{
  "schema_version": 2,
  "benchmark": "Speakmac vs Wispr Flow 27-clip licensed real-speech benchmark corpus",
  "benchmark_date": "2026-07-28",
  "corpus_version": "2026-07-28-real-speech-v1",
  "selection": {
    "svarah": "18 deterministic spread offsets; first eligible 3.0-10.5 second, 7-22 word row",
    "librispeech": "9 deterministic offsets; first eligible 7-22 word row from a new speaker"
  },
  "normalization": {
    "change_notice": "Source audio was resampled and loudness-normalized for this benchmark.",
    "ffmpeg_filter": "loudnorm=I=-20:TP=-2:LRA=7",
    "sample_rate_hz": 16000,
    "channels": 1,
    "codec": "pcm_s16le"
  },
  "sources": {
    "svarah": {
      "title": "Svarah: Evaluating English ASR Systems on Indian Accents",
      "creators": [
        "Tahir Javed",
        "Sakshi Joshi",
        "Vignesh Nagarajan",
        "Sai Sundaresan",
        "Janki Nawale",
        "Abhigyan Raman",
        "Kaushal Santosh Bhogale",
        "Pratyush Kumar",
        "Mitesh M. Khapra"
      ],
      "official_url": "https://huggingface.co/datasets/ai4bharat/Svarah",
      "acquisition_url": "https://huggingface.co/datasets/amrithanandini/svarah",
      "license": "CC BY 4.0",
      "license_url": "https://creativecommons.org/licenses/by/4.0/",
      "attribution_note": "18 normalized derivatives from the Svarah test material. No endorsement is implied.",
      "provenance_limit": "The historical corpus acquired these fixed rows through the listed public mirror. The official repository license was rechecked on 2026-08-31; official row-level access remained gated, so mirror-to-official bit identity was not independently rechecked."
    },
    "librispeech": {
      "title": "LibriSpeech: an ASR corpus based on public domain audio books",
      "creators": [
        "Vassil Panayotov",
        "Guoguo Chen",
        "Daniel Povey",
        "Sanjeev Khudanpur"
      ],
      "official_url": "https://www.openslr.org/12",
      "acquisition_url": "https://huggingface.co/datasets/openslr/librispeech_asr",
      "license": "CC BY 4.0",
      "license_url": "https://creativecommons.org/licenses/by/4.0/",
      "attribution_note": "9 normalized derivatives from LibriSpeech test-clean. No endorsement is implied."
    }
  },
  "samples": [
    {
      "id": "svarah-01-row-0",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 5.927,
      "reference_words": 14,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 0,
        "selection_offset": 0
      },
      "audio": {
        "path": "audio/svarah-01-row-0.wav",
        "sha256": "8b6eaf9f1deb75dd260e752851f461c306e4c7a3b4662a9c9b4db5b9b33ee24e"
      },
      "reference": {
        "path": "reference/svarah-01-row-0.txt",
        "sha256": "173facc1baee3535e8f68158859b1afa39d2d6e9c02dd9a07118a961c102aef3"
      }
    },
    {
      "id": "svarah-02-row-76",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 7.04,
      "reference_words": 11,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 76,
        "selection_offset": 74
      },
      "audio": {
        "path": "audio/svarah-02-row-76.wav",
        "sha256": "d6db880b612737f226d503913ec5ee22b4f4b2b02564f4e19856a878bd91a95d"
      },
      "reference": {
        "path": "reference/svarah-02-row-76.txt",
        "sha256": "704a9ddf46ba12544ed417d9a57cd4f5d59f07145f96ba69dec7bad66b2f8ecb"
      }
    },
    {
      "id": "svarah-03-row-148",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 6.267,
      "reference_words": 8,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 148,
        "selection_offset": 148
      },
      "audio": {
        "path": "audio/svarah-03-row-148.wav",
        "sha256": "4b75bf3a1388d36636e0f6885b1c8979e6e57d7a2ef8cbf43ec782b18ff3add5"
      },
      "reference": {
        "path": "reference/svarah-03-row-148.txt",
        "sha256": "cf31b5d6dac095897b60dfbc118644c2516c03e035211cbfe30102ac9b559534"
      }
    },
    {
      "id": "svarah-04-row-225",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 6.08,
      "reference_words": 13,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 225,
