cf55d1933161070a3c2ce778253a2250d175f8fc42e87b1",
          "src/reaper/wave/wave_io-inl.h": "fe5f7375d75f0b0892a367cffe405efb113173a0aa16e9ab3d8f2cd730d2c8a6",
          "src/reaper/wave/wave_io.h": "04245f9d0b38c24c9c5c8e17df758d350e550c31023eef73a93dd77a6db04294",
          "magphase-port/demos/demo_copy_synthesis_lossless.py": "d3aa2abd2f7facff5dc821e5db9851b75216dc507f607bd8a77bf27df736ce68",
          "magphase-port/demos/demo_copy_synthesis_low_dim.py": "9aad8c21dd0cfce4f390008d8da3204df6dec154ad0de849e3243167d8bae25d",
          "magphase-port/development/compare_mags.py": "382f9a6ada1cd00edec29da28acda9de990e364e27af59a9f0afa48b4c97a6cd",
          "magphase-port/development/run_merlin_experiment.py": "870158543e6e6244c363ad1bbeda31184feb46f2ead86d244e205480530c4c9d",
          "magphase-port/scripts/batch_convert_label_state_aligned_to_variable_frame_rate.py": "6705bdb715d6c87caf7da424458c4ba0d726764dd23eab5fc598487dc8bcda4d",
          "magphase-port/scripts/batch_feature_extraction_for_tts.py": "e5e901d8c8807e48d002ecb8751d3beb7d3fbb856db9b223f9ff9b9f66ffff34",
          "magphase-port/scripts/batch_waveform_generation.py": "c0df72751851ab044bb980b006df6f0e5c000bc849d444fc80a275e1e19faaa3",
          "magphase-port/src/__init__.py": "2e262bac3afb815e57a323ef5b5a60df817d61eca72fb3b86620dd1b30b7e899",
          "magphase-port/src/libaudio.py": "f93e14c379b87719a015017a56032a5920a5a4164034e7a3aaa5816dca4ce96a",
          "magphase-port/src/libplot.py": "c4f77b57d88edb05e13107f471aa2269d0cf3f713d3cbaa5593a8c13b3b7a7c5",
          "magphase-port/src/libutils.py": "6d4b040230702f765f6ba8cc71ed386d17ff61b79c6fe5eba0cc81c6722c8139",
          "magphase-port/src/magphase.py": "2ce23c57862dfca7b106570aeecf988e8e60cd7cbbf8ba583fe16591db548ba5",
          "magphase-port/tools/bin/acep": "5c39b46a97b6686ea7a20d65c4f088a88f6f6763b36e8be24b42d9ff7882bc41",
          "magphase-port/tools/bin/acorr": "46716b512c5ba931027885b5ece6cfcf349eda0f16ad6e4d61b1926b725dd4fb",
          "magphase-port/tools/bin/agcep": "c7390a5789f6eeb92097994e07328dd3546d3b9ef241417e60bcc73f44804add",
          "magphase-port/tools/bin/amcep": "8572c9ba36f259c0399de0a83c820b80b0c24ed84404f6a9dcbf9163c0d3a7b0",
          "magphase-port/tools/bin/average": "fe96d954ddd7c0b39373cc9a112ba4f1ae75538fe2bc75297d066b9bdc35abdf",
          "magphase-port/tools/bin/b2mc": "744ded3ef7efb8b598967c5665abadad60d8fa0c5343ca4975ad6f6621567f8a",
          "magphase-port/tools/bin/bcp": "c422138a94001d184184ea85dbe1ef61cca7223432322f2c73d662fb15f63b85",
          "magphase-port/tools/bin/bcut": "37f39bf7d7eb9c6fd596f45815edc8ec26505ec603784027d62aa61fbc6ee05a",
          "magphase-port/tools/bin/bell": "daa10b3db0d317c45ba972cc043f10edc755a512502d533a800a64ed73cfb631",
          "magphase-port/tools/bin/c2acr": "64a1675213a654cd79e6c963b4e178039333c7559a1f4cbf2a211fd6bb45d1e6",
          "magphase-port/tools/bin/c2ir": "690c6e5c51713a4a0ed73371e1619aa4063f8d15313a40b2d8c38d947c69956a",
          "magphase-port/tools/bin/c2ndps": "8fe7b507ea94acd4b828098a445a39650637650abab842b8114206cc7492df54",
          "magphase-port/tools/bin/c2sp": "3e998a84d4c14da2b3153255d14f1206282664cf129f015103ca48a777a68028",
          "magphase-port/tools/bin/cat2": "d313117626b4070af4072c050dae05b23facd6f3f47c6c105481d411d4cdf15d",
          "magphase-port/tools/bin/cdist": "53b595eab3a735d309112062974158b2ee96aab984fba30f2be09d5cbcc9703b",
          "magphase-port/tools/bin/clip": "3be76a715b549d3c0d1fd37b16a0fdecd017d9151a63f850589c44664c126a51",
          "magphase-port/tools/bin/da": "67acd139295ea8ba0a3e680461a089b9e602fb64546c60c7274207c58e6b3f3d",
          "magphase-port/tools/bin/dawrite": "90d651ab82a337287fe7880fcdc92db28c59fdefe2b505be146aa304464024fc",
          "magphase-port/tools/bin/dct": "9411d80bcf436a93a7f93127e20ac4e15810e039bbc2bb4bdcac9246e4057805",
          "magphase-port/tools/bin/decimate": "56180d12440263cab80d46fde0a8a90d197cee2ea74859806a3af1d63ddc41a8",
          "magphase-port/tools/bin/delay": "7908a326d8126881676848ed4689f7c83689e1c822e583dae534901c76f123a5",
          "magphase-port/tools/bin/delta": "c5823e0faca3216746e2e2f918d71ca911dda6cd74176da569d1ff83e23187bb",
          "magphase-port/tools/bin/df2": "c40a799fc544e52adbae018ec9ff1326f848f72a05c90830daf45580afbc16a7",
          "magphase-port/tools/bin/dfs": "8a6be1c077880127bfbbcb37667ec1f37dbb278c60db381cd8ad91c3d8533fa7",
          "magphase-port/tools/bin/dmp": "86dd8ff58e18e3df549d2f2dacc52581580c7cb5960d3aa2b85965d37638b5ec",
          "magphase-port/tools/bin/ds": "6dc22c2c8431435c9cc97dafd64e395b1ba1b9bb8d9f93f0fee324bbdc612406",
          "magphase-port/tools/bin/dtw": "752522451acb7b29783b5530fbffc8ecdbed80886bf49562b271c99f799eee6e",
          "magphase-port/tools/bin/echo2": "0e049e56bff4410c93192836b5628564fa3b3236b2869e69e84cbccc83f20d8a",
          "magphase-port/tools/bin/excite": "b1a0e0583a9da38b6dd48ae83dd271de88ed3125c311df7ddc9ca376e7848a9b",
          "magphase-port/tools/bin/extract": "b0cf062d990ad2744017cc8976e4a62ad6d0a2518ed007f9261b463109c31c22",
          "magphase-port/tools/bin/fd": "0301afe9fc38b14c9bdf1540cf3e7d50ba5436c2fb55857b298cabd5f2d1bee2",
          "magphase-port/tools/bin/fdrw": "a240de9f7e0ae9d88d35ce210157c34937037f85e9087bffabfe652847c72c23",
          "magphase-port/tools/bin/fft": "c233aaf01b2e2edd3e82e42b890d7c096c5c874be2cb278ce2d133309ff6ef51",
          "magphase-port/tools/bin/fft2": "4cea31927bd183699125df5dd631ac519a896f598ff341dbdc72868d58933b50",
          "magphase-port/tools/bin/fftcep": "721f510647cee78c42002d0c76df5c7340af74e58627e4e633fa30207feda722",
          "magphase-port/tools/bin/fftr": "a04e5f4974e552ac4d80b21da9f34a9f892dee19700faac9ac826d886eb2d3c3",
          "magphase-port/tools/bin/fftr2": "d2190225cc4c6814d2938ec090ed41140100952903af709220c5b40128c65d37",
          "magphase-port/tools/bin/fig": "f0c5835c73583b69b6de482bb249d67b922bf48a00d812e3dd605ae01a913fef",
          "magphase-port/tools/bin/frame": "bf021097935a9c6c6a8dee6b3868659c02e7c0e6e71a8ee0813c43742f2a27a2",
          "magphase-port/tools/bin/freqt": "a8bb2eb2a0d223fd4b8616137164f61daed9439c184b702d5da8a4caf58e8520",
          "magphase-port/tools/bin/gc2gc": "21ffa2bd3e8fd73303b9905832a391b21b1f0766512c7d2f602b96564d26f0b5",
          "magphase-port/tools/bin/gcep": "36830f5342bdc118bd49dca50f0691c38f575a7d0d13e669fe779f3330680603",
          "magphase-port/tools/bin/glogsp": "60452cb49b78101b4deafd3f3efa3f49b19c6100939bdce427b5f935c938715f",
          "magphase-port/tools/bin/glsadf": "942fee269873ffccc61d74b4cc8d3d319efb77deccefd70439c4067da8b35bb8",
          "magphase-port/tools/bin/gmm": "bf2a075114349df929139dca0fdbddfd25eff88c99eab5cdb2ec4b5ec0d55ff6",
          "magphase-port/tools/bin/gmmp": "8e7e30f7ce1ac731e8db59cea060c89d75aaf0c4525b2548ca4d7c77a24b7eaa",
          "magphase-port/tools/bin/gnorm": "b7814955b7cf8e78b5dee07dac1416fb6059a84d91207b075490ba46cf4a30bf",
          "magphase-port/tools/bin/grlogsp": "9e17bb41a8ec045f8976f1e1027575c82627af17a1845699c27475c549cf8f19",
          "magphase-port/tools/bin/grpdelay": "cdbf349bb082749e35f1813c8d53e06c51ee8f44d8d80e32367ae2aab99a9830",
          "magphase-port/tools/bin/gseries": "288bae1d4b2c451f355878c10babeeb6dc1ffdb8966d35de8490f79695368ffa",
          "magphase-port/tools/bin/gwave": "6be98a889a70b98c39432d6aee76a7762aad614ccc9ab72740c51f48018fdf56",
          "magphase-port/tools/bin/histogram": "d9a6a5cba0c4606b56f521bdb740f32fc9515ac799b4952dda817247e1d491e8",
          "magphase-port/tools/bin/idct": "08e7efe8e14c763d59888da1582c406ebb995869e9ed5b4a37502b4550f8a4d8",
          "magphase-port/tools/bin/ifft": "7f1636d7a728c283872e5d23c4bf3d82f6c736517b82f280167a29ea79f56ad8",
          "magphase-port/tools/bin/ifft2": "9c524e9f97f473d5c799ed3fefe274ca97f2a55f82078f02573f4a3faf9b0d56",
          "magphase-port/tools/bin/ifftr": "e490066adb03bc26faa121881bd9332246c52ba001b54ed514be0540ca1d5cdd",
          "magphase-port/tools/bin/ignorm": "afbfaf4dc75256378b6bd5d828d58eaa08be3bf844a2b1203f12918551e73f83",
          "magphase-port/tools/bin/impulse": "e30f20abd9dab73b260ac7148943ec75a437daaf1614d3dc60b28680d4058843",
          "magphase-port/tools/bin/imsvq": "cf768f25268290632f248ba45e5d5a629b2632823990d4e5eac086a6127cd6ec",
          "magphase-port/tools/bin/interpolate": "cfb42c03d4f399f1940a4d792593c505c2afbbae4b5c18e1d483405c4fbd975c",
          "magphase-port/tools/bin/ivq": "81efaf6a11eb8b415b956bdcde5ab46f4a280cd1c68576c3a2619cf6a3d1926d",
          "magphase-port/tools/bin/lbg": "3e16cabf272d89d407dcaf99a7371a364440fd3be27fb4f328a3b5e01c182de4",
          "magphase-port/tools/bin/levdur": "29ae50e18ce8824cac169e1ec1de61e5dbb433f5315ca2a16d7376ca6916b039",
          "magphase-port/tools/bin/linear_intpl": "6794c493cd3809e157acd7a7d7ba71fde5ebc88f5df73d32c0d2f6f29db3f230",
          "magphase-port/tools/bin/lmadf": "d3baaae1058a8a9f5af78049130209f7838c70e4bee6a1c482eaca428978b15d",
          "magphase-port/tools/bin/lpc": "68bb10e385881eb0ad4b265c8621d038b1f3597b726232541260d504d451ad2e",
          "magphase-port/tools/bin/lpc2c": "aa0e9205c65b146bf9fbf5448abefc8d3e8923dfcd807c2553fa8e8a3b8f01ec",
          "magphase-port/tools/bin/lpc2lsp": "8c947b1e9d8b29406851d64359dbb49585ae82e3313d75ee3dd48dcdd7c3374b",
          "magphase-port/tools/bin/lpc2par": "d4053fa760cbc66bb728ea2cc1739c23fd6d9467ee09e9996aa6032a98bc6189",
          "magphase-port/tools/bin/lsp2lpc": "201b51f92ffc79b02d678a40ec458068f882c5b68f82c4f2e9c64a2202d8cc63",
          "magphase-port/tools/bin/lsp2sp": "adffb889caf0f9fe70f8aa14f08c183b3c4c812cd572268130e9efb9bb6725fa",
          "magphase-port/tools/bin/lspcheck": "b06790986e2adfc7dbb30759e855c65c2ebf5e138bc789852ff5cf885f61f610",
          "magphase-port/tools/bin/lspdf": "859638a6c09ef92b392bfe2f65abbb863a6c005cc2895ce1b9c30a50a9845faf",
          "magphase-port/tools/bin/ltcdf": "73653ad5f3de10845bb8812dcdb3fb8f4dba79f751ee6d0720e4791fb4e4f1f4",
          "magphase-port/tools/bin/mc2b": "1d3383e57cb44a6a9909d6bef203393a874e7fc0f69781f589969d1de4fbe1d3",
          "magphase-port/tools/bin/mcep": "4fbc8cf7426a0fdbed0fa40831bd2aa632cf924dca7cf9a690302594bdc46f40",
          "magphase-port/tools/bin/merge": "788ac6e3cefc4b0e28486944b7edf32790ec0ab90e6eaa92150db5e4a5d8a482",
          "magphase-port/tools/bin/mfcc": "646921d59adb06b4f5f9de7e3b3f105e7bfa8fb6b9e2d557730472d5afe65ffb",
          "magphase-port/tools/bin/mgc2mgc": "62bf524ec15ad49198206ce9d27bb25fdccd096eeffb77e403db7873a28b51c0",
          "magphase-port/tools/bin/mgc2mgclsp": "3dfa7f5377e2ea86644a70273db5bbae348628172f456d4ca71f7d8f9722e450",
          "magphase-port/tools/bin/mgc2sp": "216b14ab9f04e9c26ea41b981a8f518625291ff6593db772f43559561a3ed281",
          "magphase-port/tools/bin/mgcep": "7dac6de30fe85ffe45788d4ea9181f4faaf2c29d961be338b310697d4d63681c",
          "magphase-port/tools/bin/mgclsp2mgc": "14eacd029f883d307b583f7d30b59772477f9a32010dff4f02f65314eabed271",
          "magphase-port/tools/bin/mgclsp2sp": "f036552a4481972bad013f331cdf2aa060f081b165965acd27a6a6533e209505",
          "magphase-port/tools/bin/mglsadf": "830d1e888a5fd9c154152ab48d18353ccf9da75a6c5f438ee04b60388812389a",
          "magphase-port/tools/bin/minmax": "068f725ca9b2ce6165840f917776db406ce31c64f2dd0f96d6891ee752378217",
          "magphase-port/tools/bin/mlpg": "8e2463e42e630d7455f7ee1bdbdcdb04c107c42eeafd8c7bc6bc200eef07b5ab",
          "magphase-port/tools/bin/mlsacheck": "15c0ee2792f8ba14db2af60778006a941f8e5d75579c58a99be2e33dbdeb975b",
          "magphase-port/tools/bin/mlsadf": "1236ef0b9e24b2d630111b8eb5a0c2e2048addef3c8b109aafbbf81d75f97dff",
          "magphase-port/tools/bin/msvq": "46981501bcefaffb36d289f8d7b4b5cb7de660d99f71e3dd7dc51db87cbf1316",
          "magphase-port/tools/bin/nacorr": "f4766aa2f015f2a594e8b26aafb3a6962e96cd2741ea4671e827ba2afa6c8ed9",
          "magphase-port/tools/bin/nan": "2d41e6c83077bee3802386a057f8479ee09bc1edf635384b7a991c427195b83b",
          "magphase-port/tools/bin/ndps2c": "4fd438ef0f7d9d5539abd11aad4ee0b40aed59db98c4a52282dbb0b9190cca79",
          "magphase-port/tools/bin/norm0": "dada655cd695715961ef7ffa0183f0d78172e468b98aebee3a4a8d3102faec38",
          "magphase-port/tools/bin/nrand": "5f984449fa7e3106cdde7c345a1e52fc9b347ee6aa525d6207f3c729df10f46b",
          "magphase-port/tools/bin/par2lpc": "f0a036dbd4e8b59c5cae17c6d30a670f434253ddbe60b829bfe2bcbf2be6e92f",
          "magphase-port/tools/bin/pca": "e785ab05695cf48c96741e376988785a78690dec72c281801026a557162aca14",
          "magphase-port/tools/bin/pcas": "b7043e6fc3cf74fac614be763fa122944af6f8b90fd89bf9aec0e3c835cbe3a5",
          "magphase-port/tools/bin/phase": "95e1e5c744eef8ae3d572289a851dac1dfb4972012594395bca8de3102210a3e",
          "magphase-port/tools/bin/pitch": "b47a1923c41afb0c0c2a98e7d31ef103549c1cbead561458630ce574dc860896",
          "magphase-port/tools/bin/poledf": "386a6ac1d81c391652d7cd1c12fbb88eda7da375f234b7eb90b6355748316697",
          "magphase-port/tools/bin/psgr": "e5d3d2aa81463155a3ded9424bf2ca2aa252f4e95204584934ca40e209774de3",
          "magphase-port/tools/bin/ramp": "af26ccdd2fec68434b42651f590f38ef300b1077182f4843270a942e67617a3e",
          "magphase-port/tools/bin/raw2wav": "b01300b07cf5052d39c708eb9a09e353044a2aae10439105c4fe70b3804ea525",
          "magphase-port/tools/bin/rawtowav": "37b1b942c2594b23eb5096d95b82a11cc345cee193e4d39cf685879d5150644c",
          "magphase-port/tools/bin/reaper": "01640f6835cb86bb6a75ca8429b5caadf3de8ed37d2c9f82558301db417aed2b",
          "magphase-port/tools/bin/reverse": "442954de5cc79e69bd01dd0f2d1c4673d2eb5436b4189db39ec705725ee718fa",
          "magphase-port/tools/bin/rmse": "ee3eb9e0720494852ff7e173fa4a2157cde3e168bb354e5e54502790e3438f45",
          "magphase-port/tools/bin/root_pol": "75e9e2f96fdd4d60a2de67df7a112dd7e8b72ee3ca6e103138927a41be890a4d",
          "magphase-port/tools/bin/sin": "322fbd2ccfc237a36e5571d3d7790f44af1d0bfc5920c6ade1c4270c48e734a6",
          "magphase-port/tools/bin/smcep": "6d8e0442e3fb7c2888e8481943179c92b83d69b970058cb1deda832104e8de4e",
          "magphase-port/tools/bin/snr": "d4afb5620db8263b74127044d26af3f791904fa421385324fa1abbb5dfab1352",
          "magphase-port/tools/bin/sopr": "976fc7a13ea941300066bb9fef15f1013b996983f74d30675ad1274266bb6596",
          "magphase-port/tools/bin/spec": "7b662ea342aec02f2b06e8e603a7a6ccea6e06b30566d542c43f9377d443ebaf",
          "magphase-port/tools/bin/sptk": "e7e3b5f278e745c277c05a5e3203bb0a1a41087d5723840aa0d3049ca8b5bbd1",
          "magphase-port/tools/bin/step": "af78363bd8c080ec3a72124c1341a9af98dfe842b5afa0db9fec2365b0e8b047",
          "magphase-port/tools/bin/str": "c46a0d7a711751e03141c1f0b19475b57ccb5b7c6c543de2db6640967041d947",
          "magphase-port/tools/bin/swab": "ee7d6b0e3bcd35ed245ea22d1ba9576478422304397b8dfa9f25c400d2b42196",
          "magphase-port/tools/bin/symmetrize": "ef6790da8e1a9cf9b9a662d0c8535d4058d74a2fa42a6d01fa19ea77a034f89e",
          "magphase-port/tools/bin/train": "326f872acf949511494385e38e756e37d970c290b0c090320c8cf555a811bae1",
          "magphase-port/tools/bin/transpose": "cf9d418125ddcc6e40eb71a107ff7faa2ce7d47f0cd603f3f6d4b0210869060e",
          "magphase-port/tools/bin/uels": "149ea7d498f39b1ed8c9c53a26344bc8360c49932752b44c89bf07cff823c215",
          "magphase-port/tools/bin/ulaw": "f7afc82c2c4690932e067e52af3cb106e5327a3b95519b2725fff474e6ae2fbf",
          "magphase-port/tools/bin/us": "7d578ce3ed86e56406228b396d8a0f689f4bb2eb764ca6908106aeaaf9c8bbd8",
          "magphase-port/tools/bin/us16": "68f3b772ff4a32b4455f0ac493c3d2c21c89323b738482a8a5cdc0fd8920e504",
          "magphase-port/tools/bin/uscd": "6abcabe445a388256572a24a38f3bd585c419fc35de4e22d922982454029a5ca",
          "magphase-port/tools/bin/vc": "edc1e4d59dacf786e53504bcf63346796c94bfedc6b074d6fb70929ecae38386",
          "magphase-port/tools/bin/vopr": "2a6d408ae88b9f5bc027473f007802ea1530c3fac0560b0405fc2b2e129b810b",
          "magphase-port/tools/bin/vq": "95ad7257d3b7bea72876d8759e6c49812a4d45bf9f6851fb2b13fa427e536dc3",
          "magphase-port/tools/bin/vstat": "03a3ef7e632de5a10b6c4f999c3ad9dbcfb18102de070831ef107146a657bc4e",
          "magphase-port/tools/bin/vsum": "e218676a8336e8c653ebac67cff178165c2f66375f15cce6abe806145a63c3b1",
          "magphase-port/tools/bin/wav2raw": "b8527b413a56256bac2835925b274f9f57d6700fb439385ad7bcd32bfea06687",
          "magphase-port/tools/bin/wavjoin": "676f9b222e5f722191753c1fa485df1de6aafc114dcb85fac6b14fe7fd682359",
          "magphase-port/tools/bin/wavsplit": "fbeafafd81a921a5bceec6ebadc5ad09f428f17e7b73681dab42b31713c2d5f4",
          "magphase-port/tools/bin/window": "3966223d56088569e9e655170b334def9688364c2f9058f415ad1cc5857cdd04",
          "magphase-port/tools/bin/x2x": "0979e5e94c5d4ae2eaf8e891d611e40e43101b7b84464dafb1b99d9ee7729e02",
          "magphase-port/tools/bin/xgr": "69e33f237e18d04ebedeb9d4ebccebc4fb30d54fe9e8e2090bb5908cc600da8b",
          "magphase-port/tools/bin/zcross": "cebcfa6c2d3111ea3ca277cc362acd3a72ba037149b61c6ba00811c403562f0a",
          "magphase-port/tools/bin/zerodf": "0d52228cd34542b501e0be48db04e717a105c11ce384bac6092bf1cc3ebb68f8"
        },
        "sources": {
          "magphase": {
            "repository": "https://github.com/CSTR-Edinburgh/magphase.git",
            "revision": "ec6ae285c3128d3f416857f05d2a7e4be0bd5707"
          },
          "reaper": {
            "repository": "https://github.com/google/REAPER.git",
            "revision": "1d6e9b95e6b08b500fccbc9a043989dbda747276"
          }
        },
        "note": "上流コードと重みをキャッシュへ取得。第三者コードを本リポジトリに複製しない。"
      }
    },
    "native_sample_rates": {
      "world_harvest": 16000,
      "facodec": 16000,
      "hifiglot": 44100,
      "golf": 24000,
      "sifigan": 24000,
      "llsm2": 24000,
      "magphase": 48000
    },
    "evaluation": "native-full-MOS + strict-length + original-rate-common-prefix-v1",
    "runtime_python": "/home/runner/work/check-nonml-saikyo-vocoder/check-nonml-saikyo-vocoder/.cache/references/env/bin/python",
    "host_python": "3.12.3"
  },
  "execution": {
    "started_at": "2026-10-12T11:37:45.778718+00:00",
    "repository_commit": "6f01a683f11d590ca85938b1fae091ec85d2428a"
  },
  "resynthesis": [
    {
      "case": "smoke_arctic",
      "method": "world_harvest",
      "status": "measured",
      "started_at": "2026-10-12T11:37:48.958150+00:00",
      "native_sample_rate": 16000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.003323128999994651,
      "native_input_samples": 53600,
      "native_input_sha256": "009a7e8ad1682c8688445329d05d765678e600abb0539d65e68282173e082c6f",
      "output_sha256": "bf7a652faba5f367fb234e24c49c3cd8a1519a85d80f4ef6531176cf887f09b3",
      "output_samples": 53600,
      "length_delta_samples": 0,
      "analysis_s": 0.7094794960000002,
      "synthesis_s": 0.05369671700000822,
      "total_s": 0.7631762130000084,
      "rtf": 0.22781379492537562,
      "output_peak": 0.32031564700733184,
      "output_rms": 0.04469001053221412,
      "finished_at": "2026-10-12T11:37:49.728174+00:00",
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "arctic",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 147734,
        "output_samples": 147735,
        "length_delta_samples": 1,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -26.991771868243454,
        "output_peak": 0.32096343818409534
      },
      "common_prefix": {
        "samples": 147734,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 0.9999932311232951,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 147734,
          "output_samples": 147734,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 14.81950551201502,
              "spectral_convergence": 0.45990603998031715,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 7.040018759632789e-07
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 12.834918644610848,
              "spectral_convergence": 0.22443782979088564,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.852450084524139e-07
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 11.496181356131968,
              "spectral_convergence": 0.2512459058956549,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.766215533669349e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 0.00021192180604849,
                "output_power": 0.00024215567792329037,
                "delta_db": 0.5791901011566339,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.0007506719378509437,
                "output_power": 0.0008907706226267766,
                "delta_db": 0.7431570461761455,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.0005447977950658482,
                "output_power": 0.0007180293213980116,
                "delta_db": 1.199068382163119,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 9.431507243494021e-05,
                "output_power": 0.00011581073028641326,
                "delta_db": 0.891676975995624,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 1.3572749160000328e-05,
                "output_power": 1.6397245146349068e-05,
                "delta_db": 0.8210306671253182,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.7343096646047786e-05,
                "output_power": 1.5731162391033775e-05,
                "delta_db": -0.4236582999403612,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 1.0821264092698321e-05,
                "output_power": 1.6482566883418016e-07,
                "delta_db": -18.17253149402549,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 6.343414671079065e-11,
                "output_power": 6.287469393776286e-11,
                "delta_db": -0.038472182919799286,
                "reference_active": false
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 7,
            "band_count": 8,
            "active_band_mae_db": 3.261473280940384,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": 0.850708995786134,
          "pitch": null,
          "output_rms_dbfs": -26.9917424712863,
          "output_peak": 0.32096343818409534,
          "reference_rms_dbfs": -27.842451467072436,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 13.050201837585945
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 147734,
          "output_samples": 147734,
          "intelligibility": {
            "estoi": {
              "value": 0.9571978357834486,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04053941026622261,
              "rms_dbfs": -27.842451467072436,
              "peak": 0.232452392578125,
              "peak_dbfs": -12.673319587326667,
              "crest_factor_db": 15.169131879745771,
              "mean": 1.0310790979268482e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.044710844862464884,
              "rms_dbfs": -26.9917424712863,
              "peak": 0.32096343818409534,
              "peak_dbfs": -9.870888728802766,
              "crest_factor_db": 17.120853742483536,
              "mean": -1.6963730873061194e-09,
              "zero_energy": false
            },
            "rms_difference_db": 0.8507089957861353
          },
          "full_scale": {
            "reference": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 334,
            "uncovered_tail_samples": 440,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 1.1315037067683916e-06,
            "threshold_dbfs": -59.46344018972689,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 41,
            "quiet_samples": 18081,
            "reference_mean_power": 5.009816483048026e-07,
            "output_mean_power": 1.5709082952324152e-06,
            "output_rms_dbfs": -58.03849166988663,
            "output_zero_energy": false,
            "output_peak": 0.03099043370585131,
            "output_energy_fraction": 9.617607947835265e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 327,
            "reference_estimated_voiced_frames": 184,
            "output_estimated_voiced_frames": 188,
            "both_estimated_voiced_frames": 178,
            "reference_voiced_output_unvoiced_frames": 6,
            "reference_unvoiced_output_voiced_frames": 10,
            "paired_voiced_fraction": 0.967391304347826,
            "absolute_difference_cents_median": 4.615487763393991,
            "absolute_difference_cents_p95": 23.02465655518455,
            "difference_over_600_cents_frames": 0,
            "uncovered_tail_samples": 440,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 147734,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 0.00010258359777065965,
                      "log_rms_mae_db": 2.5892912288839844,
                      "rms_envelope_nrmse": 0.15110895324768175,
                      "envelope_correlation": 0.9873679893676504,
                      "rms_gain_error_db": 0.8509815438148348,
                      "reference_mean_power": 0.0016470512022936088,
                      "output_mean_power": 0.002003573388658432
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.737071418318221e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.0004181860744866353,
                          "output_power": 0.0005416313253894939,
                          "reference_active": true,
                          "delta_db": 1.1233420713620792
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.655525977123722e-05,
                          "output_power": 0.00010411652728796673,
                          "reference_active": true,
                          "delta_db": 0.802262100553698
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.295514528933935e-06,
                          "output_power": 6.476884275559677e-06,
                          "reference_active": true,
                          "delta_db": 0.1233490690985118
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.6829844136714297
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 9.51361232071073e-05,
                      "log_rms_mae_db": 1.9107704373227123,
                      "rms_envelope_nrmse": 0.13341689375668794,
                      "envelope_correlation": 0.9918626501720911,
                      "rms_gain_error_db": 0.8510860688613353,
                      "reference_mean_power": 0.0016619096793530144,
                      "output_mean_power": 0.0020216968022128334
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.993659628933029e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00039685314515544806,
                          "output_power": 0.0005196663541866435,
                          "reference_active": true,
                          "delta_db": 1.1709477278187495
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.1281752657875335e-05,
                          "output_power": 4.854899290469858e-05,
                          "reference_active": true,
                          "delta_db": 0.7042209796269099
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.147467230270734e-07,
                          "output_power": 6.661245127222294e-07,
                          "reference_active": true,
                          "delta_db": -0.3059675732228594
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7270454268895062
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 7.858062960145444e-05,
                      "log_rms_mae_db": 2.0070661548760227,
                      "rms_envelope_nrmse": 0.14507324280407766,
                      "envelope_correlation": 0.9945131337228454,
                      "rms_gain_error_db": 0.9794723419393276,
                      "reference_mean_power": 0.0008478670875848599,
                      "output_mean_power": 0.0010623680751489471
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 4.118572979449278e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00032611065091443195,
                          "output_power": 0.00042052013914261997,
                          "reference_active": true,
                          "delta_db": 1.1042181615948863
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.0919247362893804e-05,
                          "output_power": 6.708165618669606e-05,
                          "reference_active": true,
                          "delta_db": 1.19721800687852
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.838294986205235e-06,
                          "output_power": 4.613879607238956e-06,
                          "reference_active": true,
                          "delta_db": -0.20626085327732788
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.8358990072502448
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 7.126618944542506e-05,
                      "log_rms_mae_db": 1.66036779090428,
                      "rms_envelope_nrmse": 0.1343029145722705,
                      "envelope_correlation": 0.9966343427216556,
                      "rms_gain_error_db": 0.9795565718951103,
                      "reference_mean_power": 0.0008555353845487757,
                      "output_mean_power": 0.0010719971571824603
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 3.7040306770976235e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00031681362834561343,
                          "output_power": 0.00041027548450354957,
                          "reference_active": true,
                          "delta_db": 1.1227171169167507
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.473236521731313e-05,
                          "output_power": 3.23085436107157e-05,
                          "reference_active": true,
                          "delta_db": 1.1605173053333198
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.047139621751533e-07,
                          "output_power": 3.960534581672452e-07,
                          "reference_active": true,
                          "delta_db": -1.0529150930917144
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.1120498384472617
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 5.240498613206585e-05,
                      "log_rms_mae_db": 1.4534830042833033,
                      "rms_envelope_nrmse": 0.14827311930082787,
                      "envelope_correlation": 0.994222997729622,
                      "rms_gain_error_db": 0.9639028356190783,
                      "reference_mean_power": 9.343683802972147e-05,
                      "output_mean_power": 0.00011665635882311665
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 2.735946251376838e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8948025909514686e-05,
                          "output_power": 2.417677409997102e-05,
                          "reference_active": true,
                          "delta_db": 1.0583438269208356
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.263241347636967e-06,
                          "output_power": 3.004021776277994e-06,
                          "reference_active": true,
                          "delta_db": 1.2297220782262803
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.834787408003976e-07,
                          "output_power": 8.238645641717678e-07,
                          "reference_active": true,
                          "delta_db": 0.2182860685003556
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.8354506578824905
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 4.4293590932304836e-05,
                      "log_rms_mae_db": 1.1473865216158918,
                      "rms_envelope_nrmse": 0.12928194772271745,
                      "envelope_correlation": 0.9977941049098276,
                      "rms_gain_error_db": 0.963283654805639,
                      "reference_mean_power": 9.429068361359236e-05,
                      "output_mean_power": 0.0001177056066189313
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 2.4918519689831095e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8481977963071213e-05,
                          "output_power": 2.3627138540833105e-05,
                          "reference_active": true,
                          "delta_db": 1.0666267976701307
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.0691683146494508e-06,
                          "output_power": 1.3506515270938242e-06,
                          "reference_active": true,
                          "delta_db": 1.014972342146451
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.6063349891805814e-08,
                          "output_power": 3.5102629262021626e-08,
                          "reference_active": true,
                          "delta_db": -1.1801587095722206
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.0872526164629341
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 1.2860287854104408e-05,
                      "log_rms_mae_db": 1.9062796198247656,
                      "rms_envelope_nrmse": 0.24172793015876323,
                      "envelope_correlation": 0.9441853082772157,
                      "rms_gain_error_db": 0.47394400669466463,
                      "reference_mean_power": 9.424710941015776e-06,
                      "output_mean_power": 1.0511445424700352e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.625314523028166e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8781674075282524e-06,
                          "output_power": 1.9156942131885586e-06,
                          "reference_active": true,
                          "delta_db": 0.08591887532418666
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.6792080290845895e-06,
                          "output_power": 1.6279672738444566e-06,
                          "reference_active": true,
                          "delta_db": -0.13458831894600334
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.330459106871231e-07,
                          "output_power": 4.341444305971277e-07,
                          "reference_active": true,
                          "delta_db": -0.8913038190065277
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.3706036710922392
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 8.463536202804461e-06,
                      "log_rms_mae_db": 1.2608194589094692,
                      "rms_envelope_nrmse": 0.14510858148984707,
                      "envelope_correlation": 0.9777741293360636,
                      "rms_gain_error_db": 0.4750832160465795,
                      "reference_mean_power": 9.508840352637213e-06,
                      "output_mean_power": 1.0608057803285293e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.426890844734256e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.627718067426583e-06,
                          "output_power": 1.6578649841685355e-06,
                          "reference_active": true,
                          "delta_db": 0.07969974780015923
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 9.572772525291917e-07,
                          "output_power": 8.691170376611786e-07,
                          "reference_active": true,
                          "delta_db": -0.4195947537455413
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.121992613475366e-08,
                          "output_power": 3.20349191490002e-08,
                          "reference_active": true,
                          "delta_db": -1.094835769576341
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5313767570406805