        "selection_offset": 222
      },
      "audio": {
        "path": "audio/svarah-04-row-225.wav",
        "sha256": "9a22bc772cbc906bd0853188119f055ed340d7911e5f7b58e4fe2aa1bdbc32c1"
      },
      "reference": {
        "path": "reference/svarah-04-row-225.txt",
        "sha256": "66b3877b385df7d7a0d71b9da3152c29db3c079e34b47324bf006aceb96cb59f"
      }
    },
    {
      "id": "svarah-05-row-298",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 7.931,
      "reference_words": 12,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 298,
        "selection_offset": 296
      },
      "audio": {
        "path": "audio/svarah-05-row-298.wav",
        "sha256": "36b96206e819577975ccbe23d98e5b6d4430b686f1700631ed8a143ea753689c"
      },
      "reference": {
        "path": "reference/svarah-05-row-298.txt",
        "sha256": "bc6758f76ee1a8701c8e07906d08f79161220c03c891ba010b00e9860f1a25ed"
      }
    },
    {
      "id": "svarah-06-row-374",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 7.226,
      "reference_words": 18,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 374,
        "selection_offset": 370
      },
      "audio": {
        "path": "audio/svarah-06-row-374.wav",
        "sha256": "dcc426ae908901bfc60354294f13be529632f03e5ce4175ded6263e8c3711acd"
      },
      "reference": {
        "path": "reference/svarah-06-row-374.txt",
        "sha256": "ac319b9df29dfa23c8ef7f6f131b76d26af697b20f9e444810a809ad06969d49"
      }
    },
    {
      "id": "svarah-07-row-445",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 8.425,
      "reference_words": 17,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 445,
        "selection_offset": 444
      },
      "audio": {
        "path": "audio/svarah-07-row-445.wav",
        "sha256": "c5c7244832503d710d5441de01337559698d4acd84f156cf8a865430d3fe50e8"
      },
      "reference": {
        "path": "reference/svarah-07-row-445.txt",
        "sha256": "b22760ee3653e48ee645d77d3537b9909150d3435d6e2ee5344236ab7e1380fd"
      }
    },
    {
      "id": "svarah-08-row-521",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 3.406,
      "reference_words": 7,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 521,
        "selection_offset": 518
      },
      "audio": {
        "path": "audio/svarah-08-row-521.wav",
        "sha256": "c14fa18b09e054e3ed31f3fa4121cb51ee3526021ecac2a5544b491413cb431e"
      },
      "reference": {
        "path": "reference/svarah-08-row-521.txt",
        "sha256": "398ac80b9297aca20ccd1f636ef5f704327d95d61a4418d22db8aad51571a3ce"
      }
    },
    {
      "id": "svarah-09-row-592",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 3.098,
      "reference_words": 7,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 592,
        "selection_offset": 592
      },
      "audio": {
        "path": "audio/svarah-09-row-592.wav",
        "sha256": "f51689cde8f81da2cfb2c91e2f1765ac52842a5f78b9c422d36f20636f1106b6"
      },
      "reference": {
        "path": "reference/svarah-09-row-592.txt",
        "sha256": "103c20f0ffedd7b104aad3b0a7c029ba121b4bcae17bb7b7ee6d6cb6cffc02a2"
      }
    },
    {
      "id": "svarah-10-row-666",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 4.0,
      "reference_words": 10,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 666,
        "selection_offset": 666
      },
      "audio": {
        "path": "audio/svarah-10-row-666.wav",
        "sha256": "d1d04e76b83565606ea26802f56a1b772a23900208164e64e24ba4f9ae4a51f4"
      },
      "reference": {
        "path": "reference/svarah-10-row-666.txt",
        "sha256": "0df3ea8cc01734f3c1c8b6a6be10ab66611a68437a8136fc663da5b3ad7d4308"
      }
    },
    {
      "id": "svarah-11-row-742",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 3.16,
      "reference_words": 10,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 742,
        "selection_offset": 740
      },
      "audio": {
        "path": "audio/svarah-11-row-742.wav",
        "sha256": "49fd30abe7028ca276e4e57122afeec6db7f6ab89f52d9a9a8606c2ee211e33f"
      },
      "reference": {
        "path": "reference/svarah-11-row-742.txt",