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_arctic",
      "method": "facodec",
      "status": "measured",
      "started_at": "2026-10-12T11:37:57.930155+00:00",
      "native_sample_rate": 16000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.003356862999993382,
      "native_input_samples": 53600,
      "native_input_sha256": "009a7e8ad1682c8688445329d05d765678e600abb0539d65e68282173e082c6f",
      "output_sha256": "14a821bffda99fdd99e97f8db032c6b3a31183d5bdcbc9e2546c5e7c4d6e7a45",
      "output_samples": 53600,
      "length_delta_samples": 0,
      "analysis_s": 0.7581828630000018,
      "synthesis_s": 2.6250750039999957,
      "total_s": 3.3832578669999975,
      "rtf": 1.0099277214925366,
      "output_peak": 0.2541073262691498,
      "output_rms": 0.03835902456646158,
      "finished_at": "2026-10-12T11:38:01.320236+00:00",
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "arctic",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 147734,
        "output_samples": 147735,
        "length_delta_samples": 1,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -28.3205342169301,
        "output_peak": 0.2538327042841333
      },
      "common_prefix": {
        "samples": 147734,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 0.9999932311232951,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 147734,
          "output_samples": 147734,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 15.108145553016405,
              "spectral_convergence": 0.18869840897804405,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 7.040018759632789e-07
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 13.557197858236842,
              "spectral_convergence": 0.1870012486258208,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.852450084524139e-07
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 12.102053586776167,
              "spectral_convergence": 0.19730348970981543,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.766215533669349e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 0.00021192180604849,
                "output_power": 0.00021452165041276045,
                "delta_db": 0.052954830623855656,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.0007506719378509437,
                "output_power": 0.0006667271282663164,
                "delta_db": -0.515020546994851,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.0005447977950658482,
                "output_power": 0.00048621443098380697,
                "delta_db": -0.49407496525494504,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 9.431507243494021e-05,
                "output_power": 7.765504728618052e-05,
                "delta_db": -0.8441141413248598,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 1.3572749160000328e-05,
                "output_power": 1.26429248864982e-05,
                "delta_db": -0.30820265168574995,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.7343096646047786e-05,
                "output_power": 1.42507615916442e-05,
                "delta_db": -0.8528856964373521,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 1.0821264092698321e-05,
                "output_power": 1.293818173849458e-07,
                "delta_db": -19.224047488712927,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 6.343414671079065e-11,
                "output_power": 4.679799393840456e-11,
                "delta_db": -1.320958654653746,
                "reference_active": false
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 7,
            "band_count": 8,
            "active_band_mae_db": 3.1844714744335056,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": -0.47805335309030206,
          "pitch": null,
          "output_rms_dbfs": -28.32050482016274,
          "output_peak": 0.2538327042841333,
          "reference_rms_dbfs": -27.842451467072436,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 13.58913233267647
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 147734,
          "output_samples": 147734,
          "intelligibility": {
            "estoi": {
              "value": 0.9449384079301993,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04053941026622261,
              "rms_dbfs": -27.842451467072436,
              "peak": 0.232452392578125,
              "peak_dbfs": -12.673319587326667,
              "crest_factor_db": 15.169131879745771,
              "mean": 1.0310790979268482e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.03836849452405692,
              "rms_dbfs": -28.32050482016274,
              "peak": 0.2538327042841333,
              "peak_dbfs": -11.909048466331004,
              "crest_factor_db": 16.411456353831735,
              "mean": 9.301877133608924e-06,
              "zero_energy": false
            },
            "rms_difference_db": -0.47805335309030283
          },
          "full_scale": {
            "reference": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 334,
            "uncovered_tail_samples": 440,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 1.1315037067683916e-06,
            "threshold_dbfs": -59.46344018972689,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 41,
            "quiet_samples": 18081,
            "reference_mean_power": 5.009816483048026e-07,
            "output_mean_power": 6.837278909748252e-07,
            "output_rms_dbfs": -61.651167037778905,
            "output_zero_energy": false,
            "output_peak": 0.0027567861815530528,
            "output_energy_fraction": 5.68428399294372e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 327,
            "reference_estimated_voiced_frames": 184,
            "output_estimated_voiced_frames": 186,
            "both_estimated_voiced_frames": 182,
            "reference_voiced_output_unvoiced_frames": 2,
            "reference_unvoiced_output_voiced_frames": 4,
            "paired_voiced_fraction": 0.9891304347826086,
            "absolute_difference_cents_median": 3.011258631603317,
            "absolute_difference_cents_p95": 11.345912144947622,
            "difference_over_600_cents_frames": 0,
            "uncovered_tail_samples": 440,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 147734,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 0.00010258359777065965,
                      "log_rms_mae_db": 1.4628308858146586,
                      "rms_envelope_nrmse": 0.09768996820529703,
                      "envelope_correlation": 0.9918512111934735,
                      "rms_gain_error_db": -0.47778419919000864,
                      "reference_mean_power": 0.0016470512022936088,
                      "output_mean_power": 0.0014754642137399685
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.737071418318221e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.0004181860744866353,
                          "output_power": 0.00037128401693128167,
                          "reference_active": true,
                          "delta_db": -0.5166331272279352
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.655525977123722e-05,
                          "output_power": 7.761836091534874e-05,
                          "reference_active": true,
                          "delta_db": -0.4732899672102858
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.295514528933935e-06,
                          "output_power": 6.5525652890091e-06,
                          "reference_active": true,
                          "delta_db": 0.17380126357438938
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.3879081193375368
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 9.51361232071073e-05,
                      "log_rms_mae_db": 1.2698024639041068,
                      "rms_envelope_nrmse": 0.09126436787071564,
                      "envelope_correlation": 0.9925017635698516,
                      "rms_gain_error_db": -0.47791945268397285,
                      "reference_mean_power": 0.0016619096793530144,
                      "output_mean_power": 0.001488728395381073
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.993659628933029e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00039685314515544806,
                          "output_power": 0.00035388541030753243,
                          "reference_active": true,
                          "delta_db": -0.49767168266234085
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.1281752657875335e-05,
                          "output_power": 3.7134282946594994e-05,
                          "reference_active": true,
                          "delta_db": -0.4598308519643731
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.147467230270734e-07,
                          "output_power": 7.665256879510485e-07,
                          "reference_active": true,
                          "delta_db": 0.30374540711567877
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.4204159805807976
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 7.858062960145444e-05,
                      "log_rms_mae_db": 1.4761595671807792,
                      "rms_envelope_nrmse": 0.09493599486137144,
                      "envelope_correlation": 0.9936578980435306,
                      "rms_gain_error_db": -0.43010042227817064,
                      "reference_mean_power": 0.0008478670875848599,
                      "output_mean_power": 0.000767923105498391
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 4.118572979449278e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00032611065091443195,
                          "output_power": 0.0002759951553961301,
                          "reference_active": true,
                          "delta_db": -0.7246352441733803
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.0919247362893804e-05,
                          "output_power": 4.490314434515439e-05,
                          "reference_active": true,
                          "delta_db": -0.5460522216163124
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.838294986205235e-06,
                          "output_power": 4.589071476517915e-06,
                          "reference_active": true,
                          "delta_db": -0.2296752130355464
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5001208929417463
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 7.126618944542506e-05,
                      "log_rms_mae_db": 1.2724384576764154,
                      "rms_envelope_nrmse": 0.08571319442049666,
                      "envelope_correlation": 0.9947556688113763,
                      "rms_gain_error_db": -0.43009109161957815,
                      "reference_mean_power": 0.0008555353845487757,
                      "output_mean_power": 0.0007748700362323842
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 3.7040306770976235e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00031681362834561343,
                          "output_power": 0.00026755839479737875,
                          "reference_active": true,
                          "delta_db": -0.7338527356770986
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.473236521731313e-05,
                          "output_power": 2.2300692294661796e-05,
                          "reference_active": true,
                          "delta_db": -0.44947305664850884
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.047139621751533e-07,
                          "output_power": 4.859796486178782e-07,
                          "reference_active": true,
                          "delta_db": -0.16427236340859963
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.44919938524473574
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 5.240498613206585e-05,
                      "log_rms_mae_db": 1.3805467925296449,
                      "rms_envelope_nrmse": 0.14768937127079623,
                      "envelope_correlation": 0.9888554949739563,
                      "rms_gain_error_db": -0.867115949662791,
                      "reference_mean_power": 9.343683802972147e-05,
                      "output_mean_power": 7.652556384880694e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 2.735946251376838e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8948025909514686e-05,
                          "output_power": 1.644097456841282e-05,
                          "reference_active": true,
                          "delta_db": -0.6163641235466779
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.263241347636967e-06,
                          "output_power": 1.6969695529563829e-06,
                          "reference_active": true,
                          "delta_db": -1.250568184676445
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.834787408003976e-07,
                          "output_power": 9.129957896334391e-07,
                          "reference_active": true,
                          "delta_db": 0.6644155809010209
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.843782629708048
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 4.4293590932304836e-05,
                      "log_rms_mae_db": 1.254940851112486,
                      "rms_envelope_nrmse": 0.13077357558201932,
                      "envelope_correlation": 0.9924842854868091,
                      "rms_gain_error_db": -0.8676682043414081,
                      "reference_mean_power": 9.429068361359236e-05,
                      "output_mean_power": 7.721505123956954e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 2.4918519689831095e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8481977963071213e-05,
                          "output_power": 1.603452523984727e-05,
                          "reference_active": true,
                          "delta_db": -0.6169234248397599
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.0691683146494508e-06,
                          "output_power": 7.617159779790312e-07,
                          "reference_active": true,
                          "delta_db": -1.4725301415774092
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.6063349891805814e-08,
                          "output_power": 5.127375024581464e-08,
                          "reference_active": true,
                          "delta_db": 0.46539565382574855
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.8516164067476392
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 1.2860287854104408e-05,
                      "log_rms_mae_db": 1.8191543656329277,
                      "rms_envelope_nrmse": 0.19022310373417242,
                      "envelope_correlation": 0.9626791569029383,
                      "rms_gain_error_db": -0.40522916444999824,
                      "reference_mean_power": 9.424710941015776e-06,
                      "output_mean_power": 8.585095343781002e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.625314523028166e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8781674075282524e-06,
                          "output_power": 1.7273086108048239e-06,
                          "reference_active": true,
                          "delta_db": -0.363643617897424
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.6792080290845895e-06,
                          "output_power": 1.5220625745918215e-06,
                          "reference_active": true,
                          "delta_db": -0.4267199475638398
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.330459106871231e-07,
                          "output_power": 5.120992270496716e-07,
                          "reference_active": true,
                          "delta_db": -0.17410495646102989
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.3214895073074312
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 8.463536202804461e-06,
                      "log_rms_mae_db": 1.4233099952418196,
                      "rms_envelope_nrmse": 0.13387365379560034,
                      "envelope_correlation": 0.9784449174354545,
                      "rms_gain_error_db": -0.40589935943009114,
                      "reference_mean_power": 9.508840352637213e-06,
                      "output_mean_power": 8.66039339232699e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.426890844734256e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.627718067426583e-06,
                          "output_power": 1.5166579143488868e-06,
                          "reference_active": true,
                          "delta_db": -0.30691548391821993
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 9.572772525291917e-07,
                          "output_power": 8.785122473545741e-07,
                          "reference_active": true,
                          "delta_db": -0.37289918655522236
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.121992613475366e-08,
                          "output_power": 3.8766970671540654e-08,
                          "reference_active": true,
                          "delta_db": -0.26645343501755486
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.3154227018303324
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_arctic",
      "method": "hifiglot",
      "status": "measured",
      "started_at": "2026-10-12T11:38:10.465924+00:00",
      "native_sample_rate": 44100,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 1.3053999992962417e-05,
      "native_input_samples": 147734,
      "native_input_sha256": "bd719b730a7005fc41842c648e42de53638f4d1a1f791be8da1a831d38988742",
      "output_sha256": "6f965336992754e6d48adc0bafec25754339bdd52d038cf51a2ab36da7f44392",
      "output_samples": 147968,
      "length_delta_samples": 234,
      "analysis_s": 0.07701992400001245,
      "synthesis_s": 10.522214810999998,
      "total_s": 10.59923473500001,
      "rtf": 3.1639720837011143,
      "output_peak": 0.24100691080093384,
      "output_rms": 0.03733230802729508,
      "finished_at": "2026-10-12T11:38:21.070393+00:00",
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "arctic",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 147734,
        "output_samples": 147968,
        "length_delta_samples": 234,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -28.558303189669623,
        "output_peak": 0.24100691080093384
      },
      "common_prefix": {
        "samples": 147734,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 0.9984185769896193,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 147734,
          "output_samples": 147734,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 9.784330178935472,
              "spectral_convergence": 0.4661262646080092,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 7.040018759632789e-07
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 9.263860256300411,
              "spectral_convergence": 0.29791160306340575,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.852450084524139e-07
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 8.262440191887537,
              "spectral_convergence": 0.3498319205619211,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.766215533669349e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 0.00021192180604849,
                "output_power": 0.00019093445185316754,
                "delta_db": -0.4529134772336396,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.0007506719378509437,
                "output_power": 0.0007112719257243159,
                "delta_db": -0.23414513861167138,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.0005447977950658482,
                "output_power": 0.000371305414215435,
                "delta_db": -1.6650405924173939,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 9.431507243494021e-05,
                "output_power": 8.515453676353354e-05,
                "delta_db": -0.4437331187629537,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 1.3572749160000328e-05,
                "output_power": 1.1625415574865572e-05,
                "delta_db": -0.6725933629936854,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.7343096646047786e-05,
                "output_power": 1.814379712984249e-05,
                "delta_db": 0.196015370403202,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 1.0821264092698321e-05,
                "output_power": 7.419979010525117e-06,
                "delta_db": -1.6387531935608288,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 6.343414671079065e-11,
                "output_power": 5.317519239925485e-08,
                "delta_db": 29.23385967651332,
                "reference_active": false
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 7,
            "band_count": 8,
            "active_band_mae_db": 0.757599179140482,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": -0.7089784070254785,
          "pitch": null,
          "output_rms_dbfs": -28.551429874097916,
          "output_peak": 0.24100691080093384,
          "reference_rms_dbfs": -27.842451467072436,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 9.103543542374473
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 147734,
          "output_samples": 147734,
          "intelligibility": {
            "estoi": {
              "value": 0.828710129223041,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04053941026622261,
              "rms_dbfs": -27.842451467072436,
              "peak": 0.232452392578125,
              "peak_dbfs": -12.673319587326667,
              "crest_factor_db": 15.169131879745771,
              "mean": 1.0310790979268482e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.0373618615096208,
              "rms_dbfs": -28.551429874097916,
              "peak": 0.24100691080093384,
              "peak_dbfs": -12.359410079649546,
              "crest_factor_db": 16.192019794448367,
              "mean": -7.510051673474398e-05,
              "zero_energy": false
            },
            "rms_difference_db": -0.7089784070254801
          },
          "full_scale": {
            "reference": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 334,
            "uncovered_tail_samples": 440,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 1.1315037067683916e-06,
            "threshold_dbfs": -59.46344018972689,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 41,
            "quiet_samples": 18081,
            "reference_mean_power": 5.009816483048026e-07,
            "output_mean_power": 4.661384496468513e-07,
            "output_rms_dbfs": -63.314850726075356,
            "output_zero_energy": false,
            "output_peak": 0.005348001141101122,
            "output_energy_fraction": 4.086955454569798e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 327,
            "reference_estimated_voiced_frames": 184,
            "output_estimated_voiced_frames": 191,
            "both_estimated_voiced_frames": 178,
            "reference_voiced_output_unvoiced_frames": 6,
            "reference_unvoiced_output_voiced_frames": 13,
            "paired_voiced_fraction": 0.967391304347826,
            "absolute_difference_cents_median": 6.788082198884432,
            "absolute_difference_cents_p95": 40.76079817370715,
            "difference_over_600_cents_frames": 0,
            "uncovered_tail_samples": 440,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 147734,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 0.00010258359777065965,
                      "log_rms_mae_db": 1.5558256408268833,
                      "rms_envelope_nrmse": 0.11227907200179496,
                      "envelope_correlation": 0.9920072807530015,
                      "rms_gain_error_db": -0.7089391740598178,
                      "reference_mean_power": 0.0016470512022936088,
                      "output_mean_power": 0.0013989854026906427
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.737071418318221e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.0004181860744866353,
                          "output_power": 0.0003533824367641938,
                          "reference_active": true,
                          "delta_db": -0.7312460568910806
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.655525977123722e-05,
                          "output_power": 7.146971325627935e-05,
                          "reference_active": true,
                          "delta_db": -0.8317142445485324
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.295514528933935e-06,
                          "output_power": 9.20938136329459e-06,
                          "reference_active": true,
                          "delta_db": 1.6519922706614496
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.071650857367021
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 9.51361232071073e-05,
                      "log_rms_mae_db": 1.1498658631662642,
                      "rms_envelope_nrmse": 0.09190633484863381,
                      "envelope_correlation": 0.9968407895222436,
                      "rms_gain_error_db": -0.7088737654733,
                      "reference_mean_power": 0.0016619096793530144,
                      "output_mean_power": 0.001411627273879232
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.993659628933029e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00039685314515544806,
                          "output_power": 0.00033543181310088726,
                          "reference_active": true,
                          "delta_db": -0.7302557686201137
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.1281752657875335e-05,
                          "output_power": 3.435008698808726e-05,
                          "reference_active": true,
                          "delta_db": -0.7983028621144848
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.147467230270734e-07,
                          "output_power": 8.616136964171678e-07,
                          "reference_active": true,
                          "delta_db": 0.8116042111369569
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7800542806238518
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 7.858062960145444e-05,
                      "log_rms_mae_db": 2.510109677967643,
                      "rms_envelope_nrmse": 0.18261871291022352,
                      "envelope_correlation": 0.9825543883610621,
                      "rms_gain_error_db": -1.1968202661592977,
                      "reference_mean_power": 0.0008478670875848599,
                      "output_mean_power": 0.0006436440376904559
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 4.118572979449278e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00032611065091443195,
                          "output_power": 0.00026382064824288933,
                          "reference_active": true,
                          "delta_db": -0.9205620015784423
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.0919247362893804e-05,
                          "output_power": 3.8591306381081404e-05,
                          "reference_active": true,
                          "delta_db": -1.2039249530079041
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.838294986205235e-06,
                          "output_power": 6.6371147324105925e-06,
                          "reference_active": true,
                          "delta_db": 1.3728698178256329
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.1657855908039931
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 7.126618944542506e-05,
                      "log_rms_mae_db": 1.8369992535079689,
                      "rms_envelope_nrmse": 0.16510889466751935,
                      "envelope_correlation": 0.9877586238487975,
                      "rms_gain_error_db": -1.1967272146814885,
                      "reference_mean_power": 0.0008555353845487757,
                      "output_mean_power": 0.00064947921186734
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 3.7040306770976235e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00031681362834561343,
                          "output_power": 0.0002561350055008981,
                          "reference_active": true,
                          "delta_db": -0.923349185749248
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.473236521731313e-05,
                          "output_power": 1.954486989446311e-05,
                          "reference_active": true,
                          "delta_db": -1.0223286722068878
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.047139621751533e-07,
                          "output_power": 6.037343668771319e-07,
                          "reference_active": true,
                          "delta_db": 0.7780057924287
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.9078945501282787
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 5.240498613206585e-05,
                      "log_rms_mae_db": 2.981730098786596,
                      "rms_envelope_nrmse": 0.2535492993729409,
                      "envelope_correlation": 0.9512795150997416,
                      "rms_gain_error_db": -0.48345985010088577,
                      "reference_mean_power": 9.343683802972147e-05,
                      "output_mean_power": 8.359343050592016e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 2.735946251376838e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8948025909514686e-05,
                          "output_power": 1.2870479585573349e-05,
                          "reference_active": true,
                          "delta_db": -1.6796923982888399
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.263241347636967e-06,
                          "output_power": 2.1988951834181275e-06,
                          "reference_active": true,
                          "delta_db": -0.1252634076487747
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.834787408003976e-07,
                          "output_power": 4.6577289413317294e-07,
                          "reference_active": true,
                          "delta_db": -2.2585300569084277
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.3544952876153473
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 4.4293590932304836e-05,
                      "log_rms_mae_db": 2.5449683401365366,
                      "rms_envelope_nrmse": 0.22032910426952884,
                      "envelope_correlation": 0.9616770571406632,
                      "rms_gain_error_db": -0.48422559710864144,
                      "reference_mean_power": 9.429068361359236e-05,
                      "output_mean_power": 8.434245239790389e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 2.4918519689831095e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8481977963071213e-05,
                          "output_power": 1.2339685878140839e-05,
                          "reference_active": true,
                          "delta_db": -1.7544434374032436
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.0691683146494508e-06,
                          "output_power": 1.0914324284472967e-06,
                          "reference_active": true,
                          "delta_db": 0.08950773583843878
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.6063349891805814e-08,
                          "output_power": 3.3094055411093286e-08,
                          "reference_active": true,
                          "delta_db": -1.4360552850023434
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.093335486081342
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 1.2860287854104408e-05,
                      "log_rms_mae_db": 4.340878409213484,
                      "rms_envelope_nrmse": 0.46055362045923387,
                      "envelope_correlation": 0.8028753281309269,
                      "rms_gain_error_db": -0.36365264080639675,
                      "reference_mean_power": 9.424710941015776e-06,
                      "output_mean_power": 8.667678118292942e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.625314523028166e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8781674075282524e-06,
                          "output_power": 1.9608322088117793e-06,
                          "reference_active": true,
                          "delta_db": 0.1870613224607675
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.6792080290845895e-06,
                          "output_power": 1.823859745958865e-06,
                          "reference_active": true,
                          "delta_db": 0.3588693606912311
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.330459106871231e-07,
                          "output_power": 5.549917884712523e-07,
                          "reference_active": true,
                          "delta_db": 0.17521941483307105
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.2403833659950232
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 8.463536202804461e-06,
                      "log_rms_mae_db": 3.9324458101060045,
                      "rms_envelope_nrmse": 0.39935485341206145,
                      "envelope_correlation": 0.8260365153068502,
                      "rms_gain_error_db": -0.36294701883011765,
                      "reference_mean_power": 9.508840352637213e-06,
                      "output_mean_power": 8.74647086963423e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.426890844734256e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.627718067426583e-06,
                          "output_power": 2.0064055286201197e-06,
                          "reference_active": true,
                          "delta_db": 0.9083953173089334
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 9.572772525291917e-07,
                          "output_power": 9.748609134172655e-07,
                          "reference_active": true,
                          "delta_db": 0.07904918873581453
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.121992613475366e-08,
                          "output_power": 6.236063748828567e-08,
                          "reference_active": true,
                          "delta_db": 1.7980333696402748
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.928492625228341
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_arctic",
      "method": "golf",
      "status": "measured",
      "started_at": "2026-10-12T11:38:40.373827+00:00",
      "native_sample_rate": 24000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.0027964249999854474,
      "native_input_samples": 80400,
      "native_input_sha256": "417fa558968c888366980ad216aed458790c0513178edaec37a4398c78785cb7",
      "output_sha256": "23302787c8383c70abd348661fecdc011988111a002f9c58ba395b41b66f0196",
      "output_samples": 80160,
      "length_delta_samples": -240,
      "analysis_s": 0.3247736350000139,
      "synthesis_s": 0.08763207499998771,
      "total_s": 0.4124057100000016,
      "rtf": 0.12310618208955271,
      "output_peak": 0.2396697998046875,
      "output_rms": 0.040866796013439975,
      "finished_at": "2026-10-12T11:38:40.792603+00:00",
      "analysis_stages": {
        "external_f0_s": 0.03458668700000089
      },
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "arctic",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 147734,
        "output_samples": 147294,
        "length_delta_samples": -440,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -27.768987734063487,
        "output_peak": 0.23975399371243228
      },
      "common_prefix": {
        "samples": 147294,
        "sample_rate": 44100,
        "reference_coverage": 0.9970216740899184,
        "output_coverage": 1.0,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 147294,
          "output_samples": 147294,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 9.159911687216843,
              "spectral_convergence": 0.5005368528843914,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 7.040018759632789e-07
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 7.689353310146066,
              "spectral_convergence": 0.17637127246801787,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.852450084524139e-07
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 7.018250914508004,
              "spectral_convergence": 0.1848243479893894,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.766215533669349e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 0.00021254068604361208,
                "output_power": 0.00023383542688258245,
                "delta_db": 0.41468230868487616,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.0007529226819213252,
                "output_power": 0.0008165428356597538,
                "delta_db": 0.352285921330567,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.0005464306766619265,
                "output_power": 0.0005037742313028526,
                "delta_db": -0.3529912367361127,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 9.459637084317756e-05,
                "output_power": 8.232223265323987e-05,
                "delta_db": -0.603573347924734,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 1.3613614237810666e-05,
                "output_power": 1.2108636385144988e-05,
                "delta_db": -0.5087820226095023,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.7394901154332135e-05,
                "output_power": 1.4594513279744567e-05,
                "delta_db": -0.7623234918784781,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 1.0853658630064307e-05,
                "output_power": 8.302181409731581e-06,
                "delta_db": -1.1638393915129703,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 6.33043423473613e-11,
                "output_power": 1.0468913239117067e-10,
                "delta_db": 2.184680992407214,
                "reference_active": false
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 7,
            "band_count": 8,
            "active_band_mae_db": 0.5940682458110345,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": 0.06051095215089792,
          "pitch": null,
          "output_rms_dbfs": -27.768987734063487,
          "output_peak": 0.23975399371243228,
          "reference_rms_dbfs": -27.829498686214382,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 7.955838637290303
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 147294,
          "output_samples": 147294,
          "intelligibility": {
            "estoi": {
              "value": 0.9380415312278868,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.040599909517143026,
              "rms_dbfs": -27.829498686214382,
              "peak": 0.232452392578125,
              "peak_dbfs": -12.673319587326667,
              "crest_factor_db": 15.156179098887716,
              "mean": 9.548475195654609e-06,
              "zero_energy": false
            },
            "output": {
              "rms": 0.04088373958266272,
              "rms_dbfs": -27.768987734063487,
              "peak": 0.23975399371243228,
              "peak_dbfs": -12.404682996361762,
              "crest_factor_db": 15.364304737701724,
              "mean": 0.00021267467320097046,
              "zero_energy": false
            },
            "rms_difference_db": 0.06051095215089575
          },
          "full_scale": {
            "reference": {
              "sample_count": 147294,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 147294,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 334,
            "uncovered_tail_samples": 0,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 1.1315037067683916e-06,
            "threshold_dbfs": -59.46344018972689,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 41,
            "quiet_samples": 18081,
            "reference_mean_power": 5.009816483048026e-07,
            "output_mean_power": 1.0980571373646955e-06,
            "output_rms_dbfs": -59.593750607997904,
            "output_zero_energy": false,
            "output_peak": 0.003899288051962267,
            "output_energy_fraction": 8.064196515701062e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 327,
            "reference_estimated_voiced_frames": 184,
            "output_estimated_voiced_frames": 190,
            "both_estimated_voiced_frames": 180,
            "reference_voiced_output_unvoiced_frames": 4,
            "reference_unvoiced_output_voiced_frames": 10,
            "paired_voiced_fraction": 0.9782608695652174,
            "absolute_difference_cents_median": 2.1644002374641462,
            "absolute_difference_cents_p95": 12.258855336781572,
            "difference_over_600_cents_frames": 0,
            "uncovered_tail_samples": 0,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 147294,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147181,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 668,
                      "rms_floor": 0.00010258359777065965,
                      "log_rms_mae_db": 1.3552554314242065,
                      "rms_envelope_nrmse": 0.0934359564226052,
                      "envelope_correlation": 0.9903464555373455,
                      "rms_gain_error_db": 0.060968883143503974,
                      "reference_mean_power": 0.001651982359010943,
                      "output_mean_power": 0.0016753374381897476
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 668,
                      "frequency_resolution_hz": 0.300081654872074,
                      "power_floor": 5.744018305296206e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.0004196903500213713,
                          "output_power": 0.00042748535243247596,
                          "reference_active": true,
                          "delta_db": 0.07992254886736187
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.601058049494274e-05,
                          "output_power": 8.656284088075075e-05,
                          "reference_active": true,
                          "delta_db": 0.027796222032913587
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.249746476833883e-06,
                          "output_power": 6.902794910616853e-06,
                          "reference_active": true,
                          "delta_db": 0.4316256983589392
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.17978148975307154
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147184,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 662,
                      "rms_floor": 9.51361232071073e-05,
                      "log_rms_mae_db": 0.9089241171567788,
                      "rms_envelope_nrmse": 0.0685572026452342,
                      "envelope_correlation": 0.9945168823046757,
                      "rms_gain_error_db": 0.06078473824553953,
                      "reference_mean_power": 0.0016669300800600478,
                      "output_mean_power": 0.0016904248071580347
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 662,
                      "frequency_resolution_hz": 0.3028014281790717,
                      "power_floor": 4.999390164704366e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00039804656700590164,
                          "output_power": 0.0004060405291760442,
                          "reference_active": true,
                          "delta_db": 0.08635502448611154
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.070759102696087e-05,
                          "output_power": 4.158464882530161e-05,
                          "reference_active": true,
                          "delta_db": 0.09257635615313722
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.106058178354036e-07,
                          "output_power": 8.203514427921725e-07,
                          "reference_active": true,
                          "delta_db": 0.6237118744461081
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.26754775169511896
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147181,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 668,
                      "rms_floor": 7.858062960145444e-05,
                      "log_rms_mae_db": 1.4211157605000933,
                      "rms_envelope_nrmse": 0.10876201746247635,
                      "envelope_correlation": 0.9894525179447986,
                      "rms_gain_error_db": -0.08091746109999721,
                      "reference_mean_power": 0.0008504056101716861,
                      "output_mean_power": 0.0008347076052180391
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 668,
                      "frequency_resolution_hz": 0.300081654872074,
                      "power_floor": 4.1141825082020993e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.0003260428461345841,
                          "output_power": 0.0003087999002176395,
                          "reference_active": true,
                          "delta_db": -0.23597524250203689
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.068107845009785e-05,
                          "output_power": 4.6462884305364595e-05,
                          "reference_active": true,
                          "delta_db": -0.3773968145597462
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.819510284248267e-06,
                          "output_power": 3.204682153242527e-06,
                          "reference_active": true,
                          "delta_db": -1.7721794954048864
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7951838508222231
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147184,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 662,
                      "rms_floor": 7.126618944542506e-05,
                      "log_rms_mae_db": 1.0846863834068012,
                      "rms_envelope_nrmse": 0.08239068392758434,
                      "envelope_correlation": 0.9935157362846413,
                      "rms_gain_error_db": -0.08103249104646745,
                      "reference_mean_power": 0.0008581200808697384,
                      "output_mean_power": 0.0008422573623155044
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 662,
                      "frequency_resolution_hz": 0.3028014281790717,
                      "power_floor": 3.699203202928049e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00031666980773638207,
                          "output_power": 0.0002991689044288494,
                          "reference_active": true,
                          "delta_db": -0.24690207352780033
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.4374188154129783e-05,
                          "output_power": 2.247325963213933e-05,
                          "reference_active": true,
                          "delta_db": -0.35264090105534157
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.036188338185189e-07,
                          "output_power": 3.7373490138485486e-07,
                          "reference_active": true,
                          "delta_db": -1.295383065280923
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.6316420132880216
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147181,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 668,
                      "rms_floor": 5.240498613206585e-05,
                      "log_rms_mae_db": 1.4173617887941206,
                      "rms_envelope_nrmse": 0.17129400594307392,
                      "envelope_correlation": 0.9800263652209414,
                      "rms_gain_error_db": -0.7054028758495947,
                      "reference_mean_power": 9.371658640718695e-05,
                      "output_mean_power": 7.966657996962923e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 668,
                      "frequency_resolution_hz": 0.300081654872074,
                      "power_floor": 2.7393505704813312e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.89533166257441e-05,
                          "output_power": 1.5297252320203742e-05,
                          "reference_active": true,
                          "delta_db": -0.9307178759149997
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.260639278022334e-06,
                          "output_power": 1.5706332773044302e-06,
                          "reference_active": true,
                          "delta_db": -1.5815647439147484
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.824691212322689e-07,
                          "output_power": 3.94846797004228e-07,
                          "reference_active": true,
                          "delta_db": -2.970385885205982
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.8275561683452437
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147184,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 662,
                      "rms_floor": 4.4293590932304836e-05,
                      "log_rms_mae_db": 1.0900688133920449,
                      "rms_envelope_nrmse": 0.12081250788326622,
                      "envelope_correlation": 0.9922115654074547,
                      "rms_gain_error_db": -0.7063238500159658,
                      "reference_mean_power": 9.457554664511423e-05,
                      "output_mean_power": 8.037971744157561e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 662,
                      "frequency_resolution_hz": 0.3028014281790717,
                      "power_floor": 2.4951355350330786e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8474912368370368e-05,
                          "output_power": 1.4678431272408311e-05,
                          "reference_active": true,
                          "delta_db": -0.999027433750796
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.0678094506400485e-06,
                          "output_power": 6.869468542385521e-07,
                          "reference_active": true,
                          "delta_db": -1.9157062116721093
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.579146989910142e-08,
                          "output_power": 2.4027024524473548e-08,
                          "reference_active": true,
                          "delta_db": -2.800845928682091
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.9051931913683322
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147181,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 668,
                      "rms_floor": 1.2860287854104408e-05,
                      "log_rms_mae_db": 1.7373681381868147,
                      "rms_envelope_nrmse": 0.23718097374654293,
                      "envelope_correlation": 0.943221917736212,
                      "rms_gain_error_db": -0.484583230833845,
                      "reference_mean_power": 9.45292576588201e-06,
                      "output_mean_power": 8.454889271756832e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 668,
                      "frequency_resolution_hz": 0.300081654872074,
                      "power_floor": 5.62140142230449e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8902552847161388e-06,
                          "output_power": 1.8497442082842434e-06,
                          "reference_active": true,
                          "delta_db": -0.0940878474207457
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.674194811457055e-06,
                          "output_power": 1.712739763154092e-06,
                          "reference_active": true,
                          "delta_db": 0.09885388810732318
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.277062024833127e-07,
                          "output_power": 6.413237302555556e-07,
                          "reference_active": true,
                          "delta_db": 0.8468511121539274
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.34659761589399873
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147184,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 662,
                      "rms_floor": 8.463536202804461e-06,
                      "log_rms_mae_db": 1.2643269955130354,
                      "rms_envelope_nrmse": 0.15990354765794068,
                      "envelope_correlation": 0.9710071583866848,
                      "rms_gain_error_db": -0.483959958404456,
                      "reference_mean_power": 9.537563555554211e-06,
                      "output_mean_power": 8.531815379333583e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 662,
                      "frequency_resolution_hz": 0.3028014281790717,
                      "power_floor": 4.421546407190113e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.637428033868421e-06,
                          "output_power": 1.77684761207621e-06,
                          "reference_active": true,
                          "delta_db": 0.35487961439426036
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 9.534065142525167e-07,
                          "output_power": 9.773431302834773e-07,
                          "reference_active": true,
                          "delta_db": 0.10768949702051159
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.077861959432027e-08,
                          "output_power": 5.3840323013414975e-08,
                          "reference_active": true,
                          "delta_db": 1.2067513646399899
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5564401586849206
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_arctic",
      "method": "sifigan",
      "status": "measured",
      "started_at": "2026-10-12T11:38:45.708552+00:00",
      "native_sample_rate": 24000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.0028099609999969744,
      "native_input_samples": 80400,
      "native_input_sha256": "417fa558968c888366980ad216aed458790c0513178edaec37a4398c78785cb7",
      "output_sha256": "1ea386b8d46006cbaa66a730a0c30013b3bc77ea30883c6ce49d613fdd0bda13",
      "output_samples": 80520,
      "length_delta_samples": 120,
      "analysis_s": 1.131860773999989,
      "synthesis_s": 2.2207960080000078,
      "total_s": 3.352656781999997,
      "rtf": 1.0007930692537303,
      "output_peak": 0.24281077086925507,
      "output_rms": 0.040298774052798955,
      "finished_at": "2026-10-12T11:38:49.067851+00:00",
      "analysis_stages": {
        "external_f0_s": 0.6790806509999925
      },
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "arctic",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 147734,
        "output_samples": 147956,
        "length_delta_samples": 222,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -27.89049032817796,
        "output_peak": 0.24464781020910228
      },
      "common_prefix": {
        "samples": 147734,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 0.9984995539214361,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 147734,
          "output_samples": 147734,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 9.474964743428924,
              "spectral_convergence": 0.4536666375339895,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 7.040018759632789e-07
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 7.933236046951942,
              "spectral_convergence": 0.23733315914970526,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.852450084524139e-07
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 7.020604253818325,
              "spectral_convergence": 0.21881747303049257,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.766215533669349e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 0.00021192180604849,
                "output_power": 0.000235665036919388,
                "delta_db": 0.4611950924685249,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.0007506719378509437,
                "output_power": 0.0007443232538788605,
                "delta_db": -0.036885937909798494,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.0005447977950658482,
                "output_power": 0.0005164088112686194,
                "delta_db": -0.23241697365331804,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 9.431507243494021e-05,
                "output_power": 9.455179728915083e-05,
                "delta_db": 0.0108868592888251,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 1.3572749160000328e-05,
                "output_power": 1.3152488632580285e-05,
                "delta_db": -0.13659887788336542,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.7343096646047786e-05,
                "output_power": 1.4323380321181681e-05,
                "delta_db": -0.8308112057903116,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 1.0821264092698321e-05,
                "output_power": 9.322245738960957e-06,
                "delta_db": -0.6475744915149231,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 6.343414671079065e-11,
                "output_power": 1.6795101095437612e-10,
                "delta_db": 4.228595200551634,
                "reference_active": false
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 7,
            "band_count": 8,
            "active_band_mae_db": 0.33662420550129524,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": -0.04167901135604792,
          "pitch": null,
          "output_rms_dbfs": -27.884130478428485,
          "output_peak": 0.24464781020910228,
          "reference_rms_dbfs": -27.842451467072436,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 8.142935014733064
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 147734,
          "output_samples": 147734,
          "intelligibility": {
            "estoi": {
              "value": 0.9191371194354236,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04053941026622261,
              "rms_dbfs": -27.842451467072436,
              "peak": 0.232452392578125,
              "peak_dbfs": -12.673319587326667,
              "crest_factor_db": 15.169131879745771,
              "mean": 1.0310790979268482e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.040345348951270114,
              "rms_dbfs": -27.884130478428485,
              "peak": 0.24464781020910228,
              "peak_dbfs": -12.229173343297255,
              "crest_factor_db": 15.65495713513123,
              "mean": -2.7765848654740428e-05,
              "zero_energy": false
            },
            "rms_difference_db": -0.04167901135604879
          },
          "full_scale": {
            "reference": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 334,
            "uncovered_tail_samples": 440,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 1.1315037067683916e-06,
            "threshold_dbfs": -59.46344018972689,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 41,
            "quiet_samples": 18081,
            "reference_mean_power": 5.009816483048026e-07,
            "output_mean_power": 7.167341547337389e-06,
            "output_rms_dbfs": -51.44641899480561,
            "output_zero_energy": false,
            "output_peak": 0.04563410804817108,
            "output_energy_fraction": 0.0005389061474506982,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 327,
            "reference_estimated_voiced_frames": 184,
            "output_estimated_voiced_frames": 189,
            "both_estimated_voiced_frames": 179,
            "reference_voiced_output_unvoiced_frames": 5,
            "reference_unvoiced_output_voiced_frames": 10,
            "paired_voiced_fraction": 0.9728260869565217,
            "absolute_difference_cents_median": 3.2913197555522142,
            "absolute_difference_cents_p95": 26.743293955463443,
            "difference_over_600_cents_frames": 0,
            "uncovered_tail_samples": 440,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 147734,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 0.00010258359777065965,
                      "log_rms_mae_db": 2.672786592277021,
                      "rms_envelope_nrmse": 0.12893735354239996,
                      "envelope_correlation": 0.9803926251691051,
                      "rms_gain_error_db": -0.04165391147902823,
                      "reference_mean_power": 0.0016470512022936088,
                      "output_mean_power": 0.0016313295731361633
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.737071418318221e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.0004181860744866353,
                          "output_power": 0.0004047930724863239,
                          "reference_active": true,
                          "delta_db": -0.1413649529937769
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.655525977123722e-05,
                          "output_power": 8.172144509421687e-05,
                          "reference_active": true,
                          "delta_db": -0.2495742639459232
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.295514528933935e-06,
                          "output_power": 5.08793458700054e-06,
                          "reference_active": true,
                          "delta_db": -0.9248971137123443
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.43861211021734814
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 9.51361232071073e-05,
                      "log_rms_mae_db": 1.7816878973290904,
                      "rms_envelope_nrmse": 0.0791049866788256,
                      "envelope_correlation": 0.9918043220831484,
                      "rms_gain_error_db": -0.042034341985996114,
                      "reference_mean_power": 0.0016619096793530144,
                      "output_mean_power": 0.001645902038301328
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.993659628933029e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00039685314515544806,
                          "output_power": 0.00038498798788967506,
                          "reference_active": true,
                          "delta_db": -0.13182647357402866
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.1281752657875335e-05,
                          "output_power": 3.939261094737641e-05,
                          "reference_active": true,
                          "delta_db": -0.20343360556970666
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.147467230270734e-07,
                          "output_power": 6.097909637104397e-07,
                          "reference_active": true,
                          "delta_db": -0.6897118846843993
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.3416573212760448
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 7.858062960145444e-05,
                      "log_rms_mae_db": 2.318528455803697,
                      "rms_envelope_nrmse": 0.15557368416061004,
                      "envelope_correlation": 0.9780479642560024,
                      "rms_gain_error_db": -0.1543874227363376,
                      "reference_mean_power": 0.0008478670875848599,
                      "output_mean_power": 0.0008182556917453044
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 4.118572979449278e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00032611065091443195,
                          "output_power": 0.00031905818757953567,
                          "reference_active": true,
                          "delta_db": -0.09495089440960156
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.0919247362893804e-05,
                          "output_power": 5.025838646652249e-05,
                          "reference_active": true,
                          "delta_db": -0.056734341648921285
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.838294986205235e-06,
                          "output_power": 4.411877385010171e-06,
                          "reference_active": true,
                          "delta_db": -0.40068909336714265
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.18412477647522185
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 7.126618944542506e-05,
                      "log_rms_mae_db": 1.6898319602950076,
                      "rms_envelope_nrmse": 0.09879015047101962,
                      "envelope_correlation": 0.9904317862983238,
                      "rms_gain_error_db": -0.1543169881224526,
                      "reference_mean_power": 0.0008555353845487757,
                      "output_mean_power": 0.0008256695674536792
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 3.7040306770976235e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00031681362834561343,
                          "output_power": 0.0003096119850291023,
                          "reference_active": true,
                          "delta_db": -0.09986091524718907
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.473236521731313e-05,
                          "output_power": 2.4415410307763185e-05,
                          "reference_active": true,
                          "delta_db": -0.05601623882733475
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.047139621751533e-07,
                          "output_power": 4.725268747179707e-07,
                          "reference_active": true,
                          "delta_db": -0.2861880520445822
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.1473550687063687
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 5.240498613206585e-05,
                      "log_rms_mae_db": 2.13869459491216,
                      "rms_envelope_nrmse": 0.17970896292060537,
                      "envelope_correlation": 0.9744971552666083,
                      "rms_gain_error_db": -0.03180329677234052,
                      "reference_mean_power": 9.343683802972147e-05,
                      "output_mean_power": 9.27551011801981e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 2.735946251376838e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8948025909514686e-05,
                          "output_power": 1.902151140499203e-05,
                          "reference_active": true,
                          "delta_db": 0.016810520901864984
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.263241347636967e-06,
                          "output_power": 2.035795539810554e-06,
                          "reference_active": true,
                          "delta_db": -0.45996710198069346
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.834787408003976e-07,
                          "output_power": 7.142226937962946e-07,
                          "reference_active": true,
                          "delta_db": -0.4019357125847557
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.29290444515577135
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 4.4293590932304836e-05,
                      "log_rms_mae_db": 1.3736240604088954,
                      "rms_envelope_nrmse": 0.10524577594185303,
                      "envelope_correlation": 0.9906668804927011,
                      "rms_gain_error_db": -0.03237628548883259,
                      "reference_mean_power": 9.429068361359236e-05,
                      "output_mean_power": 9.359036819936414e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 2.4918519689831095e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8481977963071213e-05,
                          "output_power": 1.8470283208512757e-05,
                          "reference_active": true,
                          "delta_db": -0.00274893464188227
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.0691683146494508e-06,
                          "output_power": 9.638213493898548e-07,
                          "reference_active": true,
                          "delta_db": -0.45049537704735076
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.6063349891805814e-08,
                          "output_power": 3.9753106944116e-08,
                          "reference_active": true,
                          "delta_db": -0.639844411854274
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.36436290784783565
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 1.2860287854104408e-05,
                      "log_rms_mae_db": 2.488510946019934,
                      "rms_envelope_nrmse": 0.33376298157530077,
                      "envelope_correlation": 0.8796993054768368,
                      "rms_gain_error_db": -0.369790679215834,
                      "reference_mean_power": 9.424710941015776e-06,
                      "output_mean_power": 8.65543643330501e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.625314523028166e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8781674075282524e-06,
                          "output_power": 1.730915392564304e-06,
                          "reference_active": true,
                          "delta_db": -0.35458459811152165
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.6792080290845895e-06,
                          "output_power": 1.5845056288798145e-06,
                          "reference_active": true,
                          "delta_db": -0.2521071580654643
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.330459106871231e-07,
                          "output_power": 4.789054626319997e-07,
                          "reference_active": true,
                          "delta_db": -0.465148251192844
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.35728000245661
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 8.463536202804461e-06,
                      "log_rms_mae_db": 1.4787452570483008,
                      "rms_envelope_nrmse": 0.1649095184478934,
                      "envelope_correlation": 0.9652585279502507,
                      "rms_gain_error_db": -0.36874919109648296,
                      "reference_mean_power": 9.508840352637213e-06,
                      "output_mean_power": 8.734793390894203e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.426890844734256e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.627718067426583e-06,
                          "output_power": 1.5109731862135967e-06,
                          "reference_active": true,
                          "delta_db": -0.3232242671189303
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 9.572772525291917e-07,
                          "output_power": 9.035795826653208e-07,
                          "reference_active": true,
                          "delta_db": -0.25071330019148047
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.121992613475366e-08,
                          "output_power": 3.639788361458399e-08,
                          "reference_active": true,
                          "delta_db": -0.540310772303182
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.3714161132045309
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_arctic",
      "method": "llsm2",
      "status": "measured",
      "started_at": "2026-10-12T11:38:56.174276+00:00",
      "native_sample_rate": 24000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.00271461299999487,
      "native_input_samples": 80400,
      "native_input_sha256": "417fa558968c888366980ad216aed458790c0513178edaec37a4398c78785cb7",
      "output_sha256": "3d5c5158beda00e630d2446ea5f833d379c52b5ce658a23c3912a3cd6b6b7bd0",
      "output_samples": 80640,
      "length_delta_samples": 240,
      "analysis_s": 0.19737292499999626,
      "synthesis_s": 0.025730019000008042,
      "total_s": 0.2231029440000043,
      "rtf": 0.06659789373134456,
      "output_peak": 0.22467108070850372,
      "output_rms": 0.04000703938217067,
      "finished_at": "2026-10-12T11:38:56.404641+00:00",
      "analysis_stages": {
        "external_f0_s": 0.04687874600000441
      },
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "arctic",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 147734,
        "output_samples": 148176,
        "length_delta_samples": 442,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -27.953498925960993,
        "output_peak": 0.22720066730670443
      },
      "common_prefix": {
        "samples": 147734,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 0.997017060792571,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 147734,
          "output_samples": 147734,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 8.127784067357496,
              "spectral_convergence": 0.10980461963019779,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 7.040018759632789e-07
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 6.8408322543415965,
              "spectral_convergence": 0.10654292245554338,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.852450084524139e-07
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 6.216771298720576,
              "spectral_convergence": 0.11731183576151775,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.766215533669349e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 0.00021192180604849,
                "output_power": 0.00020996730288554683,
                "delta_db": -0.04023976885129639,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.0007506719378509437,
                "output_power": 0.0007392576750312449,
                "delta_db": -0.06654338965511367,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.0005447977950658482,
                "output_power": 0.0005274510381059375,
                "delta_db": -0.14053189783877856,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 9.431507243494021e-05,
                "output_power": 9.300695780227815e-05,
                "delta_db": -0.060656635091135794,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 1.3572749160000328e-05,
                "output_power": 1.254013825087229e-05,
                "delta_db": -0.3436549843213257,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.7343096646047786e-05,
                "output_power": 1.6000857088464037e-05,
                "delta_db": -0.3498339791639471,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 1.0821264092698321e-05,
                "output_power": 8.522817842723712e-06,
                "delta_db": -1.0369478970191253,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 6.343414671079065e-11,
                "output_power": 1.0156569008036088e-10,
                "delta_db": 2.044239213470846,
                "reference_active": false
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 7,
            "band_count": 8,
            "active_band_mae_db": 0.29120122170581747,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": -0.09807385710468443,
          "pitch": null,
          "output_rms_dbfs": -27.94052532417712,
          "output_peak": 0.22720066730670443,
          "reference_rms_dbfs": -27.842451467072436,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 7.061795873473223
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 147734,
          "output_samples": 147734,
          "intelligibility": {
            "estoi": {
              "value": 0.9769844115069563,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04053941026622261,
              "rms_dbfs": -27.842451467072436,
              "peak": 0.232452392578125,
              "peak_dbfs": -12.673319587326667,
              "crest_factor_db": 15.169131879745771,
              "mean": 1.0310790979268482e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.040084247386882016,
              "rms_dbfs": -27.94052532417712,
              "peak": 0.22720066730670443,
              "peak_dbfs": -12.87180794802373,
              "crest_factor_db": 15.068717376153389,
              "mean": -2.9813475780708075e-05,
              "zero_energy": false
            },
            "rms_difference_db": -0.09807385710468353
          },
          "full_scale": {
            "reference": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 334,
            "uncovered_tail_samples": 440,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 1.1315037067683916e-06,
            "threshold_dbfs": -59.46344018972689,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 41,
            "quiet_samples": 18081,
            "reference_mean_power": 5.009816483048026e-07,
            "output_mean_power": 9.482704283044246e-07,
            "output_rms_dbfs": -60.23067792649086,
            "output_zero_energy": false,
            "output_peak": 0.027123104807009985,
            "output_energy_fraction": 7.223151586511838e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 327,
            "reference_estimated_voiced_frames": 184,
            "output_estimated_voiced_frames": 188,
            "both_estimated_voiced_frames": 183,
            "reference_voiced_output_unvoiced_frames": 1,
            "reference_unvoiced_output_voiced_frames": 5,
            "paired_voiced_fraction": 0.9945652173913043,
            "absolute_difference_cents_median": 0.510577149651648,
            "absolute_difference_cents_p95": 4.619487412913877,
            "difference_over_600_cents_frames": 0,
            "uncovered_tail_samples": 440,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 147734,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 0.00010258359777065965,
                      "log_rms_mae_db": 0.7682459613355594,
                      "rms_envelope_nrmse": 0.03511969942456301,
                      "envelope_correlation": 0.9987040251727023,
                      "rms_gain_error_db": -0.09803919605077482,
                      "reference_mean_power": 0.0016470512022936088,
                      "output_mean_power": 0.0016102866067912837
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.737071418318221e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.0004181860744866353,
                          "output_power": 0.00041502558548347075,
                          "reference_active": true,
                          "delta_db": -0.032946958931554166
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.655525977123722e-05,
                          "output_power": 8.20078548144775e-05,
                          "reference_active": true,
                          "delta_db": -0.23438012483746734
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.295514528933935e-06,
                          "output_power": 4.515606674585775e-06,
                          "reference_active": true,
                          "delta_db": -1.4431512423803685
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5701594420497966
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 9.51361232071073e-05,
                      "log_rms_mae_db": 0.4633613909064657,
                      "rms_envelope_nrmse": 0.023365931590560492,
                      "envelope_correlation": 0.9994893998809793,
                      "rms_gain_error_db": -0.09803317636122534,
                      "reference_mean_power": 0.0016619096793530144,
                      "output_mean_power": 0.0016248156730080782
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.993659628933029e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00039685314515544806,
                          "output_power": 0.0003940036485076497,
                          "reference_active": true,
                          "delta_db": -0.031295830938050176
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.1281752657875335e-05,
                          "output_power": 3.9540709001597345e-05,
                          "reference_active": true,
                          "delta_db": -0.18713675064372987
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.147467230270734e-07,
                          "output_power": 5.111429088267973e-07,
                          "reference_active": true,
                          "delta_db": -1.45609832749391
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5581769696918967
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 7.858062960145444e-05,
                      "log_rms_mae_db": 0.9114966712530835,
                      "rms_envelope_nrmse": 0.039971182978169255,
                      "envelope_correlation": 0.9986761766125589,
                      "rms_gain_error_db": -0.10383837368066821,
                      "reference_mean_power": 0.0008478670875848599,
                      "output_mean_power": 0.0008278352970869532
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 4.118572979449278e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00032611065091443195,
                          "output_power": 0.0003181672147175935,
                          "reference_active": true,
                          "delta_db": -0.10709557188884332
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.0919247362893804e-05,
                          "output_power": 4.71170049831767e-05,
                          "reference_active": true,
                          "delta_db": -0.3370429922203998
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.838294986205235e-06,
                          "output_power": 3.943324949628815e-06,
                          "reference_active": true,
                          "delta_db": -0.8882977677350813
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.4441454439481081
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 7.126618944542506e-05,
                      "log_rms_mae_db": 0.4957640388384907,
                      "rms_envelope_nrmse": 0.024724720723331806,
                      "envelope_correlation": 0.9995276714152812,
                      "rms_gain_error_db": -0.10380453941309442,
                      "reference_mean_power": 0.0008555353845487757,
                      "output_mean_power": 0.000835328929830576
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 3.7040306770976235e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00031681362834561343,
                          "output_power": 0.00030854389110311983,
                          "reference_active": true,
                          "delta_db": -0.1148690314117799
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.473236521731313e-05,
                          "output_power": 2.3003823309370318e-05,
                          "reference_active": true,
                          "delta_db": -0.3146562784842335
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.047139621751533e-07,
                          "output_power": 3.816939005225674e-07,
                          "reference_active": true,
                          "delta_db": -1.2133009905427234
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5476087668129123
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 5.240498613206585e-05,
                      "log_rms_mae_db": 0.8600263388295011,
                      "rms_envelope_nrmse": 0.051830699084393625,
                      "envelope_correlation": 0.9979573804115001,
                      "rms_gain_error_db": -0.09624995134915809,
                      "reference_mean_power": 9.343683802972147e-05,
                      "output_mean_power": 9.138883435780046e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 2.735946251376838e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8948025909514686e-05,
                          "output_power": 1.8908520957530813e-05,
                          "reference_active": true,
                          "delta_db": -0.009064106635253014
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.263241347636967e-06,
                          "output_power": 1.9313427744961017e-06,
                          "reference_active": true,
                          "delta_db": -0.6887150958412424
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.834787408003976e-07,
                          "output_power": 6.986691991473079e-07,
                          "reference_active": true,
                          "delta_db": -0.49755618867639995
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.39844513038429846
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 4.4293590932304836e-05,
                      "log_rms_mae_db": 0.5352781680460713,
                      "rms_envelope_nrmse": 0.02820669809265977,
                      "envelope_correlation": 0.9994320453833985,
                      "rms_gain_error_db": -0.09619835776474005,
                      "reference_mean_power": 9.429068361359236e-05,
                      "output_mean_power": 9.222506046551923e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 2.4918519689831095e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8481977963071213e-05,
                          "output_power": 1.8388510504129273e-05,
                          "reference_active": true,
                          "delta_db": -0.022018958954586346
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.0691683146494508e-06,
                          "output_power": 8.89978441594234e-07,
                          "reference_active": true,
                          "delta_db": -0.7966659310221667
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.6063349891805814e-08,
                          "output_power": 3.377021738824289e-08,
                          "reference_active": true,
                          "delta_db": -1.3482166190881246
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7223005030216259
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 1.2860287854104408e-05,
                      "log_rms_mae_db": 1.3908034980945734,
                      "rms_envelope_nrmse": 0.1378050409756749,
                      "envelope_correlation": 0.9808279482358293,
                      "rms_gain_error_db": -0.34437690942565924,
                      "reference_mean_power": 9.424710941015776e-06,
                      "output_mean_power": 8.70623425152279e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.625314523028166e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8781674075282524e-06,
                          "output_power": 1.723081222784983e-06,
                          "reference_active": true,
                          "delta_db": -0.37428550068205235
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.6792080290845895e-06,
                          "output_power": 1.4863397782546557e-06,
                          "reference_active": true,
                          "delta_db": -0.5298640136647703
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.330459106871231e-07,
                          "output_power": 3.844114093288406e-07,
                          "reference_active": true,
                          "delta_db": -1.4196834692128775
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7746109945199001
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 8.463536202804461e-06,
                      "log_rms_mae_db": 0.8744048587829556,
                      "rms_envelope_nrmse": 0.08021428773788165,
                      "envelope_correlation": 0.9933053218171964,
                      "rms_gain_error_db": -0.34475731620791217,
                      "reference_mean_power": 9.508840352637213e-06,
                      "output_mean_power": 8.783180832424002e-06
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.426890844734256e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.627718067426583e-06,
                          "output_power": 1.4741042978153031e-06,
                          "reference_active": true,
                          "delta_db": -0.4305097171424309
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 9.572772525291917e-07,
                          "output_power": 8.342464669401056e-07,
                          "reference_active": true,
                          "delta_db": -0.5974336295598861
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.121992613475366e-08,
                          "output_power": 2.90847389769523e-08,
                          "reference_active": true,
                          "delta_db": -1.5144203865193249
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.8474545777405473
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_arctic",
      "method": "magphase",
      "status": "measured",
      "started_at": "2026-10-12T11:38:58.507574+00:00",
      "native_sample_rate": 48000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "PCM16 (MagPhase/REAPER file API only)",
      "resampling_and_input_io_s": 0.006063879999999244,
      "native_input_samples": 160799,
      "native_input_sha256": "5a4b6cd8a0e7c547f3626e0d21d3ab3af048cb9b98af9c1072afddde274f4c07",
      "output_sha256": "dfea06ca1566bd39001ca7dafcca038a2321299c155321d61a3a7bdee7ea3250",
      "output_samples": 161032,
      "length_delta_samples": 233,
      "analysis_s": 2.1354286159999845,
      "synthesis_s": 0.6914883379999992,
      "total_s": 2.8269169539999837,
      "rtf": 0.8438610550563077,
      "output_peak": 0.2457974213917105,
      "output_rms": 0.03914238959361408,
      "finished_at": "2026-10-12T11:39:01.345725+00:00",
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "arctic",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 147734,
        "output_samples": 147949,
        "length_delta_samples": 215,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -28.145121544193437,
        "output_peak": 0.24486297016808867
      },
      "common_prefix": {
        "samples": 147734,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 0.9985467965312371,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 147734,
          "output_samples": 147734,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 5.478088113207686,
              "spectral_convergence": 0.26697415746929654,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 7.040018759632789e-07
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 4.777519613825908,
              "spectral_convergence": 0.13270323952396074,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.852450084524139e-07
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 4.234615996427011,
              "spectral_convergence": 0.1491123770165633,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 5.766215533669349e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 0.00021192180604849,
                "output_power": 0.00019329007265768297,
                "delta_db": -0.39966097149535607,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.0007506719378509437,
                "output_power": 0.0006898921133102242,
                "delta_db": -0.3666900084729144,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.0005447977950658482,
                "output_power": 0.0005142172288233798,
                "delta_db": -0.2508871764270401,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 9.431507243494021e-05,
                "output_power": 9.41310214153404e-05,
                "delta_db": -0.0084833138202551,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 1.3572749160000328e-05,
                "output_power": 1.4754384922865437e-05,
                "delta_db": 0.36253286546737007,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.7343096646047786e-05,
                "output_power": 1.728228637119123e-05,
                "delta_db": -0.015254470370657892,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 1.0821264092698321e-05,
                "output_power": 1.1471770599496325e-05,
                "delta_db": 0.2535245772838156,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 6.343414671079065e-11,
                "output_power": 1.1427572146576899e-10,
                "delta_db": 2.556308695381446,
                "reference_active": false
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 7,
            "band_count": 8,
            "active_band_mae_db": 0.23671905476248703,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": -0.2963544158469022,
          "pitch": null,
          "output_rms_dbfs": -28.13880588291934,
          "output_peak": 0.24486297016808867,
          "reference_rms_dbfs": -27.842451467072436,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 4.8300745744868685
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 147734,
          "output_samples": 147734,
          "intelligibility": {
            "estoi": {
              "value": 0.9704880841202749,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04053941026622261,
              "rms_dbfs": -27.842451467072436,
              "peak": 0.232452392578125,
              "peak_dbfs": -12.673319587326667,
              "crest_factor_db": 15.169131879745771,
              "mean": 1.0310790979268482e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.03917957366250814,
              "rms_dbfs": -28.13880588291934,
              "peak": 0.24486297016808867,
              "peak_dbfs": -12.221537737081603,
              "crest_factor_db": 15.917268145837738,
              "mean": -0.000790661080269352,
              "zero_energy": false
            },
            "rms_difference_db": -0.2963544158469027
          },
          "full_scale": {
            "reference": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 147734,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 334,
            "uncovered_tail_samples": 440,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 1.1315037067683916e-06,
            "threshold_dbfs": -59.46344018972689,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 41,
            "quiet_samples": 18081,
            "reference_mean_power": 5.009816483048026e-07,
            "output_mean_power": 1.2108637878617448e-06,
            "output_rms_dbfs": -59.16904708638267,
            "output_zero_energy": false,
            "output_peak": 0.025509073797980807,
            "output_energy_fraction": 9.654235066180397e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 327,
            "reference_estimated_voiced_frames": 184,
            "output_estimated_voiced_frames": 185,
            "both_estimated_voiced_frames": 181,
            "reference_voiced_output_unvoiced_frames": 3,
            "reference_unvoiced_output_voiced_frames": 4,
            "paired_voiced_fraction": 0.9836956521739131,
            "absolute_difference_cents_median": 3.5804848896916024,
            "absolute_difference_cents_p95": 9.913063115432701,
            "difference_over_600_cents_frames": 0,
            "uncovered_tail_samples": 440,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 147734,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 0.00010258359777065965,
                      "log_rms_mae_db": 1.043326358133119,
                      "rms_envelope_nrmse": 0.06241242925418057,
                      "envelope_correlation": 0.9966575888870117,
                      "rms_gain_error_db": -0.29669344229773287,
                      "reference_mean_power": 0.0016470512022936088,
                      "output_mean_power": 0.0015382883683538143
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.737071418318221e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.0004181860744866353,
                          "output_power": 0.0003879587680859513,
                          "reference_active": true,
                          "delta_db": -0.3258399518818244
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.655525977123722e-05,
                          "output_power": 8.480661227670587e-05,
                          "reference_active": true,
                          "delta_db": -0.0886374909425125
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.295514528933935e-06,
                          "output_power": 6.702157632430928e-06,
                          "reference_active": true,
                          "delta_db": 0.27183407532305387
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.22877050604913027
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 9.51361232071073e-05,
                      "log_rms_mae_db": 0.6349598487309452,
                      "rms_envelope_nrmse": 0.044475691158611386,
                      "envelope_correlation": 0.9988393893187798,
                      "rms_gain_error_db": -0.29679406747658155,
                      "reference_mean_power": 0.0016619096793530144,
                      "output_mean_power": 0.0015521297046573533
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.993659628933029e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00039685314515544806,
                          "output_power": 0.00036734013312484905,
                          "reference_active": true,
                          "delta_db": -0.3356144749585104
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.1281752657875335e-05,
                          "output_power": 4.026875164032201e-05,
                          "reference_active": true,
                          "delta_db": -0.1078996103370129
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.147467230270734e-07,
                          "output_power": 7.708202259962338e-07,
                          "reference_active": true,
                          "delta_db": 0.3280092895623476
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.257174458285957
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 7.858062960145444e-05,
                      "log_rms_mae_db": 1.1602554107262046,
                      "rms_envelope_nrmse": 0.05539883683305805,
                      "envelope_correlation": 0.9978745598470535,
                      "rms_gain_error_db": -0.24615837820939407,
                      "reference_mean_power": 0.0008478670875848599,
                      "output_mean_power": 0.0008011464993802617
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 4.118572979449278e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00032611065091443195,
                          "output_power": 0.0003080256998473851,
                          "reference_active": true,
                          "delta_db": -0.24778030300721107
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.0919247362893804e-05,
                          "output_power": 4.7790792733035724e-05,
                          "reference_active": true,
                          "delta_db": -0.275377412548906
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.838294986205235e-06,
                          "output_power": 4.556188898077306e-06,
                          "reference_active": true,
                          "delta_db": -0.2609062186594338
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.26135464473851694
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 7.126618944542506e-05,
                      "log_rms_mae_db": 0.7760495109848857,
                      "rms_envelope_nrmse": 0.04049345179312594,
                      "envelope_correlation": 0.9991208038849616,
                      "rms_gain_error_db": -0.24608312223901624,
                      "reference_mean_power": 0.0008555353845487757,
                      "output_mean_power": 0.0008084062532457414
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 3.7040306770976235e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 0.00031681362834561343,
                          "output_power": 0.0002995383881889515,
                          "reference_active": true,
                          "delta_db": -0.24351366799790958
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.473236521731313e-05,
                          "output_power": 2.3255786585675304e-05,
                          "reference_active": true,
                          "delta_db": -0.26734617668368077
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.047139621751533e-07,
                          "output_power": 4.527405673456566e-07,
                          "reference_active": true,
                          "delta_db": -0.4719590828063131
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.3276063091626345
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 5.240498613206585e-05,
                      "log_rms_mae_db": 0.8160765798902857,
                      "rms_envelope_nrmse": 0.05187534897168628,
                      "envelope_correlation": 0.9978818006256741,
                      "rms_gain_error_db": 0.007490749473056106,
                      "reference_mean_power": 9.343683802972147e-05,
                      "output_mean_power": 9.359813777649565e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 2.735946251376838e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8948025909514686e-05,
                          "output_power": 1.947047483207802e-05,
                          "reference_active": true,
                          "delta_db": 0.11812573003570374
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.263241347636967e-06,
                          "output_power": 2.297536846673904e-06,
                          "reference_active": true,
                          "delta_db": 0.06531616358343023
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 7.834787408003976e-07,
                          "output_power": 8.01561456333905e-07,
                          "reference_active": true,
                          "delta_db": 0.09909609000283183
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.0941793278739886
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 4.4293590932304836e-05,
                      "log_rms_mae_db": 0.5263081324998228,
                      "rms_envelope_nrmse": 0.028166238538091994,
                      "envelope_correlation": 0.999332887411409,
                      "rms_gain_error_db": 0.006720197996628552,
                      "reference_mean_power": 9.429068361359236e-05,
                      "output_mean_power": 9.443670033580618e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 2.4918519689831095e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8481977963071213e-05,
                          "output_power": 1.8971711128099652e-05,
                          "reference_active": true,
                          "delta_db": 0.11358055186076756
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.0691683146494508e-06,
                          "output_power": 1.1006813985203103e-06,
                          "reference_active": true,
                          "delta_db": 0.1261554723924211
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.6063349891805814e-08,
                          "output_power": 4.707227115620348e-08,
                          "reference_active": true,
                          "delta_db": 0.0940963454756491
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.11127745657627924
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147621,
                    "uncovered_tail_samples": 113,
                    "metrics": {
                      "frame_count": 670,
                      "rms_floor": 1.2860287854104408e-05,
                      "log_rms_mae_db": 1.2182475423375119,
                      "rms_envelope_nrmse": 0.18311978869232273,
                      "envelope_correlation": 0.9688602124503056,
                      "rms_gain_error_db": 0.41927542251560923,
                      "reference_mean_power": 9.424710941015776e-06,
                      "output_mean_power": 1.0379957644621934e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 670,
                      "frequency_resolution_hz": 0.29918588873812757,
                      "power_floor": 5.625314523028166e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.8781674075282524e-06,
                          "output_power": 2.551205384807495e-06,
                          "reference_active": true,
                          "delta_db": 1.3301112307955032
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.6792080290845895e-06,
                          "output_power": 1.818589164102975e-06,
                          "reference_active": true,
                          "delta_db": 0.3463009690926466
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.330459106871231e-07,
                          "output_power": 4.7537611583224984e-07,
                          "reference_active": true,
                          "delta_db": -0.49727258143556663
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7245615937745722
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 147624,
                    "uncovered_tail_samples": 110,
                    "metrics": {
                      "frame_count": 664,
                      "rms_floor": 8.463536202804461e-06,
                      "log_rms_mae_db": 0.802682304919757,
                      "rms_envelope_nrmse": 0.13604780505530373,
                      "envelope_correlation": 0.9801219834530436,
                      "rms_gain_error_db": 0.4217603896905896,
                      "reference_mean_power": 9.508840352637213e-06,
                      "output_mean_power": 1.0478608024147724e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 664,
                      "frequency_resolution_hz": 0.30188937568455637,
                      "power_floor": 4.426890844734256e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 1.627718067426583e-06,
                          "output_power": 2.271273478710443e-06,
                          "reference_active": true,
                          "delta_db": 1.4469024563450148
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 9.572772525291917e-07,
                          "output_power": 1.0352539528905286e-06,
                          "reference_active": true,
                          "delta_db": 0.3400915839997088
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.121992613475366e-08,
                          "output_power": 3.103086797739903e-08,
                          "reference_active": true,
                          "delta_db": -1.2331328561042212
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.0067089654829815
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_ready",
      "method": "world_harvest",
      "status": "measured",
      "started_at": "2026-10-12T11:37:49.728583+00:00",
      "native_sample_rate": 16000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.004284565999995493,
      "native_input_samples": 80000,
      "native_input_sha256": "272edc24c29f9bdefd63f8ea08c58dc4488fe68f37f59af32352ec31bc99c761",
      "output_sha256": "60977b2a9c9e5a21e23f7179c66fda795a588c450aacec790683f52bea2b8585",
      "output_samples": 80000,
      "length_delta_samples": 0,
      "analysis_s": 1.3105030210000166,
      "synthesis_s": 0.13900450299999534,
      "total_s": 1.449507524000012,
      "rtf": 0.2899015048000024,
      "output_peak": 0.6228445127861545,
      "output_rms": 0.05988977704123537,
      "finished_at": "2026-10-12T11:37:51.186212+00:00",
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "ready",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "ok",
        "reference_samples": 220500,
        "output_samples": 220500,
        "length_delta_samples": 0,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": [
          {
            "window_ms": 5,
            "window_samples": 220,
            "hop_samples": 55,
            "log_magnitude_mae_db": 13.706921473637687,
            "spectral_convergence": 0.4256170921182679,
            "spectral_convergence_status": "ok",
            "amplitude_floor": 1.3526499269056254e-06
          },
          {
            "window_ms": 20,
            "window_samples": 882,
            "hop_samples": 220,
            "log_magnitude_mae_db": 11.308300642418027,
            "spectral_convergence": 0.366146828033146,
            "spectral_convergence_status": "ok",
            "amplitude_floor": 1.2238389615019609e-06
          },
          {
            "window_ms": 80,
            "window_samples": 3528,
            "hop_samples": 882,
            "log_magnitude_mae_db": 10.376212562188721,
            "spectral_convergence": 0.37582109282693693,
            "spectral_convergence_status": "ok",
            "amplitude_floor": 8.265562353627178e-07
          }
        ],
        "band_energy": {
          "bands": [
            {
              "low_hz": 0.0,
              "high_hz": 200.0,
              "reference_power": 5.873507160680166e-07,
              "output_power": 7.442924532683023e-07,
              "delta_db": 1.0284611293943793,
              "reference_active": true
            },
            {
              "low_hz": 200.0,
              "high_hz": 500.0,
              "reference_power": 0.00012765279053574292,
              "output_power": 0.00016048515418281173,
              "delta_db": 0.9940455073744613,
              "reference_active": true
            },
            {
              "low_hz": 500.0,
              "high_hz": 1000.0,
              "reference_power": 0.00018664008076538165,
              "output_power": 0.00026103908270137024,
              "delta_db": 1.4570062094012726,
              "reference_active": true
            },
            {
              "low_hz": 1000.0,
              "high_hz": 2000.0,
              "reference_power": 0.0014726428134665878,
              "output_power": 0.002634417845834215,
              "delta_db": 2.5258723730430166,
              "reference_active": true
            },
            {
              "low_hz": 2000.0,
              "high_hz": 4000.0,
              "reference_power": 0.0004214136033266785,
              "output_power": 0.0005153600798485183,
              "delta_db": 0.8740222414133542,
              "reference_active": true
            },
            {
              "low_hz": 4000.0,
              "high_hz": 8000.0,
              "reference_power": 1.5641285560683588e-05,
              "output_power": 1.8362426774995465e-05,
              "delta_db": 0.6965763198902687,
              "reference_active": true
            },
            {
              "low_hz": 8000.0,
              "high_hz": 16000.0,
              "reference_power": 9.365447124805669e-07,
              "output_power": 9.000457427361759e-08,
              "delta_db": -10.172639342667505,
              "reference_active": true
            },
            {
              "low_hz": 16000.0,
              "high_hz": 22050.0,
              "reference_power": 1.0738535418483273e-08,
              "output_power": 5.751954914476512e-11,
              "delta_db": -22.711295806324575,
              "reference_active": true
            }
          ],
          "power_floor": 1e-12,
          "active_band_count": 8,
          "band_count": 8,
          "active_band_mae_db": 5.057489866188605,
          "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
          "status": "ok"
        },
        "rms_gain_error_db": 2.07722285596235,
        "pitch": null,
        "output_rms_dbfs": -24.448451966803695,
        "output_peak": 0.6238571639547419,
        "reference_rms_dbfs": -26.52567482276605,
        "rms_gain_status": "ok",
        "mr_stft_log_magnitude_mae_db": 11.797144892748145
      },
      "common_prefix": {
        "samples": 220500,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 1.0,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 220500,
          "output_samples": 220500,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 13.706921473637687,
              "spectral_convergence": 0.4256170921182679,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.3526499269056254e-06
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 11.308300642418027,
              "spectral_convergence": 0.366146828033146,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.2238389615019609e-06
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 10.376212562188721,
              "spectral_convergence": 0.37582109282693693,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 8.265562353627178e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 5.873507160680166e-07,
                "output_power": 7.442924532683023e-07,
                "delta_db": 1.0284611293943793,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.00012765279053574292,
                "output_power": 0.00016048515418281173,
                "delta_db": 0.9940455073744613,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.00018664008076538165,
                "output_power": 0.00026103908270137024,
                "delta_db": 1.4570062094012726,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 0.0014726428134665878,
                "output_power": 0.002634417845834215,
                "delta_db": 2.5258723730430166,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 0.0004214136033266785,
                "output_power": 0.0005153600798485183,
                "delta_db": 0.8740222414133542,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.5641285560683588e-05,
                "output_power": 1.8362426774995465e-05,
                "delta_db": 0.6965763198902687,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 9.365447124805669e-07,
                "output_power": 9.000457427361759e-08,
                "delta_db": -10.172639342667505,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 1.0738535418483273e-08,
                "output_power": 5.751954914476512e-11,
                "delta_db": -22.711295806324575,
                "reference_active": true
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 8,
            "band_count": 8,
            "active_band_mae_db": 5.057489866188605,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": 2.07722285596235,
          "pitch": null,
          "output_rms_dbfs": -24.448451966803695,
          "output_peak": 0.6238571639547419,
          "reference_rms_dbfs": -26.52567482276605,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 11.797144892748145
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 220500,
          "output_samples": 220500,
          "intelligibility": {
            "estoi": {
              "value": 0.7157757289103673,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04717547252141775,
              "rms_dbfs": -26.52567482276605,
              "peak": 0.4219970703125,
              "peak_dbfs": -7.493811281786975,
              "crest_factor_db": 19.031863540979074,
              "mean": -7.474896696960033e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.05992077222373725,
              "rms_dbfs": -24.448451966803695,
              "peak": 0.6238571639547419,
              "peak_dbfs": -4.098296668118949,
              "crest_factor_db": 20.350155298684747,
              "mean": -2.253728190008224e-09,
              "zero_energy": false
            },
            "rms_difference_db": 2.0772228559623542
          },
          "full_scale": {
            "reference": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 500,
            "uncovered_tail_samples": 0,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 4.337883984370161e-06,
            "threshold_dbfs": -53.62722067264567,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 111,
            "quiet_samples": 48951,
            "reference_mean_power": 1.7511803859325063e-07,
            "output_mean_power": 1.6512795019999234e-07,
            "output_rms_dbfs": -67.82179410139678,
            "output_zero_energy": false,
            "output_peak": 0.00580257233989908,
            "output_energy_fraction": 1.0209835879993948e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 493,
            "reference_estimated_voiced_frames": 375,
            "output_estimated_voiced_frames": 384,
            "both_estimated_voiced_frames": 371,
            "reference_voiced_output_unvoiced_frames": 4,
            "reference_unvoiced_output_voiced_frames": 13,
            "paired_voiced_fraction": 0.9893333333333333,
            "absolute_difference_cents_median": 0.19224974425137262,
            "absolute_difference_cents_p95": 5.608772417901257,
            "difference_over_600_cents_frames": 5,
            "uncovered_tail_samples": 0,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 220500,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.00020852380901383914,
                      "log_rms_mae_db": 1.3102613390662203,
                      "rms_envelope_nrmse": 0.3383733355086049,
                      "envelope_correlation": 0.9692092660998494,
                      "rms_gain_error_db": 2.0771236598671243,
                      "reference_mean_power": 0.0022283113150001854,
                      "output_mean_power": 0.0035949117335466567
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.0625302814618026e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.329204569103168e-05,
                          "output_power": 0.0001723082550413708,
                          "reference_active": true,
                          "delta_db": 3.712492406989151
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.715947137354789e-05,
                          "output_power": 6.217920163428543e-05,
                          "reference_active": true,
                          "delta_db": 5.591412369633398
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.337117092557393e-06,
                          "output_power": 1.692459007786148e-05,
                          "reference_active": true,
                          "delta_db": 5.012114276318755
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 4.772006350980434
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.00017912166381814563,
                      "log_rms_mae_db": 1.5409379758596338,
                      "rms_envelope_nrmse": 0.33095887318255246,
                      "envelope_correlation": 0.9714314681492944,
                      "rms_gain_error_db": 2.0770813735815445,
                      "reference_mean_power": 0.002241741518969235,
                      "output_mean_power": 0.0036165433298762637
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.7748664101209405e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 6.50676491549953e-05,
                          "output_power": 0.00015939031062433786,
                          "reference_active": true,
                          "delta_db": 3.890968007097946
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.974133673656654e-06,
                          "output_power": 2.9533411854813705e-05,
                          "reference_active": true,
                          "delta_db": 5.686301101993289
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.542662112687682e-07,
                          "output_power": 1.2824166081516533e-06,
                          "reference_active": true,
                          "delta_db": 5.5869940103888
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 5.054754373160011
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 8.012931504544701e-05,
                      "log_rms_mae_db": 0.4511707122134686,
                      "rms_envelope_nrmse": 0.1860957855754044,
                      "envelope_correlation": 0.9935269584713566,
                      "rms_gain_error_db": 1.2241117279302585,
                      "reference_mean_power": 0.0002904911032130505,
                      "output_mean_power": 0.0003850738336977634
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.3656212534604894e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.531485150701641e-06,
                          "output_power": 1.6039890320127366e-05,
                          "reference_active": true,
                          "delta_db": 2.260408186646188
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.53385574612736e-06,
                          "output_power": 8.35260194572108e-06,
                          "reference_active": true,
                          "delta_db": 2.6535408697667693
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 9.811645139753325e-07,
                          "output_power": 1.6959781175786788e-06,
                          "reference_active": true,
                          "delta_db": 2.376784118757678
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.430244391723545
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 7.683290217697679e-05,
                      "log_rms_mae_db": 0.45324510413942226,
                      "rms_envelope_nrmse": 0.18205050068378578,
                      "envelope_correlation": 0.9944637251796024,
                      "rms_gain_error_db": 1.224133759194885,
                      "reference_mean_power": 0.0002922422310524987,
                      "output_mean_power": 0.00038739708690871177
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.0716920690025388e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.590477024775088e-06,
                          "output_power": 1.6671548240626467e-05,
                          "reference_active": true,
                          "delta_db": 2.4013572416556066
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.3745634613779177e-06,
                          "output_power": 4.342251060845616e-06,
                          "reference_active": true,
                          "delta_db": 2.6213114913303484
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.08439418728906e-08,
                          "output_power": 1.1336371460829421e-07,
                          "reference_active": true,
                          "delta_db": 2.702567253136051
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.5750786620406685
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.0001998844441775553,
                      "log_rms_mae_db": 0.9953534187856522,
                      "rms_envelope_nrmse": 0.3672819632357638,
                      "envelope_correlation": 0.9655119015073639,
                      "rms_gain_error_db": 2.2345696062524407,
                      "reference_mean_power": 0.0018349331226473488,
                      "output_mean_power": 0.003069567591117678
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.254653489774897e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.003134454909151e-05,
                          "output_power": 0.0001895457020923443,
                          "reference_active": true,
                          "delta_db": 3.7445382811632073
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.7426031096981302e-05,
                          "output_power": 6.171492291313465e-05,
                          "reference_active": true,
                          "delta_db": 5.491917061569788
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.403046539135653e-06,
                          "output_power": 1.611714213001529e-05,
                          "reference_active": true,
                          "delta_db": 4.746493274314828
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 4.660982872349274
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.0001632401576896862,
                      "log_rms_mae_db": 1.0521816374963775,
                      "rms_envelope_nrmse": 0.360546382104069,
                      "envelope_correlation": 0.9668840921984544,
                      "rms_gain_error_db": 2.2343310631335993,
                      "reference_mean_power": 0.0018460278065843946,
                      "output_mean_power": 0.003087957715806242
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.961502025877799e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.302776729830403e-05,
                          "output_power": 0.00017891627834127025,
                          "reference_active": true,
                          "delta_db": 3.891618327656521
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.934967558355204e-06,
                          "output_power": 2.8402021074025487e-05,
                          "reference_active": true,
                          "delta_db": 5.538040898238247
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.3641003921582925e-07,
                          "output_power": 1.4038366601685208e-06,
                          "reference_active": true,
                          "delta_db": 5.074218469267379
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 4.834625898387382
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 1.2279001954331291e-05,
                      "log_rms_mae_db": 0.8768884258876363,
                      "rms_envelope_nrmse": 0.15485392247023627,
                      "envelope_correlation": 0.9897739071052621,
                      "rms_gain_error_db": 1.0813696105149586,
                      "reference_mean_power": 2.821460992899086e-05,
                      "output_mean_power": 3.619186901990836e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.5952398033270154e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.299736325352575e-07,
                          "output_power": 1.1076331880350148e-06,
                          "reference_active": true,
                          "delta_db": 1.253316645657262
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.4173147699880047e-07,
                          "output_power": 2.3925938524835655e-07,
                          "reference_active": true,
                          "delta_db": 2.274026695498324
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.106516197351276e-08,
                          "output_power": 7.562498526077318e-08,
                          "reference_active": true,
                          "delta_db": 3.863916806539287
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.4637533825649576
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 1.0791287994309055e-05,
                      "log_rms_mae_db": 0.8722144891045633,
                      "rms_envelope_nrmse": 0.14613406870930132,
                      "envelope_correlation": 0.9936053575923423,
                      "rms_gain_error_db": 1.081637565645078,
                      "reference_mean_power": 2.8382119245712924e-05,
                      "output_mean_power": 3.640898541815365e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.2726489446417807e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.282257748405607e-07,
                          "output_power": 9.776525684879347e-07,
                          "reference_active": true,
                          "delta_db": 1.2791849928989996
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.7755829625744676e-08,
                          "output_power": 9.763387072882504e-08,
                          "reference_active": true,
                          "delta_db": 2.2800468080930587
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 1.7360058548804467e-09,
                          "output_power": 3.4530118624695265e-09,
                          "reference_active": true,
                          "delta_db": 2.986468847750538
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.181900216247532
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_ready",
      "method": "facodec",
      "status": "measured",
      "started_at": "2026-10-12T11:38:01.320693+00:00",
      "native_sample_rate": 16000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.0043345490000206155,
      "native_input_samples": 80000,
      "native_input_sha256": "272edc24c29f9bdefd63f8ea08c58dc4488fe68f37f59af32352ec31bc99c761",
      "output_sha256": "494669bff233a6f64ede6d2c5f3a8d38dcd2c54a55c368ac6909bf69f96f810a",
      "output_samples": 80000,
      "length_delta_samples": 0,
      "analysis_s": 1.1790952570000002,
      "synthesis_s": 3.9942248509999843,
      "total_s": 5.173320107999984,
      "rtf": 1.034664021599997,
      "output_peak": 0.35467514395713806,
      "output_rms": 0.041149328589184464,
      "finished_at": "2026-10-12T11:38:06.502619+00:00",
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "ready",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "ok",
        "reference_samples": 220500,
        "output_samples": 220500,
        "length_delta_samples": 0,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": [
          {
            "window_ms": 5,
            "window_samples": 220,
            "hop_samples": 55,
            "log_magnitude_mae_db": 14.925855531445801,
            "spectral_convergence": 0.3805259406039923,
            "spectral_convergence_status": "ok",
            "amplitude_floor": 1.3526499269056254e-06
          },
          {
            "window_ms": 20,
            "window_samples": 882,
            "hop_samples": 220,
            "log_magnitude_mae_db": 12.711501195601718,
            "spectral_convergence": 0.4844920526904217,
            "spectral_convergence_status": "ok",
            "amplitude_floor": 1.2238389615019609e-06
          },
          {
            "window_ms": 80,
            "window_samples": 3528,
            "hop_samples": 882,
            "log_magnitude_mae_db": 11.961962416236853,
            "spectral_convergence": 0.6809038479064725,
            "spectral_convergence_status": "ok",
            "amplitude_floor": 8.265562353627178e-07
          }
        ],
        "band_energy": {
          "bands": [
            {
              "low_hz": 0.0,
              "high_hz": 200.0,
              "reference_power": 5.873507160680166e-07,
              "output_power": 8.470009004099757e-08,
              "delta_db": -8.410136305478735,
              "reference_active": true
            },
            {
              "low_hz": 200.0,
              "high_hz": 500.0,
              "reference_power": 0.00012765279053574292,
              "output_power": 0.00016367410388966962,
              "delta_db": 1.079496587755175,
              "reference_active": true
            },
            {
              "low_hz": 500.0,
              "high_hz": 1000.0,
              "reference_power": 0.00018664008076538165,
              "output_power": 9.22199206707451e-05,
              "delta_db": -3.0618016942390103,
              "reference_active": true
            },
            {
              "low_hz": 1000.0,
              "high_hz": 2000.0,
              "reference_power": 0.0014726428134665878,
              "output_power": 0.001168662597829849,
              "delta_db": -1.004082773745376,
              "reference_active": true
            },
            {
              "low_hz": 2000.0,
              "high_hz": 4000.0,
              "reference_power": 0.0004214136033266785,
              "output_power": 0.00026410678109118366,
              "delta_db": -2.0292899851584965,
              "reference_active": true
            },
            {
              "low_hz": 4000.0,
              "high_hz": 8000.0,
              "reference_power": 1.5641285560683588e-05,
              "output_power": 5.991645487001855e-06,
              "delta_db": -4.167263357842017,
              "reference_active": true
            },
            {
              "low_hz": 8000.0,
              "high_hz": 16000.0,
              "reference_power": 9.365447124805669e-07,
              "output_power": 4.923931529455143e-08,
              "delta_db": -12.792165111286991,
              "reference_active": true
            },
            {
              "low_hz": 16000.0,
              "high_hz": 22050.0,
              "reference_power": 1.0738535418483273e-08,
              "output_power": 2.731194999479064e-11,
              "delta_db": -25.945923452666197,
              "reference_active": true
            }
          ],
          "power_floor": 1e-12,
          "active_band_count": 8,
          "band_count": 8,
          "active_band_mae_db": 7.311269908521499,
          "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
          "status": "ok"
        },
        "rms_gain_error_db": -1.1831687714694024,
        "pitch": null,
        "output_rms_dbfs": -27.708843594235454,
        "output_peak": 0.35731838274800654,
        "reference_rms_dbfs": -26.52567482276605,
        "rms_gain_status": "ok",
        "mr_stft_log_magnitude_mae_db": 13.199773047761456
      },
      "common_prefix": {
        "samples": 220500,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 1.0,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 220500,
          "output_samples": 220500,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 14.925855531445801,
              "spectral_convergence": 0.3805259406039923,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.3526499269056254e-06
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 12.711501195601718,
              "spectral_convergence": 0.4844920526904217,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.2238389615019609e-06
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 11.961962416236853,
              "spectral_convergence": 0.6809038479064725,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 8.265562353627178e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 5.873507160680166e-07,
                "output_power": 8.470009004099757e-08,
                "delta_db": -8.410136305478735,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.00012765279053574292,
                "output_power": 0.00016367410388966962,
                "delta_db": 1.079496587755175,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.00018664008076538165,
                "output_power": 9.22199206707451e-05,
                "delta_db": -3.0618016942390103,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 0.0014726428134665878,
                "output_power": 0.001168662597829849,
                "delta_db": -1.004082773745376,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 0.0004214136033266785,
                "output_power": 0.00026410678109118366,
                "delta_db": -2.0292899851584965,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.5641285560683588e-05,
                "output_power": 5.991645487001855e-06,
                "delta_db": -4.167263357842017,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 9.365447124805669e-07,
                "output_power": 4.923931529455143e-08,
                "delta_db": -12.792165111286991,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 1.0738535418483273e-08,
                "output_power": 2.731194999479064e-11,
                "delta_db": -25.945923452666197,
                "reference_active": true
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 8,
            "band_count": 8,
            "active_band_mae_db": 7.311269908521499,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": -1.1831687714694024,
          "pitch": null,
          "output_rms_dbfs": -27.708843594235454,
          "output_peak": 0.35731838274800654,
          "reference_rms_dbfs": -26.52567482276605,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 13.199773047761456
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 220500,
          "output_samples": 220500,
          "intelligibility": {
            "estoi": {
              "value": 0.34502210178108533,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04717547252141775,
              "rms_dbfs": -26.52567482276605,
              "peak": 0.4219970703125,
              "peak_dbfs": -7.493811281786975,
              "crest_factor_db": 19.031863540979074,
              "mean": -7.474896696960033e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.04116781528920055,
              "rms_dbfs": -27.708843594235447,
              "peak": 0.35731838274800654,
              "peak_dbfs": -8.93889280661697,
              "crest_factor_db": 18.769950787618484,
              "mean": 8.351107741376886e-06,
              "zero_energy": false
            },
            "rms_difference_db": -1.183168771469397
          },
          "full_scale": {
            "reference": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 500,
            "uncovered_tail_samples": 0,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 4.337883984370161e-06,
            "threshold_dbfs": -53.62722067264567,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 111,
            "quiet_samples": 48951,
            "reference_mean_power": 1.7511803859325063e-07,
            "output_mean_power": 2.5407099414644184e-07,
            "output_rms_dbfs": -65.95044913073144,
            "output_zero_energy": false,
            "output_peak": 0.006037012870586404,
            "output_energy_fraction": 3.3280697584465026e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 493,
            "reference_estimated_voiced_frames": 375,
            "output_estimated_voiced_frames": 360,
            "both_estimated_voiced_frames": 353,
            "reference_voiced_output_unvoiced_frames": 22,
            "reference_unvoiced_output_voiced_frames": 7,
            "paired_voiced_fraction": 0.9413333333333334,
            "absolute_difference_cents_median": 7.03027779184326,
            "absolute_difference_cents_p95": 26.748388997590737,
            "difference_over_600_cents_frames": 1,
            "uncovered_tail_samples": 0,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 220500,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.00020852380901383914,
                      "log_rms_mae_db": 1.3855318766776734,
                      "rms_envelope_nrmse": 0.1937470507014923,
                      "envelope_correlation": 0.9646916200544612,
                      "rms_gain_error_db": -1.1841832912332653,
                      "reference_mean_power": 0.0022283113150001854,
                      "output_mean_power": 0.0016965143459042717
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.0625302814618026e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.329204569103168e-05,
                          "output_power": 4.962273379778331e-05,
                          "reference_active": true,
                          "delta_db": -1.6937615708278442
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.715947137354789e-05,
                          "output_power": 2.1319548241652746e-05,
                          "reference_active": true,
                          "delta_db": 0.9427409326574081
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.337117092557393e-06,
                          "output_power": 5.9897801155197985e-06,
                          "reference_active": true,
                          "delta_db": 0.5010414873621721
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.0458479969491414
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.00017912166381814563,
                      "log_rms_mae_db": 1.3835959376933098,
                      "rms_envelope_nrmse": 0.17408195047878777,
                      "envelope_correlation": 0.9754492078517071,
                      "rms_gain_error_db": -1.1839700531622683,
                      "reference_mean_power": 0.002241741518969235,
                      "output_mean_power": 0.0017068231713260097
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.7748664101209405e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 6.50676491549953e-05,
                          "output_power": 4.5510757038274716e-05,
                          "reference_active": true,
                          "delta_db": -1.5525105654711164
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.974133673656654e-06,
                          "output_power": 1.0235510984828946e-05,
                          "reference_active": true,
                          "delta_db": 1.0842601707041883
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.542662112687682e-07,
                          "output_power": 3.1282619219000514e-07,
                          "reference_active": true,
                          "delta_db": -0.5402662439726489
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.0590123267159848
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 8.012931504544701e-05,
                      "log_rms_mae_db": 1.515403916776559,
                      "rms_envelope_nrmse": 0.32033175566868843,
                      "envelope_correlation": 0.8972727486829567,
                      "rms_gain_error_db": -0.6908492964870581,
                      "reference_mean_power": 0.0002904911032130505,
                      "output_mean_power": 0.0002477695352830159
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.3656212534604894e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.531485150701641e-06,
                          "output_power": 6.270749032350668e-06,
                          "reference_active": true,
                          "delta_db": -1.8184115580431468
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.53385574612736e-06,
                          "output_power": 4.9203865632606625e-06,
                          "reference_active": true,
                          "delta_db": 0.3553152579868138
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 9.811645139753325e-07,
                          "output_power": 8.964585108871337e-07,
                          "reference_active": true,
                          "delta_db": -0.3921163785628575
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.8552810648642728
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 7.683290217697679e-05,
                      "log_rms_mae_db": 1.4844915509009455,
                      "rms_envelope_nrmse": 0.3108108369445424,
                      "envelope_correlation": 0.9022044406108098,
                      "rms_gain_error_db": -0.69084964723792,
                      "reference_mean_power": 0.0002922422310524987,
                      "output_mean_power": 0.0002492631103985017
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.0716920690025388e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.590477024775088e-06,
                          "output_power": 6.430770601230177e-06,
                          "reference_active": true,
                          "delta_db": -1.7357719157936615
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.3745634613779177e-06,
                          "output_power": 2.601561382726581e-06,
                          "reference_active": true,
                          "delta_db": 0.3965029653819129
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.08439418728906e-08,
                          "output_power": 5.179722041983995e-08,
                          "reference_active": true,
                          "delta_db": -0.699108878169896
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.9437945864484902
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.0001998844441775553,
                      "log_rms_mae_db": 1.2748478665993552,
                      "rms_envelope_nrmse": 0.20365840164509239,
                      "envelope_correlation": 0.9652365994408377,
                      "rms_gain_error_db": -1.2835697674893474,
                      "reference_mean_power": 0.0018349331226473488,
                      "output_mean_power": 0.0013654105793959512
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.254653489774897e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.003134454909151e-05,
                          "output_power": 6.222869848359243e-05,
                          "reference_active": true,
                          "delta_db": -1.0926939543647487
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.7426031096981302e-05,
                          "output_power": 2.00660414178348e-05,
                          "reference_active": true,
                          "delta_db": 0.6126321962517505
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.403046539135653e-06,
                          "output_power": 6.007821442363167e-06,
                          "reference_active": true,
                          "delta_db": 0.4607830823084072
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7220364109749688
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.0001632401576896862,
                      "log_rms_mae_db": 1.202590370554248,
                      "rms_envelope_nrmse": 0.18188767734843878,
                      "envelope_correlation": 0.9766035291664764,
                      "rms_gain_error_db": -1.2833473440969956,
                      "reference_mean_power": 0.0018460278065843946,
                      "output_mean_power": 0.0013737367104363686
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.961502025877799e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.302776729830403e-05,
                          "output_power": 5.877588113852494e-05,
                          "reference_active": true,
                          "delta_db": -0.9428887444709845
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.934967558355204e-06,
                          "output_power": 9.027708007651071e-06,
                          "reference_active": true,
                          "delta_db": 0.5603234829717592
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.3641003921582925e-07,
                          "output_power": 4.0034256626071313e-07,
                          "reference_active": true,
                          "delta_db": -0.37462963836114577
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.6259472886012966
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 1.2279001954331291e-05,
                      "log_rms_mae_db": 2.4019795727351476,
                      "rms_envelope_nrmse": 0.326146778809146,
                      "envelope_correlation": 0.9144389530649537,
                      "rms_gain_error_db": -2.6997347133741223,
                      "reference_mean_power": 2.821460992899086e-05,
                      "output_mean_power": 1.515306824342481e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.5952398033270154e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.299736325352575e-07,
                          "output_power": 4.418481243661932e-07,
                          "reference_active": true,
                          "delta_db": -2.7379127971225925
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.4173147699880047e-07,
                          "output_power": 1.9811901616660284e-07,
                          "reference_active": true,
                          "delta_db": 1.4545984968866144
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.106516197351276e-08,
                          "output_power": 9.179701170947746e-08,
                          "reference_active": true,
                          "delta_db": 4.705549213740193
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.9660201692498
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 1.0791287994309055e-05,
                      "log_rms_mae_db": 2.3569755324249573,
                      "rms_envelope_nrmse": 0.31166905301947767,
                      "envelope_correlation": 0.9326135163786664,
                      "rms_gain_error_db": -2.699651728894015,
                      "reference_mean_power": 2.8382119245712924e-05,
                      "output_mean_power": 1.5243322832731613e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.2726489446417807e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.282257748405607e-07,
                          "output_power": 3.9403394964126585e-07,
                          "reference_active": true,
                          "delta_db": -2.6673240440600443
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.7755829625744676e-08,
                          "output_power": 8.823667824091582e-08,
                          "reference_active": true,
                          "delta_db": 1.8405332368534357
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 1.7360058548804467e-09,
                          "output_power": 3.0175630208602636e-09,
                          "reference_active": true,
                          "delta_db": 2.4010516341728327
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.302969638362104
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_ready",
      "method": "hifiglot",
      "status": "measured",
      "started_at": "2026-10-12T11:38:21.070803+00:00",
      "native_sample_rate": 44100,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 1.225299999418894e-05,
      "native_input_samples": 220500,
      "native_input_sha256": "035587c5437df50083bc61d5cbb58f9b30924e025a3a5b4adcc2d515f61fc7cd",
      "output_sha256": "dc1a49806350ead4504393f7240f65ceb51273c6ed777c37a320e6b31dce06b1",
      "output_samples": 220672,
      "length_delta_samples": 172,
      "analysis_s": 0.10221455999999307,
      "synthesis_s": 15.413723549999986,
      "total_s": 15.51593810999998,
      "rtf": 3.1031876219999956,
      "output_peak": 0.42336228489875793,
      "output_rms": 0.056271649688794015,
      "finished_at": "2026-10-12T11:38:36.592987+00:00",
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "ready",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 220500,
        "output_samples": 220672,
        "length_delta_samples": 172,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -24.994207053153282,
        "output_peak": 0.42336228489875793
      },
      "common_prefix": {
        "samples": 220500,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 0.9992205626450116,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 220500,
          "output_samples": 220500,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 7.224079209860828,
              "spectral_convergence": 0.512623283136978,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.3526499269056254e-06
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 7.949672912775516,
              "spectral_convergence": 0.513504441726716,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.2238389615019609e-06
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 7.9776011076294076,
              "spectral_convergence": 0.5226298827481729,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 8.265562353627178e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 5.873507160680166e-07,
                "output_power": 2.0861538068125213e-07,
                "delta_db": -4.4955117786035,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.00012765279053574292,
                "output_power": 7.241080683387935e-05,
                "delta_db": -2.4622692646445126,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.00018664008076538165,
                "output_power": 0.0003815814317261881,
                "delta_db": 3.1058231946394654,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 0.0014726428134665878,
                "output_power": 0.0021880491251937487,
                "delta_db": 1.7195964600709228,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 0.0004214136033266785,
                "output_power": 0.0005038438563971653,
                "delta_db": 0.7758741675199952,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.5641285560683588e-05,
                "output_power": 2.2083911358564097e-05,
                "delta_db": 1.4980355033079324,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 9.365447124805669e-07,
                "output_power": 7.816248308559302e-07,
                "delta_db": -0.7853016856901156,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 1.0738535418483273e-08,
                "output_power": 9.182442918916158e-09,
                "delta_db": -0.6798681658360177,
                "reference_active": true
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 8,
            "band_count": 8,
            "active_band_mae_db": 1.9402850275390577,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": 1.5348541196207375,
          "pitch": null,
          "output_rms_dbfs": -24.99082070314531,
          "output_peak": 0.42336228489875793,
          "reference_rms_dbfs": -26.52567482276605,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 7.717117743421917
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 220500,
          "output_samples": 220500,
          "intelligibility": {
            "estoi": {
              "value": 0.336126174681578,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04717547252141775,
              "rms_dbfs": -26.52567482276605,
              "peak": 0.4219970703125,
              "peak_dbfs": -7.493811281786975,
              "crest_factor_db": 19.031863540979074,
              "mean": -7.474896696960033e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.056293592478753755,
              "rms_dbfs": -24.99082070314531,
              "peak": 0.42336228489875793,
              "peak_dbfs": -7.465756671722397,
              "crest_factor_db": 17.525064031422914,
              "mean": 4.360211831521382e-05,
              "zero_energy": false
            },
            "rms_difference_db": 1.5348541196207393
          },
          "full_scale": {
            "reference": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 500,
            "uncovered_tail_samples": 0,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 4.337883984370161e-06,
            "threshold_dbfs": -53.62722067264567,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 111,
            "quiet_samples": 48951,
            "reference_mean_power": 1.7511803859325063e-07,
            "output_mean_power": 1.419455248634013e-07,
            "output_rms_dbfs": -68.47878294985726,
            "output_zero_energy": false,
            "output_peak": 0.004426656290888786,
            "output_energy_fraction": 9.943900035949764e-06,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 493,
            "reference_estimated_voiced_frames": 375,
            "output_estimated_voiced_frames": 383,
            "both_estimated_voiced_frames": 373,
            "reference_voiced_output_unvoiced_frames": 2,
            "reference_unvoiced_output_voiced_frames": 10,
            "paired_voiced_fraction": 0.9946666666666667,
            "absolute_difference_cents_median": 0.2947438041506784,
            "absolute_difference_cents_p95": 11.117904729926371,
            "difference_over_600_cents_frames": 6,
            "uncovered_tail_samples": 0,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 220500,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.00020852380901383914,
                      "log_rms_mae_db": 1.8838406884315242,
                      "rms_envelope_nrmse": 0.2860425186521888,
                      "envelope_correlation": 0.9458110899542325,
                      "rms_gain_error_db": 1.5352996178032055,
                      "reference_mean_power": 0.0022283113150001854,
                      "output_mean_power": 0.0031732612422010844
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.0625302814618026e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.329204569103168e-05,
                          "output_power": 0.00010305398637957824,
                          "reference_active": true,
                          "delta_db": 1.4800795233683608
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.715947137354789e-05,
                          "output_power": 1.4058833497580354e-05,
                          "reference_active": true,
                          "delta_db": -0.8655461704605385
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.337117092557393e-06,
                          "output_power": 4.170253444414875e-06,
                          "reference_active": true,
                          "delta_db": -1.0714428128270304
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.1390228355519765
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.00017912166381814563,
                      "log_rms_mae_db": 1.9039641978384687,
                      "rms_envelope_nrmse": 0.2807483605941755,
                      "envelope_correlation": 0.9465787686772509,
                      "rms_gain_error_db": 1.5350074829073348,
                      "reference_mean_power": 0.002241741518969235,
                      "output_mean_power": 0.003192171998576661
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.7748664101209405e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 6.50676491549953e-05,
                          "output_power": 9.081734816350777e-05,
                          "reference_active": true,
                          "delta_db": 1.448037002092547
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.974133673656654e-06,
                          "output_power": 6.702846483484739e-06,
                          "reference_active": true,
                          "delta_db": -0.7542423886109468
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.542662112687682e-07,
                          "output_power": 2.72185753191005e-07,
                          "reference_active": true,
                          "delta_db": -1.1446434322621568
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.1156409409885502
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 8.012931504544701e-05,
                      "log_rms_mae_db": 1.579355857749785,
                      "rms_envelope_nrmse": 0.3893072263815689,
                      "envelope_correlation": 0.9257250490787289,
                      "rms_gain_error_db": 1.5076775212599727,
                      "reference_mean_power": 0.0002904911032130505,
                      "output_mean_power": 0.0004110556178594613
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.3656212534604894e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.531485150701641e-06,
                          "output_power": 1.4616216465942219e-05,
                          "reference_active": true,
                          "delta_db": 1.8567439067443272
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.53385574612736e-06,
                          "output_power": 4.3607212880326054e-06,
                          "reference_active": true,
                          "delta_db": -0.16909368064564487
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 9.811645139753325e-07,
                          "output_power": 4.999956914851718e-07,
                          "reference_active": true,
                          "delta_db": -2.927755705887577
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.6511977644258498
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 7.683290217697679e-05,
                      "log_rms_mae_db": 1.5077122704280097,
                      "rms_envelope_nrmse": 0.3691320768003338,
                      "envelope_correlation": 0.9361011378229318,
                      "rms_gain_error_db": 1.5076625415250098,
                      "reference_mean_power": 0.0002922422310524987,
                      "output_mean_power": 0.00041353210193258123
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.0716920690025388e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.590477024775088e-06,
                          "output_power": 1.326333864558235e-05,
                          "reference_active": true,
                          "delta_db": 1.4081264910691715
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.3745634613779177e-06,
                          "output_power": 2.0255272541367966e-06,
                          "reference_active": true,
                          "delta_db": -0.6904568967273181
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.08439418728906e-08,
                          "output_power": 5.3646436421688765e-08,
                          "reference_active": true,
                          "delta_db": -0.5467646443267997
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.8817826773744297
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.0001998844441775553,
                      "log_rms_mae_db": 2.149202843028595,
                      "rms_envelope_nrmse": 0.3548607056979559,
                      "envelope_correlation": 0.8941833502231106,
                      "rms_gain_error_db": 1.5235787139446964,
                      "reference_mean_power": 0.0018349331226473488,
                      "output_mean_power": 0.002606022203707518
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.254653489774897e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.003134454909151e-05,
                          "output_power": 0.00010784075228137575,
                          "reference_active": true,
                          "delta_db": 1.2952279555218251
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.7426031096981302e-05,
                          "output_power": 1.2245923418942762e-05,
                          "reference_active": true,
                          "delta_db": -1.5320694549696499
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.403046539135653e-06,
                          "output_power": 4.297180073108372e-06,
                          "reference_active": true,
                          "delta_db": -0.9945515519736957
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.273949654155057
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.0001632401576896862,
                      "log_rms_mae_db": 2.1184423601203943,
                      "rms_envelope_nrmse": 0.3476564898075324,
                      "envelope_correlation": 0.8938396722993142,
                      "rms_gain_error_db": 1.5231147896548023,
                      "reference_mean_power": 0.0018460278065843946,
                      "output_mean_power": 0.0026214991271709233
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.961502025877799e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.302776729830403e-05,
                          "output_power": 9.808174043960408e-05,
                          "reference_active": true,
                          "delta_db": 1.2810014068795614
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.934967558355204e-06,
                          "output_power": 5.8250568297226965e-06,
                          "reference_active": true,
                          "delta_db": -1.342449887766323
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.3641003921582925e-07,
                          "output_power": 2.744824419027839e-07,
                          "reference_active": true,
                          "delta_db": -2.0138016386229363
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.5457509777562735
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 1.2279001954331291e-05,
                      "log_rms_mae_db": 3.6640634952406734,
                      "rms_envelope_nrmse": 0.619196649874388,
                      "envelope_correlation": 0.8207978173963645,
                      "rms_gain_error_db": 3.396829195172957,
                      "reference_mean_power": 2.821460992899086e-05,
                      "output_mean_power": 6.168179022285126e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.5952398033270154e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.299736325352575e-07,
                          "output_power": 1.474852237708926e-06,
                          "reference_active": true,
                          "delta_db": 2.496842159815883
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.4173147699880047e-07,
                          "output_power": 2.849385982175516e-07,
                          "reference_active": true,
                          "delta_db": 3.0328497044112557
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.106516197351276e-08,
                          "output_power": 1.2097099314337783e-07,
                          "reference_active": true,
                          "delta_db": 5.904076237551282
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 3.8112560339261403
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 1.0791287994309055e-05,
                      "log_rms_mae_db": 3.606593847318181,
                      "rms_envelope_nrmse": 0.6092571738362381,
                      "envelope_correlation": 0.8273381384123768,
                      "rms_gain_error_db": 3.396963206366596,
                      "reference_mean_power": 2.8382119245712924e-05,
                      "output_mean_power": 6.204990787141805e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.2726489446417807e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.282257748405607e-07,
                          "output_power": 1.3437529573288212e-06,
                          "reference_active": true,
                          "delta_db": 2.6605338675587915
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.7755829625744676e-08,
                          "output_power": 1.309188388178112e-07,
                          "reference_active": true,
                          "delta_db": 3.5540631808720313
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 1.7360058548804467e-09,
                          "output_power": 3.6228317972218624e-09,
                          "reference_active": true,
                          "delta_db": 3.1949698529413277
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 3.136522300457384
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_ready",
      "method": "golf",
      "status": "measured",
      "started_at": "2026-10-12T11:38:40.793019+00:00",
      "native_sample_rate": 24000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.0037092610000115656,
      "native_input_samples": 120000,
      "native_input_sha256": "dddfd7245c59e0021e5e988aeb4aeb7c71ad5e04be2c4136d770dbe5fcf09741",
      "output_sha256": "0933c56a09081db6de1534ed14df14221a8568ee1ca71a5ed4b27945cfa5e855",
      "output_samples": 119760,
      "length_delta_samples": -240,
      "analysis_s": 0.45059510699999805,
      "synthesis_s": 0.093428638000006,
      "total_s": 0.544023745000004,
      "rtf": 0.10880474900000081,
      "output_peak": 0.4308757781982422,
      "output_rms": 0.051661320820703904,
      "finished_at": "2026-10-12T11:38:41.345387+00:00",
      "analysis_stages": {
        "external_f0_s": 0.03577826299999742
      },
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "ready",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 220500,
        "output_samples": 220059,
        "length_delta_samples": -441,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -25.7298015016079,
        "output_peak": 0.4312636237752331
      },
      "common_prefix": {
        "samples": 220059,
        "sample_rate": 44100,
        "reference_coverage": 0.998,
        "output_coverage": 1.0,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 220059,
          "output_samples": 220059,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 10.990314170026819,
              "spectral_convergence": 0.4831987330796843,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.3526499269056254e-06
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 9.134164882753717,
              "spectral_convergence": 0.4796268535236363,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.2238389615019609e-06
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 8.509767663146263,
              "spectral_convergence": 0.5498726497639543,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 8.265562353627178e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 5.884231172079857e-07,
                "output_power": 1.5690864604037016e-06,
                "delta_db": 4.25957148515579,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.00012791157215647585,
                "output_power": 0.0002788972003838936,
                "delta_db": 3.385343179021656,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.00018701360704311871,
                "output_power": 0.0007546714885932222,
                "delta_db": 6.058847359058271,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 0.001477193380996421,
                "output_power": 0.0013756373249047473,
                "delta_db": -0.30933402058862,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 0.0004206612896242682,
                "output_power": 0.00025392887642097957,
                "delta_db": -2.1922045790093,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.566754374949324e-05,
                "output_power": 8.13299017834199e-06,
                "delta_db": -2.847506682495889,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 9.384096549377224e-07,
                "output_power": 2.9142657659838704e-07,
                "delta_db": -5.078633127668031,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 1.0751680935286762e-08,
                "output_power": 1.902180288074168e-10,
                "delta_db": -17.522246910133518,
                "reference_active": true
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 8,
            "band_count": 8,
            "active_band_mae_db": 5.206710917891384,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": 0.7871791234430239,
          "pitch": null,
          "output_rms_dbfs": -25.7298015016079,
          "output_peak": 0.4312636237752331,
          "reference_rms_dbfs": -26.51698062505092,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 9.544748905308934
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 220059,
          "output_samples": 220059,
          "intelligibility": {
            "estoi": {
              "value": 0.4082852590948492,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04722271675817539,
              "rms_dbfs": -26.516980625050923,
              "peak": 0.4219970703125,
              "peak_dbfs": -7.493811281786975,
              "crest_factor_db": 19.023169343263945,
              "mean": -7.551366750827335e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.051702307334742965,
              "rms_dbfs": -25.7298015016079,
              "peak": 0.4312636237752331,
              "peak_dbfs": -7.3051434436163065,
              "crest_factor_db": 18.42465805799159,
              "mean": 0.00037747512016915745,
              "zero_energy": false
            },
            "rms_difference_db": 0.7871791234430248
          },
          "full_scale": {
            "reference": {
              "sample_count": 220059,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 220059,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 499,
            "uncovered_tail_samples": 0,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 4.337883984370161e-06,
            "threshold_dbfs": -53.62722067264567,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 110,
            "quiet_samples": 48510,
            "reference_mean_power": 1.7580296517868333e-07,
            "output_mean_power": 3.953879257049371e-07,
            "output_rms_dbfs": -64.02976597237281,
            "output_zero_energy": false,
            "output_peak": 0.005019027618106649,
            "output_energy_fraction": 3.2605862475656036e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 492,
            "reference_estimated_voiced_frames": 375,
            "output_estimated_voiced_frames": 369,
            "both_estimated_voiced_frames": 367,
            "reference_voiced_output_unvoiced_frames": 8,
            "reference_unvoiced_output_voiced_frames": 2,
            "paired_voiced_fraction": 0.9786666666666667,
            "absolute_difference_cents_median": 2.9111888380541466,
            "absolute_difference_cents_p95": 24.102793353097674,
            "difference_over_600_cents_frames": 8,
            "uncovered_tail_samples": 0,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 220059,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220001,
                    "uncovered_tail_samples": 58,
                    "metrics": {
                      "frame_count": 999,
                      "rms_floor": 0.00020852380901383914,
                      "log_rms_mae_db": 1.8963604553513496,
                      "rms_envelope_nrmse": 0.21680203657830044,
                      "envelope_correlation": 0.9533745505347846,
                      "rms_gain_error_db": 0.7869842836667091,
                      "reference_mean_power": 0.002232772245597914,
                      "output_mean_power": 0.002676349665211639
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 999,
                      "frequency_resolution_hz": 0.20065520065520065,
                      "power_floor": 2.0597254148426382e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.373118717772707e-05,
                          "output_power": 0.00012686197955899623,
                          "reference_active": true,
                          "delta_db": 2.3568025713874734
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.7310312804383096e-05,
                          "output_power": 1.722305866490086e-05,
                          "reference_active": true,
                          "delta_db": -0.02194634947416558
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.378583229114294e-06,
                          "output_power": 6.850049549693232e-06,
                          "reference_active": true,
                          "delta_db": 1.0502581960398376
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 1.143002372300492
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220004,
                    "uncovered_tail_samples": 55,
                    "metrics": {
                      "frame_count": 993,
                      "rms_floor": 0.00017912166381814563,
                      "log_rms_mae_db": 1.9020657133080958,
                      "rms_envelope_nrmse": 0.17570319039303658,
                      "envelope_correlation": 0.9724575366145036,
                      "rms_gain_error_db": 0.7866886316138033,
                      "reference_mean_power": 0.0022462564210779675,
                      "output_mean_power": 0.0026923294065529634
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 993,
                      "frequency_resolution_hz": 0.20186761878604778,
                      "power_floor": 1.7710251294161318e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 6.545260209175443e-05,
                          "output_power": 0.00011404435530758543,
                          "reference_active": true,
                          "delta_db": 2.4114687772048344
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.044609212955205e-06,
                          "output_power": 8.51869942118935e-06,
                          "reference_active": true,
                          "delta_db": 0.24868342607169233
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.5390215393574344e-07,
                          "output_power": 3.5039556928079056e-07,
                          "reference_active": true,
                          "delta_db": -0.04324599951177406
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.901132734262767
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220001,
                    "uncovered_tail_samples": 58,
                    "metrics": {
                      "frame_count": 999,
                      "rms_floor": 8.012931504544701e-05,
                      "log_rms_mae_db": 4.346607351580633,
                      "rms_envelope_nrmse": 0.8721710191684702,
                      "envelope_correlation": 0.9296171097528086,
                      "rms_gain_error_db": 5.105429428873764,
                      "reference_mean_power": 0.00029107266648795534,
                      "output_mean_power": 0.0009430709502242723
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 999,
                      "frequency_resolution_hz": 0.20065520065520065,
                      "power_floor": 2.3816320452902608e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.609237032827167e-06,
                          "output_power": 2.8079527791073524e-05,
                          "reference_active": true,
                          "delta_db": 4.657008937204331
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.5643735118564604e-06,
                          "output_power": 4.031809837075947e-06,
                          "reference_active": true,
                          "delta_db": -0.5388113625072908
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 9.827841606601934e-07,
                          "output_power": 2.120723565103162e-06,
                          "reference_active": true,
                          "delta_db": 3.340259137854797
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.84535981252214
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220004,
                    "uncovered_tail_samples": 55,
                    "metrics": {
                      "frame_count": 993,
                      "rms_floor": 7.683290217697679e-05,
                      "log_rms_mae_db": 4.327935588789155,
                      "rms_envelope_nrmse": 0.8537881329226473,
                      "envelope_correlation": 0.9464163943907654,
                      "rms_gain_error_db": 5.105289179122221,
                      "reference_mean_power": 0.0002928308352722173,
                      "output_mean_power": 0.000948736751255621
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 993,
                      "frequency_resolution_hz": 0.20186761878604778,
                      "power_floor": 2.086263643848518e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.66968448171574e-06,
                          "output_power": 2.5158013209780006e-05,
                          "reference_active": true,
                          "delta_db": 4.152640373799877
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.3839055264184183e-06,
                          "output_power": 1.9081967557188544e-06,
                          "reference_active": true,
                          "delta_db": -0.9666588729317533
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.120266506222832e-08,
                          "output_power": 8.394021209168446e-08,
                          "reference_active": true,
                          "delta_db": 1.371997283534402
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.1637655100886772
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220001,
                    "uncovered_tail_samples": 58,
                    "metrics": {
                      "frame_count": 999,
                      "rms_floor": 0.0001998844441775553,
                      "log_rms_mae_db": 1.0915915065379187,
                      "rms_envelope_nrmse": 0.23000043452380242,
                      "envelope_correlation": 0.9300208522063967,
                      "rms_gain_error_db": -0.6672869698839189,
                      "reference_mean_power": 0.0018386066621304744,
                      "output_mean_power": 0.0015767401712002904
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 999,
                      "frequency_resolution_hz": 0.20065520065520065,
                      "power_floor": 2.2519386689159903e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.053607630194526e-05,
                          "output_power": 0.00011727440920649402,
                          "reference_active": true,
                          "delta_db": 1.6321278671333805
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.758412199423595e-05,
                          "output_power": 1.7080743651079405e-05,
                          "reference_active": true,
                          "delta_db": -0.12613913301341104
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.42822138760416e-06,
                          "output_power": 5.629990434321465e-06,
                          "reference_active": true,
                          "delta_db": 0.1585010510366926
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.638922683727828
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220004,
                    "uncovered_tail_samples": 55,
                    "metrics": {
                      "frame_count": 993,
                      "rms_floor": 0.0001632401576896862,
                      "log_rms_mae_db": 1.0020343971090613,
                      "rms_envelope_nrmse": 0.19654211294348733,
                      "envelope_correlation": 0.9480415433644066,
                      "rms_gain_error_db": -0.6676787973597873,
                      "reference_mean_power": 0.0018497458884908957,
                      "output_mean_power": 0.0015861497640100354
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 993,
                      "frequency_resolution_hz": 0.20186761878604778,
                      "power_floor": 1.957761283295251e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.347522796400107e-05,
                          "output_power": 0.00010802897337854723,
                          "reference_active": true,
                          "delta_db": 1.6739930673380152
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.001437008670636e-06,
                          "output_power": 8.130055363762987e-06,
                          "reference_active": true,
                          "delta_db": 0.06925512443890067
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.3661169763897524e-07,
                          "output_power": 3.3697044513715423e-07,
                          "reference_active": true,
                          "delta_db": -1.125035554625298
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.9560945821340713
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220001,
                    "uncovered_tail_samples": 58,
                    "metrics": {
                      "frame_count": 999,
                      "rms_floor": 1.2279001954331291e-05,
                      "log_rms_mae_db": 1.437189041138648,
                      "rms_envelope_nrmse": 0.25497082475163957,
                      "envelope_correlation": 0.9043924643192282,
                      "rms_gain_error_db": 0.6485585301423733,
                      "reference_mean_power": 2.8271095562234332e-05,
                      "output_mean_power": 3.2824528132122754e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 999,
                      "frequency_resolution_hz": 0.20065520065520065,
                      "power_floor": 2.575697961795058e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.356293205979798e-07,
                          "output_power": 1.6380089090924198e-06,
                          "reference_active": true,
                          "delta_db": 2.9230258942911522
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.4104938967777985e-07,
                          "output_power": 2.3274691877661593e-07,
                          "reference_active": true,
                          "delta_db": 2.1751272897920564
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.117906204652672e-08,
                          "output_power": 5.5507753870609825e-08,
                          "reference_active": true,
                          "delta_db": 2.5049060764210305
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.5343530868347464
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220004,
                    "uncovered_tail_samples": 55,
                    "metrics": {
                      "frame_count": 993,
                      "rms_floor": 1.0791287994309055e-05,
                      "log_rms_mae_db": 1.3081588856282866,
                      "rms_envelope_nrmse": 0.23579445048753045,
                      "envelope_correlation": 0.9152822381929906,
                      "rms_gain_error_db": 0.6487123925206307,
                      "reference_mean_power": 2.8439283548861125e-05,
                      "output_mean_power": 3.302097486121014e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 993,
                      "frequency_resolution_hz": 0.20186761878604778,
                      "power_floor": 2.2526160990882865e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.331096919680938e-07,
                          "output_power": 1.4885463821541523e-06,
                          "reference_active": true,
                          "delta_db": 3.0759341046222954
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.6730149415042495e-08,
                          "output_power": 9.66721778904202e-08,
                          "reference_active": true,
                          "delta_db": 2.314875753522246
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 1.747361378006809e-09,
                          "output_power": 2.1753094671187317e-09,
                          "reference_active": true,
                          "delta_db": 0.9513831768017467
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 2.1140643449820957
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_ready",
      "method": "sifigan",
      "status": "measured",
      "started_at": "2026-10-12T11:38:49.068265+00:00",
      "native_sample_rate": 24000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.003757610999997496,
      "native_input_samples": 120000,
      "native_input_sha256": "dddfd7245c59e0021e5e988aeb4aeb7c71ad5e04be2c4136d770dbe5fcf09741",
      "output_sha256": "ccd041236800dd59ae933bd9c446d7d8f755018078009d0501ae3fec4c86c947",
      "output_samples": 120120,
      "length_delta_samples": 120,
      "analysis_s": 1.8092308079999952,
      "synthesis_s": 3.3290062490000025,
      "total_s": 5.138237056999998,
      "rtf": 1.0276474113999996,
      "output_peak": 0.41929689049720764,
      "output_rms": 0.046095989684039565,
      "finished_at": "2026-10-12T11:38:54.215013+00:00",
      "analysis_stages": {
        "external_f0_s": 1.1321069440000144
      },
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "ready",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 220500,
        "output_samples": 220721,
        "length_delta_samples": 221,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -26.72013463137936,
        "output_peak": 0.42363463471793833
      },
      "common_prefix": {
        "samples": 220500,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 0.998998735960783,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 220500,
          "output_samples": 220500,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 10.29965697058859,
              "spectral_convergence": 0.2972657450676697,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.3526499269056254e-06
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 9.153398439438625,
              "spectral_convergence": 0.19313906892452393,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.2238389615019609e-06
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 8.72994336219266,
              "spectral_convergence": 0.22400171113913633,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 8.265562353627178e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 5.873507160680166e-07,
                "output_power": 5.344657652909068e-08,
                "delta_db": -10.409776104531883,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.00012765279053574292,
                "output_power": 0.0001316900401151227,
                "delta_db": 0.13522616883906885,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.00018664008076538165,
                "output_power": 0.0002143354236443541,
                "delta_db": 0.6008903997244753,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 0.0014726428134665878,
                "output_power": 0.0014308739794666957,
                "delta_db": -0.12496036289509638,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 0.0004214136033266785,
                "output_power": 0.0003395905465495613,
                "delta_db": -0.9375295868126393,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.5641285560683588e-05,
                "output_power": 1.3017178253956286e-05,
                "delta_db": -0.7975559294060277,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 9.365447124805669e-07,
                "output_power": 6.451944672043532e-07,
                "delta_db": -1.6183788178881808,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 1.0738535418483273e-08,
                "output_power": 1.6137842162922554e-10,
                "delta_db": -18.230995903824834,
                "reference_active": true
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 8,
            "band_count": 8,
            "active_band_mae_db": 4.106914159240276,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": -0.19010920150401395,
          "pitch": null,
          "output_rms_dbfs": -26.71578402427006,
          "output_peak": 0.42363463471793833,
          "reference_rms_dbfs": -26.52567482276605,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 9.394332924073291
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 220500,
          "output_samples": 220500,
          "intelligibility": {
            "estoi": {
              "value": 0.5765472409524762,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04717547252141775,
              "rms_dbfs": -26.52567482276605,
              "peak": 0.4219970703125,
              "peak_dbfs": -7.493811281786975,
              "crest_factor_db": 19.031863540979074,
              "mean": -7.474896696960033e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.04615415442245524,
              "rms_dbfs": -26.71578402427006,
              "peak": 0.42363463471793833,
              "peak_dbfs": -7.460170817750349,
              "crest_factor_db": 19.255613206519712,
              "mean": 4.0300001098981803e-05,
              "zero_energy": false
            },
            "rms_difference_db": -0.19010920150401134
          },
          "full_scale": {
            "reference": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 500,
            "uncovered_tail_samples": 0,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 4.337883984370161e-06,
            "threshold_dbfs": -53.62722067264567,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 111,
            "quiet_samples": 48951,
            "reference_mean_power": 1.7511803859325063e-07,
            "output_mean_power": 1.1818075862479789e-07,
            "output_rms_dbfs": -69.27453226528411,
            "output_zero_energy": false,
            "output_peak": 0.005221399923870944,
            "output_energy_fraction": 1.2316240203354655e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 493,
            "reference_estimated_voiced_frames": 375,
            "output_estimated_voiced_frames": 380,
            "both_estimated_voiced_frames": 372,
            "reference_voiced_output_unvoiced_frames": 3,
            "reference_unvoiced_output_voiced_frames": 8,
            "paired_voiced_fraction": 0.992,
            "absolute_difference_cents_median": 0.7825685602895722,
            "absolute_difference_cents_p95": 6.8734876612142255,
            "difference_over_600_cents_frames": 3,
            "uncovered_tail_samples": 0,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 220500,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.00020852380901383914,
                      "log_rms_mae_db": 0.9244160477645068,
                      "rms_envelope_nrmse": 0.11254321907638488,
                      "envelope_correlation": 0.9830567910068433,
                      "rms_gain_error_db": -0.1901461103236049,
                      "reference_mean_power": 0.0022283113150001854,
                      "output_mean_power": 0.00213285462257235
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.0625302814618026e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.329204569103168e-05,
                          "output_power": 7.976878397937804e-05,
                          "reference_active": true,
                          "delta_db": 0.367761279056944
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.715947137354789e-05,
                          "output_power": 1.6750279409037584e-05,
                          "reference_active": true,
                          "delta_db": -0.10481848706883254
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.337117092557393e-06,
                          "output_power": 5.203604010867873e-06,
                          "reference_active": true,
                          "delta_db": -0.1100249125517379
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.19420155955917148
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.00017912166381814563,
                      "log_rms_mae_db": 1.046780854359007,
                      "rms_envelope_nrmse": 0.07696116219391527,
                      "envelope_correlation": 0.9922759564960416,
                      "rms_gain_error_db": -0.19016391509934363,
                      "reference_mean_power": 0.002241741518969235,
                      "output_mean_power": 0.0021457007049523477
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.7748664101209405e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 6.50676491549953e-05,
                          "output_power": 7.061754715925193e-05,
                          "reference_active": true,
                          "delta_db": 0.3554751235191941
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.974133673656654e-06,
                          "output_power": 7.708853206275572e-06,
                          "reference_active": true,
                          "delta_db": -0.1469373582682124
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.542662112687682e-07,
                          "output_power": 3.358404285093412e-07,
                          "reference_active": true,
                          "delta_db": -0.2319675737861607
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.24479335185785575
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 8.012931504544701e-05,
                      "log_rms_mae_db": 0.6769551939982852,
                      "rms_envelope_nrmse": 0.13036127261565003,
                      "envelope_correlation": 0.9856721987057843,
                      "rms_gain_error_db": 0.40062004422208386,
                      "reference_mean_power": 0.0002904911032130505,
                      "output_mean_power": 0.0003185626389340989
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.3656212534604894e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.531485150701641e-06,
                          "output_power": 1.1102362085947776e-05,
                          "reference_active": true,
                          "delta_db": 0.6625481144089214
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.53385574612736e-06,
                          "output_power": 3.966552848739889e-06,
                          "reference_active": true,
                          "delta_db": -0.5805445302101204
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 9.811645139753325e-07,
                          "output_power": 9.06239544515842e-07,
                          "reference_active": true,
                          "delta_db": -0.3449882352284871
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5293602932825096
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 7.683290217697679e-05,
                      "log_rms_mae_db": 0.6367754429425413,
                      "rms_envelope_nrmse": 0.10782551299487354,
                      "envelope_correlation": 0.9906219067075759,
                      "rms_gain_error_db": 0.40067618516059933,
                      "reference_mean_power": 0.0002922422310524987,
                      "output_mean_power": 0.0003204871294655864
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.0716920690025388e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.590477024775088e-06,
                          "output_power": 1.0661426980354446e-05,
                          "reference_active": true,
                          "delta_db": 0.45975127529537924
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.3745634613779177e-06,
                          "output_power": 2.060340041249827e-06,
                          "reference_active": true,
                          "delta_db": -0.6164487798582461
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.08439418728906e-08,
                          "output_power": 5.198373460956328e-08,
                          "reference_active": true,
                          "delta_db": -0.6834986594009256
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5865662381848503
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.0001998844441775553,
                      "log_rms_mae_db": 0.674763424821533,
                      "rms_envelope_nrmse": 0.12724051290453403,
                      "envelope_correlation": 0.9790255526685455,
                      "rms_gain_error_db": -0.27136526656978566,
                      "reference_mean_power": 0.0018349331226473488,
                      "output_mean_power": 0.0017237874463540254
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.254653489774897e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.003134454909151e-05,
                          "output_power": 8.249360351386009e-05,
                          "reference_active": true,
                          "delta_db": 0.1316016179397311
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.7426031096981302e-05,
                          "output_power": 1.7051822918670652e-05,
                          "reference_active": true,
                          "delta_db": -0.09427670805474553
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.403046539135653e-06,
                          "output_power": 5.038294321403778e-06,
                          "reference_active": true,
                          "delta_db": -0.3035517432126067
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.17647668973569444
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.0001632401576896862,
                      "log_rms_mae_db": 0.593921613817399,
                      "rms_envelope_nrmse": 0.08807688809134796,
                      "envelope_correlation": 0.9901440267150707,
                      "rms_gain_error_db": -0.27142947151074726,
                      "reference_mean_power": 0.0018460278065843946,
                      "output_mean_power": 0.001734184464536202
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.961502025877799e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.302776729830403e-05,
                          "output_power": 7.479088065069016e-05,
                          "reference_active": true,
                          "delta_db": 0.10360623983908865
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.934967558355204e-06,
                          "output_power": 7.697923041561936e-06,
                          "reference_active": true,
                          "delta_db": -0.13171590498734892
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.3641003921582925e-07,
                          "output_power": 3.9709930462404733e-07,
                          "reference_active": true,
                          "delta_db": -0.4099560602589367
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.21509273502845808
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 1.2279001954331291e-05,
                      "log_rms_mae_db": 1.2533424402219606,
                      "rms_envelope_nrmse": 0.16365533409847152,
                      "envelope_correlation": 0.9824762761735233,
                      "rms_gain_error_db": -1.2696260624439037,
                      "reference_mean_power": 2.821460992899086e-05,
                      "output_mean_power": 2.1062574008602303e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.5952398033270154e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.299736325352575e-07,
                          "output_power": 5.984223811245963e-07,
                          "reference_active": true,
                          "delta_db": -1.4205646761053754
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.4173147699880047e-07,
                          "output_power": 1.172061181699577e-07,
                          "reference_active": true,
                          "delta_db": -0.8251603049193497
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.106516197351276e-08,
                          "output_power": 3.59225450023147e-08,
                          "reference_active": true,
                          "delta_db": 0.6309347515634318
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.9588865775293857
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 1.0791287994309055e-05,
                      "log_rms_mae_db": 1.1983571007375284,
                      "rms_envelope_nrmse": 0.1490340701470433,
                      "envelope_correlation": 0.9916538353478562,
                      "rms_gain_error_db": -1.2695838106653459,
                      "reference_mean_power": 2.8382119245712924e-05,
                      "output_mean_power": 2.118782802834638e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.2726489446417807e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.282257748405607e-07,
                          "output_power": 5.230497689948059e-07,
                          "reference_active": true,
                          "delta_db": -1.4372303167395715
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.7755829625744676e-08,
                          "output_power": 5.0422603792271596e-08,
                          "reference_active": true,
                          "delta_db": -0.5897055797943438
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 1.7360058548804467e-09,
                          "output_power": 1.8199509115249778e-09,
                          "reference_active": true,
                          "delta_db": 0.20508488619862003
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7440069275775119
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_ready",
      "method": "llsm2",
      "status": "measured",
      "started_at": "2026-10-12T11:38:56.405591+00:00",
      "native_sample_rate": 24000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "none",
      "resampling_and_input_io_s": 0.003710253000008379,
      "native_input_samples": 120000,
      "native_input_sha256": "dddfd7245c59e0021e5e988aeb4aeb7c71ad5e04be2c4136d770dbe5fcf09741",
      "output_sha256": "3baea50ae0c122c785cabe04e43f527a39a2a7235613bcc9380e955dd802bab8",
      "output_samples": 120240,
      "length_delta_samples": 240,
      "analysis_s": 0.26474908300002653,
      "synthesis_s": 0.04625066199997718,
      "total_s": 0.3109997450000037,
      "rtf": 0.06219994900000074,
      "output_peak": 0.4241194725036621,
      "output_rms": 0.04668167428770978,
      "finished_at": "2026-10-12T11:38:56.725091+00:00",
      "analysis_stages": {
        "external_f0_s": 0.06409111899998265
      },
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "ready",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 220500,
        "output_samples": 220941,
        "length_delta_samples": 441,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -26.61064124772139,
        "output_peak": 0.42488280739591755
      },
      "common_prefix": {
        "samples": 220500,
        "sample_rate": 44100,
        "reference_coverage": 1.0,
        "output_coverage": 0.998003992015968,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 220500,
          "output_samples": 220500,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 9.090184036770028,
              "spectral_convergence": 0.09378890283259245,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.3526499269056254e-06
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 7.748310252693219,
              "spectral_convergence": 0.12427611432457171,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.2238389615019609e-06
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 7.331475896358237,
              "spectral_convergence": 0.15940036382164097,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 8.265562353627178e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 5.873507160680166e-07,
                "output_power": 4.893576529057766e-07,
                "delta_db": -0.7927111805981379,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.00012765279053574292,
                "output_power": 0.0001255605591786987,
                "delta_db": -0.07177071932412826,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.00018664008076538165,
                "output_power": 0.0001820742838310355,
                "delta_db": -0.10756303338850437,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 0.0014726428134665878,
                "output_power": 0.001447740885113901,
                "delta_db": -0.07406583044654463,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 0.0004214136033266785,
                "output_power": 0.00041501682712492393,
                "delta_db": -0.06642844803764796,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.5641285560683588e-05,
                "output_power": 1.5200124189704188e-05,
                "delta_db": -0.12425308670135587,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 9.365447124805669e-07,
                "output_power": 6.902393450190621e-07,
                "delta_db": -1.325288049636229,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 1.0738535418483273e-08,
                "output_power": 1.7088966673904045e-10,
                "delta_db": -17.98229251053653,
                "reference_active": true
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 8,
            "band_count": 8,
            "active_band_mae_db": 2.5680466073336348,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": -0.07628924449058676,
          "pitch": null,
          "output_rms_dbfs": -26.601964067256635,
          "output_peak": 0.42488280739591755,
          "reference_rms_dbfs": -26.52567482276605,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 8.05665672860716
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 220500,
          "output_samples": 220500,
          "intelligibility": {
            "estoi": {
              "value": 0.8170171849656821,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04717547252141775,
              "rms_dbfs": -26.52567482276605,
              "peak": 0.4219970703125,
              "peak_dbfs": -7.493811281786975,
              "crest_factor_db": 19.031863540979074,
              "mean": -7.474896696960033e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.04676293882259599,
              "rms_dbfs": -26.601964067256638,
              "peak": 0.42488280739591755,
              "peak_dbfs": -7.434616839925274,
              "crest_factor_db": 19.167347227331362,
              "mean": -1.367725636674605e-06,
              "zero_energy": false
            },
            "rms_difference_db": -0.07628924449058871
          },
          "full_scale": {
            "reference": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 220500,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 500,
            "uncovered_tail_samples": 0,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 4.337883984370161e-06,
            "threshold_dbfs": -53.62722067264567,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 111,
            "quiet_samples": 48951,
            "reference_mean_power": 1.7511803859325063e-07,
            "output_mean_power": 1.3662566204493297e-07,
            "output_rms_dbfs": -68.64467720582849,
            "output_zero_energy": false,
            "output_peak": 0.004173045880391039,
            "output_energy_fraction": 1.3870166057317009e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 493,
            "reference_estimated_voiced_frames": 375,
            "output_estimated_voiced_frames": 369,
            "both_estimated_voiced_frames": 369,
            "reference_voiced_output_unvoiced_frames": 6,
            "reference_unvoiced_output_voiced_frames": 0,
            "paired_voiced_fraction": 0.984,
            "absolute_difference_cents_median": 0.05045175289020211,
            "absolute_difference_cents_p95": 3.393675023684278,
            "difference_over_600_cents_frames": 1,
            "uncovered_tail_samples": 0,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 220500,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.00020852380901383914,
                      "log_rms_mae_db": 0.34326197534774566,
                      "rms_envelope_nrmse": 0.028686474986934024,
                      "envelope_correlation": 0.9989306392823374,
                      "rms_gain_error_db": -0.07620984599322236,
                      "reference_mean_power": 0.0022283113150001854,
                      "output_mean_power": 0.002189550069989246
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.0625302814618026e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.329204569103168e-05,
                          "output_power": 7.215598390679254e-05,
                          "reference_active": true,
                          "delta_db": -0.06784490642404443
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.715947137354789e-05,
                          "output_power": 1.5053253633873584e-05,
                          "reference_active": true,
                          "delta_db": -0.568735253779112
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.337117092557393e-06,
                          "output_power": 4.036015245355334e-06,
                          "reference_active": true,
                          "delta_db": -1.2135393297962205
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.6167064966664589
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.00017912166381814563,
                      "log_rms_mae_db": 0.3543822954388748,
                      "rms_envelope_nrmse": 0.017320973025388996,
                      "envelope_correlation": 0.9996645904460563,
                      "rms_gain_error_db": -0.07611592076134054,
                      "reference_mean_power": 0.002241741518969235,
                      "output_mean_power": 0.0022027942964502343
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.7748664101209405e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 6.50676491549953e-05,
                          "output_power": 6.407392792146912e-05,
                          "reference_active": true,
                          "delta_db": -0.06683767940387007
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.974133673656654e-06,
                          "output_power": 6.888144438708975e-06,
                          "reference_active": true,
                          "delta_db": -0.6358126630476258
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.542662112687682e-07,
                          "output_power": 2.412018646580618e-07,
                          "reference_active": true,
                          "delta_db": -1.6694907186795827
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7907136870436928
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 8.012931504544701e-05,
                      "log_rms_mae_db": 0.11124821547999034,
                      "rms_envelope_nrmse": 0.03894203184947144,
                      "envelope_correlation": 0.9985530882096455,
                      "rms_gain_error_db": -0.09629259470402138,
                      "reference_mean_power": 0.0002904911032130505,
                      "output_mean_power": 0.00028412115824811977
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.3656212534604894e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.531485150701641e-06,
                          "output_power": 9.248074031184832e-06,
                          "reference_active": true,
                          "delta_db": -0.13109277978072542
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.53385574612736e-06,
                          "output_power": 4.234846891759973e-06,
                          "reference_active": true,
                          "delta_db": -0.2962998473986202
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 9.811645139753325e-07,
                          "output_power": 1.050537949107813e-06,
                          "reference_active": true,
                          "delta_db": 0.29669912681916005
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.24136391799950188
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 7.683290217697679e-05,
                      "log_rms_mae_db": 0.0815363485917743,
                      "rms_envelope_nrmse": 0.021544230494013897,
                      "envelope_correlation": 0.9996412744439525,
                      "rms_gain_error_db": -0.09625290542900691,
                      "reference_mean_power": 0.0002922422310524987,
                      "output_mean_power": 0.00028583649920715156
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.0716920690025388e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.590477024775088e-06,
                          "output_power": 9.388251441896218e-06,
                          "reference_active": true,
                          "delta_db": -0.09255496641069166
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.3745634613779177e-06,
                          "output_power": 2.2068288775543884e-06,
                          "reference_active": true,
                          "delta_db": -0.31815122505000176
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.08439418728906e-08,
                          "output_power": 6.25844943075365e-08,
                          "reference_active": true,
                          "delta_db": 0.12249404654031185
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.17773341266700174
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 0.0001998844441775553,
                      "log_rms_mae_db": 0.12586461881505662,
                      "rms_envelope_nrmse": 0.030613182589309323,
                      "envelope_correlation": 0.9988162966347879,
                      "rms_gain_error_db": -0.07301348121798633,
                      "reference_mean_power": 0.0018349331226473488,
                      "output_mean_power": 0.0018043421401756437
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.254653489774897e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.003134454909151e-05,
                          "output_power": 7.911253984924801e-05,
                          "reference_active": true,
                          "delta_db": -0.0501478578643386
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.7426031096981302e-05,
                          "output_power": 1.5370666742571737e-05,
                          "reference_active": true,
                          "delta_db": -0.5450577814404072
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.403046539135653e-06,
                          "output_power": 3.7877285726633683e-06,
                          "reference_active": true,
                          "delta_db": -1.542598582984437
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.7126014074297276
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 0.0001632401576896862,
                      "log_rms_mae_db": 0.09460132151174547,
                      "rms_envelope_nrmse": 0.019467452470696422,
                      "envelope_correlation": 0.9995556989644706,
                      "rms_gain_error_db": -0.07295212086612848,
                      "reference_mean_power": 0.0018460278065843946,
                      "output_mean_power": 0.0018152775071332598
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 1.961502025877799e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.302776729830403e-05,
                          "output_power": 7.20674886712039e-05,
                          "reference_active": true,
                          "delta_db": -0.05748634512505006
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 7.934967558355204e-06,
                          "output_power": 6.775810211598751e-06,
                          "reference_active": true,
                          "delta_db": -0.6858392247173285
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.3641003921582925e-07,
                          "output_power": 2.828885829106903e-07,
                          "reference_active": true,
                          "delta_db": -1.88279312443991
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.8753728980940961
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220441,
                    "uncovered_tail_samples": 59,
                    "metrics": {
                      "frame_count": 1001,
                      "rms_floor": 1.2279001954331291e-05,
                      "log_rms_mae_db": 0.20614960622471298,
                      "rms_envelope_nrmse": 0.03116453879019935,
                      "envelope_correlation": 0.998288829203619,
                      "rms_gain_error_db": -0.06048680117300754,
                      "reference_mean_power": 2.821460992899086e-05,
                      "output_mean_power": 2.7824371956267778e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 1001,
                      "frequency_resolution_hz": 0.20025429116338206,
                      "power_floor": 2.5952398033270154e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.299736325352575e-07,
                          "output_power": 8.20032635374113e-07,
                          "reference_active": true,
                          "delta_db": -0.052331588482132055
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.4173147699880047e-07,
                          "output_power": 1.4017639929079228e-07,
                          "reference_active": true,
                          "delta_db": -0.047914129023936075
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.106516197351276e-08,
                          "output_power": 3.333616386955522e-08,
                          "reference_active": true,
                          "delta_db": 0.30641999939407427
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.13555523896671415
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 220444,
                    "uncovered_tail_samples": 56,
                    "metrics": {
                      "frame_count": 995,
                      "rms_floor": 1.0791287994309055e-05,
                      "log_rms_mae_db": 0.16025492292490925,
                      "rms_envelope_nrmse": 0.016855521869296546,
                      "envelope_correlation": 0.9995364064609046,
                      "rms_gain_error_db": -0.06037055698067861,
                      "reference_mean_power": 2.8382119245712924e-05,
                      "output_mean_power": 2.799031362623112e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 995,
                      "frequency_resolution_hz": 0.20146185472818637,
                      "power_floor": 2.2726489446417807e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.282257748405607e-07,
                          "output_power": 7.251592820829895e-07,
                          "reference_active": true,
                          "delta_db": -0.01832635960105909
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.7755829625744676e-08,
                          "output_power": 5.727808728585168e-08,
                          "reference_active": true,
                          "delta_db": -0.036073196573321524
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 1.7360058548804467e-09,
                          "output_power": 2.2104091462149884e-09,
                          "reference_active": true,
                          "delta_db": 1.0492148338144849
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.36787146332962184
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    },
    {
      "case": "smoke_ready",
      "method": "magphase",
      "status": "measured",
      "started_at": "2026-10-12T11:39:01.346204+00:00",
      "native_sample_rate": 48000,
      "seed": 0,
      "metrics_include_timing": false,
      "input_quantization": "PCM16 (MagPhase/REAPER file API only)",
      "resampling_and_input_io_s": 0.007717811000020447,
      "native_input_samples": 240000,
      "native_input_sha256": "086184530ec1b5775e153d2ce765b5534da93e3dc8770111d230ad454fae6251",
      "output_sha256": "e89bf4634f4a719993db579a8994baf7a626eb2f524ff0adbbf2b933f9aee1ff",
      "output_samples": 239300,
      "length_delta_samples": -700,
      "analysis_s": 4.546094397000019,
      "synthesis_s": 0.957288138999985,
      "total_s": 5.503382536000004,
      "rtf": 1.1006765072000007,
      "output_peak": 0.4264111936297729,
      "output_rms": 0.04708988084790324,
      "finished_at": "2026-10-12T11:39:06.864747+00:00",
      "kind": "natural_speech",
      "language": "en",
      "style": "normal",
      "speaker": "ready",
      "mos": {
        "status": "not_requested",
        "value": null
      },
      "measurement": {
        "status": "length_mismatch",
        "reference_samples": 220500,
        "output_samples": 219857,
        "length_delta_samples": -643,
        "sample_rate": 44100,
        "reference_finite": true,
        "output_finite": true,
        "alignment": "none",
        "gain_normalization": "none",
        "mr_stft": null,
        "band_energy": null,
        "rms_gain_error_db": null,
        "pitch": null,
        "output_rms_dbfs": -26.532663220280956,
        "output_peak": 0.42698697455074064
      },
      "common_prefix": {
        "samples": 219857,
        "sample_rate": 44100,
        "reference_coverage": 0.9970839002267574,
        "output_coverage": 1.0,
        "alignment": "none",
        "gain_normalization": "none",
        "measurement": {
          "status": "ok",
          "reference_samples": 219857,
          "output_samples": 219857,
          "length_delta_samples": 0,
          "sample_rate": 44100,
          "reference_finite": true,
          "output_finite": true,
          "alignment": "none",
          "gain_normalization": "none",
          "mr_stft": [
            {
              "window_ms": 5,
              "window_samples": 220,
              "hop_samples": 55,
              "log_magnitude_mae_db": 5.579881066578077,
              "spectral_convergence": 0.2403306508787327,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.3526499269056254e-06
            },
            {
              "window_ms": 20,
              "window_samples": 882,
              "hop_samples": 220,
              "log_magnitude_mae_db": 6.967487400660371,
              "spectral_convergence": 0.222743088752754,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 1.2238389615019609e-06
            },
            {
              "window_ms": 80,
              "window_samples": 3528,
              "hop_samples": 882,
              "log_magnitude_mae_db": 7.768608011210196,
              "spectral_convergence": 0.4847445420116632,
              "spectral_convergence_status": "ok",
              "amplitude_floor": 8.265562353627178e-07
            }
          ],
          "band_energy": {
            "bands": [
              {
                "low_hz": 0.0,
                "high_hz": 200.0,
                "reference_power": 5.889419485239042e-07,
                "output_power": 5.640628300876777e-07,
                "delta_db": -0.18745006812270276,
                "reference_active": true
              },
              {
                "low_hz": 200.0,
                "high_hz": 500.0,
                "reference_power": 0.00012802701693057983,
                "output_power": 0.00011569016878968776,
                "delta_db": -0.4400517171058169,
                "reference_active": true
              },
              {
                "low_hz": 500.0,
                "high_hz": 1000.0,
                "reference_power": 0.0001871856819711674,
                "output_power": 0.00017577697251412448,
                "delta_db": -0.27310645881442974,
                "reference_active": true
              },
              {
                "low_hz": 1000.0,
                "high_hz": 2000.0,
                "reference_power": 0.0014771725771421872,
                "output_power": 0.0014116041570246623,
                "delta_db": -0.19718307985898106,
                "reference_active": true
              },
              {
                "low_hz": 2000.0,
                "high_hz": 4000.0,
                "reference_power": 0.0004224238934077821,
                "output_power": 0.0004912176977553533,
                "delta_db": 0.6552553026166164,
                "reference_active": true
              },
              {
                "low_hz": 4000.0,
                "high_hz": 8000.0,
                "reference_power": 1.5685661539919334e-05,
                "output_power": 2.5258749884507958e-05,
                "delta_db": 2.0690901277562617,
                "reference_active": true
              },
              {
                "low_hz": 8000.0,
                "high_hz": 16000.0,
                "reference_power": 9.392760521614802e-07,
                "output_power": 1.820625244273878e-06,
                "delta_db": 2.874273105615852,
                "reference_active": true
              },
              {
                "low_hz": 16000.0,
                "high_hz": 22050.0,
                "reference_power": 1.0759840145708458e-08,
                "output_power": 1.4476167044033427e-08,
                "delta_db": 1.2884776664235453,
                "reference_active": true
              }
            ],
            "power_floor": 1e-12,
            "active_band_count": 8,
            "band_count": 8,
            "active_band_mae_db": 0.9981109407892759,
            "active_policy": "reference power > max(1e-12, total reference power * 1e-6)",
            "status": "ok"
          },
          "rms_gain_error_db": -0.019670906440746087,
          "pitch": null,
          "output_rms_dbfs": -26.532663220280956,
          "output_peak": 0.42698697455074064,
          "reference_rms_dbfs": -26.51299231384021,
          "rms_gain_status": "ok",
          "mr_stft_log_magnitude_mae_db": 6.771992159482881
        },
        "extended": {
          "schema": "extended-diagnostics-v1",
          "status": "measured",
          "alignment": "none",
          "gain_normalization": "none",
          "sample_rate": 44100,
          "reference_samples": 219857,
          "output_samples": 219857,
          "intelligibility": {
            "estoi": {
              "value": 0.6648310812517151,
              "library": "pystoi",
              "version": "0.4.1",
              "extended": true,
              "reference_required": true,
              "internal_sample_rate": 10000,
              "evaluation_seed": 0,
              "internal_silence_removal_db": 40,
              "warnings": [],
              "normalization": "公式実装の行・列正規化",
              "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない",
              "status": "measured"
            }
          },
          "level": {
            "reference": {
              "rms": 0.04724440505321733,
              "rms_dbfs": -26.51299231384021,
              "peak": 0.4219970703125,
              "peak_dbfs": -7.493811281786975,
              "crest_factor_db": 19.019181032053236,
              "mean": -7.574336937617405e-05,
              "zero_energy": false
            },
            "output": {
              "rms": 0.04713753186378919,
              "rms_dbfs": -26.532663220280956,
              "peak": 0.42698697455074064,
              "peak_dbfs": -7.391707462825975,
              "crest_factor_db": 19.140955757454982,
              "mean": 0.00021644844043141996,
              "zero_energy": false
            },
            "rms_difference_db": -0.019670906440744318
          },
          "full_scale": {
            "reference": {
              "sample_count": 219857,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            },
            "output": {
              "sample_count": 219857,
              "full_scale_amplitude": 1.0,
              "at_or_above_full_scale_samples": 0,
              "at_or_above_full_scale_fraction": 0.0,
              "near_full_scale_amplitude": 0.999,
              "near_full_scale_fraction": 0.0,
              "longest_full_scale_run_samples": 0,
              "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"
            }
          },
          "quiet_regions": {
            "block_ms": 10.0,
            "block_samples": 441,
            "total_blocks": 498,
            "uncovered_tail_samples": 239,
            "meaning": "参照低エネルギー区間。人が付けた無音・無声ラベルではない",
            "status": "measured",
            "threshold_power": 4.337883984370161e-06,
            "threshold_dbfs": -53.62722067264567,
            "threshold_policy": "max(-60 dBFS, 最大参照ブロック電力より40 dB下)",
            "quiet_blocks": 109,
            "quiet_samples": 48069,
            "reference_mean_power": 1.7706885487172897e-07,
            "output_mean_power": 1.2680728906704805e-07,
            "output_rms_dbfs": -68.9685578185954,
            "output_zero_energy": false,
            "output_peak": 0.0059875078212248455,
            "output_energy_fraction": 1.2477722704040698e-05,
            "energy_fraction_denominator": "出力波形全体のエネルギー"
          },
          "pitch_consistency": {
            "status": "estimated_consistency_not_ground_truth",
            "estimator": "independent-yin-difference-v2-prefiltered",
            "fmin_hz": 65.0,
            "fmax_hz": 700.0,
            "frame_ms": 80.0,
            "hop_ms": 10.0,
            "prefilter": {
              "kind": "butterworth-forward-backward",
              "order_per_pass": 6,
              "cutoff_hz": 2000.0,
              "pad_type": "even",
              "max_pad_samples": 3528
            },
            "total_frames": 491,
            "reference_estimated_voiced_frames": 375,
            "output_estimated_voiced_frames": 372,
            "both_estimated_voiced_frames": 372,
            "reference_voiced_output_unvoiced_frames": 3,
            "reference_unvoiced_output_voiced_frames": 0,
            "paired_voiced_fraction": 0.992,
            "absolute_difference_cents_median": 6.667760554663094,
            "absolute_difference_cents_p95": 15.266020265746898,
            "difference_over_600_cents_frames": 1,
            "uncovered_tail_samples": 239,
            "scope": "自然音声の参照F0は未知。同じ独立測定器の入出力推定値の整合性のみ"
          },
          "temporal": {
            "schema": "temporal-diagnostics-v2",
            "status": "measured",
            "sample_rate": 44100,
            "samples": 219857,
            "alignment": "none",
            "normalization": "none",
            "bands": [
              {
                "name": "全帯域",
                "status": "measured",
                "filter": {
                  "kind": "none"
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 219781,
                    "uncovered_tail_samples": 76,
                    "metrics": {
                      "frame_count": 998,
                      "rms_floor": 0.00020852380901383914,
                      "log_rms_mae_db": 0.8050554130171905,
                      "rms_envelope_nrmse": 0.04530466922418516,
                      "envelope_correlation": 0.9970685821137274,
                      "rms_gain_error_db": -0.019376774272336555,
                      "reference_mean_power": 0.0022350094550475214,
                      "output_mean_power": 0.0022250597992932966
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 998,
                      "frequency_resolution_hz": 0.20085625797048645,
                      "power_floor": 2.058370945209882e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.395108325646015e-05,
                          "output_power": 7.252546863500578e-05,
                          "reference_active": true,
                          "delta_db": -0.08453996300794354
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.7387229070245606e-05,
                          "output_power": 1.6281817522358632e-05,
                          "reference_active": true,
                          "delta_db": -0.28527492726821935
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.388766642157797e-06,
                          "output_power": 5.07941543879277e-06,
                          "reference_active": true,
                          "delta_db": -0.256756424350902
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.2088571048756883
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 219784,
                    "uncovered_tail_samples": 73,
                    "metrics": {
                      "frame_count": 992,
                      "rms_floor": 0.00017912166381814563,
                      "log_rms_mae_db": 0.7481580145143083,
                      "rms_envelope_nrmse": 0.03710739495137002,
                      "envelope_correlation": 0.9979374010303406,
                      "rms_gain_error_db": -0.019202552030005003,
                      "reference_mean_power": 0.002248520709230695,
                      "output_mean_power": 0.00223860070728735
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 992,
                      "frequency_resolution_hz": 0.20207111436950148,
                      "power_floor": 1.769160003826187e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 6.564589642328533e-05,
                          "output_power": 6.440811329837093e-05,
                          "reference_active": true,
                          "delta_db": -0.08267005594429055
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.064408230629815e-06,
                          "output_power": 7.492783245744582e-06,
                          "reference_active": true,
                          "delta_db": -0.3192933486457248
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.6935605747980315e-07,
                          "output_power": 3.532923472173636e-07,
                          "reference_active": true,
                          "delta_db": -0.1931099575248074
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.19835778737160759
                    }
                  }
                ]
              },
              {
                "name": "300–1000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    300.0,
                    1000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 219781,
                    "uncovered_tail_samples": 76,
                    "metrics": {
                      "frame_count": 998,
                      "rms_floor": 8.012931504544701e-05,
                      "log_rms_mae_db": 0.41829228571330684,
                      "rms_envelope_nrmse": 0.06080074615516596,
                      "envelope_correlation": 0.9977121411088031,
                      "rms_gain_error_db": -0.34830326532147715,
                      "reference_mean_power": 0.0002913643221335893,
                      "output_mean_power": 0.00026890943755487163
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 998,
                      "frequency_resolution_hz": 0.20085625797048645,
                      "power_floor": 2.3897533292688163e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.648504422322574e-06,
                          "output_power": 8.63186269228168e-06,
                          "reference_active": true,
                          "delta_db": -0.4835547686476837
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 4.579470317502962e-06,
                          "output_power": 4.029811430116791e-06,
                          "reference_active": true,
                          "delta_db": -0.5553052406065557
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 9.830474295573735e-07,
                          "output_power": 8.605676717857032e-07,
                          "reference_active": true,
                          "delta_db": -0.5778944466897972
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5389181519813455
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 219784,
                    "uncovered_tail_samples": 73,
                    "metrics": {
                      "frame_count": 992,
                      "rms_floor": 7.683290217697679e-05,
                      "log_rms_mae_db": 0.3827752224316239,
                      "rms_envelope_nrmse": 0.05238641268737636,
                      "envelope_correlation": 0.9987000579493335,
                      "rms_gain_error_db": -0.3483448505699487,
                      "reference_mean_power": 0.000293126027399561,
                      "output_mean_power": 0.00027053278114407166
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 992,
                      "frequency_resolution_hz": 0.20207111436950148,
                      "power_floor": 2.0936583629794473e-11,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 9.709509593583371e-06,
                          "output_power": 8.615956583901039e-06,
                          "reference_active": true,
                          "delta_db": -0.5189379331232884
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 2.3881618624193633e-06,
                          "output_power": 2.094946044340822e-06,
                          "reference_active": true,
                          "delta_db": -0.5689091650220314
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 6.187797376434292e-08,
                          "output_power": 5.421297653064462e-08,
                          "reference_active": true,
                          "delta_db": -0.574328313860217
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.5540584706685122
                    }
                  }
                ]
              },
              {
                "name": "1000–3000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    1000.0,
                    3000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 219781,
                    "uncovered_tail_samples": 76,
                    "metrics": {
                      "frame_count": 998,
                      "rms_floor": 0.0001998844441775553,
                      "log_rms_mae_db": 0.40318813873460135,
                      "rms_envelope_nrmse": 0.04776962624584238,
                      "envelope_correlation": 0.9968773040021378,
                      "rms_gain_error_db": -0.008429923665205778,
                      "reference_mean_power": 0.0018404489532180932,
                      "output_mean_power": 0.0018368799932302286
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 998,
                      "frequency_resolution_hz": 0.20085625797048645,
                      "power_floor": 2.250633119630214e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.078858854616027e-05,
                          "output_power": 7.863316877319216e-05,
                          "reference_active": true,
                          "delta_db": -0.11744243363943335
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.76655855602681e-05,
                          "output_power": 1.641639807930249e-05,
                          "reference_active": true,
                          "delta_db": -0.31850162831484036
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 5.433613583459211e-06,
                          "output_power": 5.175958142850168e-06,
                          "reference_active": true,
                          "delta_db": -0.21097994314983232
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.21564133503470204
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 219784,
                    "uncovered_tail_samples": 73,
                    "metrics": {
                      "frame_count": 992,
                      "rms_floor": 0.0001632401576896862,
                      "log_rms_mae_db": 0.3723178565593941,
                      "rms_envelope_nrmse": 0.03803472995934368,
                      "envelope_correlation": 0.9979011008505009,
                      "rms_gain_error_db": -0.008360543775496188,
                      "reference_mean_power": 0.0018516105515452574,
                      "output_mean_power": 0.0018480494701474696
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 992,
                      "frequency_resolution_hz": 0.20207111436950148,
                      "power_floor": 1.9559392494828948e-10,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.369952373282e-05,
                          "output_power": 7.170452113166015e-05,
                          "reference_active": true,
                          "delta_db": -0.11918141557276717
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 8.018468234248022e-06,
                          "output_power": 7.399635898361609e-06,
                          "reference_active": true,
                          "delta_db": -0.3488106240670383
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 4.53400874682641e-07,
                          "output_power": 4.316176063904781e-07,
                          "reference_active": true,
                          "delta_db": -0.21383201657837092
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 0.22727468540605877
                    }
                  }
                ]
              },
              {
                "name": "3000–6000 Hz",
                "status": "measured",
                "filter": {
                  "kind": "butterworth-forward-backward",
                  "prototype_order": 4,
                  "edges_hz": [
                    3000.0,
                    6000.0
                  ],
                  "pad_type": "even",
                  "pad_samples": 27
                },
                "windows": [
                  {
                    "window_ms": 10.0,
                    "window_samples": 441,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 219781,
                    "uncovered_tail_samples": 76,
                    "metrics": {
                      "frame_count": 998,
                      "rms_floor": 1.2279001954331291e-05,
                      "log_rms_mae_db": 1.281867427119359,
                      "rms_envelope_nrmse": 0.20352101862144614,
                      "envelope_correlation": 0.9749703720460103,
                      "rms_gain_error_db": 1.2830646903062428,
                      "reference_mean_power": 2.8299423278076612e-05,
                      "output_mean_power": 3.802629858939388e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 998,
                      "frequency_resolution_hz": 0.20085625797048645,
                      "power_floor": 2.566036689733919e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 8.384962208800401e-07,
                          "output_power": 1.3685670544565182e-06,
                          "reference_active": true,
                          "delta_db": 2.127649714337081
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 1.4067821249168114e-07,
                          "output_power": 2.9509266478677905e-07,
                          "reference_active": true,
                          "delta_db": 3.2173157285961334
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 3.125215795609047e-08,
                          "output_power": 1.090341851977382e-07,
                          "reference_active": true,
                          "delta_db": 5.4268267180877725
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 3.5905973870069956
                    }
                  },
                  {
                    "window_ms": 40.0,
                    "window_samples": 1764,
                    "hop_samples": 220,
                    "hop_ms": 5.0,
                    "status": "measured",
                    "covered_start_sample": 0,
                    "covered_end_sample": 219784,
                    "uncovered_tail_samples": 73,
                    "metrics": {
                      "frame_count": 992,
                      "rms_floor": 1.0791287994309055e-05,
                      "log_rms_mae_db": 1.2971761957600036,
                      "rms_envelope_nrmse": 0.1875761540592229,
                      "envelope_correlation": 0.983677107642922,
                      "rms_gain_error_db": 1.2840017795232201,
                      "reference_mean_power": 2.8467952134764017e-05,
                      "output_mean_power": 3.826100777189619e-05
                    },
                    "modulation_bands": {
                      "status": "measured",
                      "envelope_rate_hz": 200.45454545454547,
                      "frames": 992,
                      "frequency_resolution_hz": 0.20207111436950148,
                      "power_floor": 2.242707104969893e-12,
                      "window": "hann",
                      "detrend": "constant",
                      "gain_normalization": "none",
                      "bands": [
                        {
                          "low_hz": 2.0,
                          "high_hz": 8.0,
                          "reference_power": 7.355378335790066e-07,
                          "output_power": 1.217645489868716e-06,
                          "reference_active": true,
                          "delta_db": 2.189158482062271
                        },
                        {
                          "low_hz": 8.0,
                          "high_hz": 20.0,
                          "reference_power": 5.623458795635236e-08,
                          "output_power": 1.1970654634931646e-07,
                          "reference_active": true,
                          "delta_db": 3.2811438384947893
                        },
                        {
                          "low_hz": 20.0,
                          "high_hz": 50.0,
                          "reference_power": 1.771713514776177e-09,
                          "output_power": 5.20906163663188e-09,
                          "reference_active": true,
                          "delta_db": 4.683659982676286
                        }
                      ],
                      "active_band_count": 3,
                      "active_band_mae_db": 3.384654101077782
                    }
                  }
                ]
              }
            ],
            "event": null,
            "silent_intervals": null,
            "limitations": [
              "有限長雑音の包絡差は独立seedでもゼロにならない",
              "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない",
              "RMS包絡は有声・無声や知覚品質の正解測定ではない"
            ]
          }
        }
      }
    }
  ],
  "calibration": [
    {
      "case": "smoke_arctic",
      "method": "identity",
      "status": "measured",
      "mos": {
        "status": "not_requested",
        "value": null
      }
    },
    {
      "case": "smoke_ready",
      "method": "identity",
      "status": "measured",
      "mos": {
        "status": "not_requested",
        "value": null
      }
    }
  ],
  "processes": [
    {
      "method": "world_harvest",
      "returncode": 0,
      "log": "logs/world_harvest-1791805065784490751.log"
    },
    {
      "method": "facodec",
      "returncode": 0,
      "log": "logs/facodec-1791805071512912437.log"
    },
    {
      "method": "hifiglot",
      "returncode": 0,
      "log": "logs/hifiglot-1791805087068164962.log"
    },
    {
      "method": "golf",
      "returncode": 0,
      "log": "logs/golf-1791805117161755446.log"
    },
    {
      "method": "sifigan",
      "returncode": 0,
      "log": "logs/sifigan-1791805121938877217.log"
    },
    {
      "method": "llsm2",
      "returncode": 0,
      "log": "logs/llsm2-1791805134637543766.log"
    },
    {
      "method": "magphase",
      "returncode": 0,
      "log": "logs/magphase-1791805137008521204.log"
    }
  ],
  "failure_count": 0,
  "process_failure_count": 0,
  "operation_evaluation": {
    "status": "not_implemented",
    "reason": "本PRは再合成比較の追加。属性操作・下流学習は未評価"
  }
}