        "sha256": "f84b55189dbcb1f08d9833f8c015b1cbd5987f2ffb20419238dc22e10f044568"
      }
    },
    {
      "id": "svarah-12-row-815",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 9.983,
      "reference_words": 18,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 815,
        "selection_offset": 814
      },
      "audio": {
        "path": "audio/svarah-12-row-815.wav",
        "sha256": "3e9d7396a97d12ee93874d85370dcef4a48e155c0475e46a00845e5d0cc3d9f8"
      },
      "reference": {
        "path": "reference/svarah-12-row-815.txt",
        "sha256": "4589ecdcd7508ae8e363aa51524ef214305ae465831c0ff819f3ba7ee3e69b33"
      }
    },
    {
      "id": "svarah-13-row-888",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 7.867,
      "reference_words": 14,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 888,
        "selection_offset": 888
      },
      "audio": {
        "path": "audio/svarah-13-row-888.wav",
        "sha256": "87af3704f1f26c0ae5b468d3fcdc8430a3450f3211c02c2e1e3e4c07c0d04fb9"
      },
      "reference": {
        "path": "reference/svarah-13-row-888.txt",
        "sha256": "992da8cce17d9b612954c296ded61d9bf50f01f2e2c43f69b974dd7f55db6962"
      }
    },
    {
      "id": "svarah-14-row-962",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 4.639,
      "reference_words": 10,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 962,
        "selection_offset": 962
      },
      "audio": {
        "path": "audio/svarah-14-row-962.wav",
        "sha256": "919462fbbf9f3a24c17d46a53200e39c1b9d72c61a4001f7cedb8cb1787cf287"
      },
      "reference": {
        "path": "reference/svarah-14-row-962.txt",
        "sha256": "2b5bbf244b3672a4c4d4b86ffab4a8d9e337c5868c3983a20fea8c501a0a8ecc"
      }
    },
    {
      "id": "svarah-15-row-1040",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 5.337,
      "reference_words": 15,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 1040,
        "selection_offset": 1036
      },
      "audio": {
        "path": "audio/svarah-15-row-1040.wav",
        "sha256": "0e303ebfb69cc99deecc1257ea4d6d65f62f589bc536b9f8fc91ea00542c54aa"
      },
      "reference": {
        "path": "reference/svarah-15-row-1040.txt",
        "sha256": "2e74c399f1321488567dafc0c806c28b9a2f9d32070ddc11c72a28fcb01996ed"
      }
    },
    {
      "id": "svarah-16-row-1110",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 3.774,
      "reference_words": 9,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 1110,
        "selection_offset": 1110
      },
      "audio": {
        "path": "audio/svarah-16-row-1110.wav",
        "sha256": "f7ba141dd8f232691a419a953504d3f3e5960e17dc56653c8384f2d926c499b4"
      },
      "reference": {
        "path": "reference/svarah-16-row-1110.txt",
        "sha256": "1ff06079545921366ccca3cfbc5c3dcb1a652b35933d8746f42e12ccc91d84d4"
      }
    },
    {
      "id": "svarah-17-row-1185",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 10.134,
      "reference_words": 20,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 1185,
        "selection_offset": 1184
      },
      "audio": {
        "path": "audio/svarah-17-row-1185.wav",
        "sha256": "94ee7a060ae0090c6767cb8daac21fb6b3c34f41885eb45eb269d6f7c779a6de"
      },
      "reference": {
        "path": "reference/svarah-17-row-1185.txt",
        "sha256": "e9ce5a820fb2a290b6a7b5884f7ae269feacb78d9165c8ebfc27993117206a63"
      }
    },
    {
      "id": "svarah-18-row-1271",
      "dataset": "Svarah",
      "subset": "test",
      "speech_kind": "real_human_indian_english",
      "category": null,
      "duration_seconds": 6.462,
      "reference_words": 11,
      "source_id": "svarah",
      "source_item": {
        "retrieval_split": "test",
        "retrieval_row_index": 1271,
        "selection_offset": 1258
      },
      "audio": {
        "path": "audio/svarah-18-row-1271.wav",
        "sha256": "7cf981724c1c80a217d47425729270e93b0c4dc16dd5ce9e360635417d82e2ff"
      },
      "reference": {
        "path": "reference/svarah-18-row-1271.txt",
        "sha256": "88d2da4ebd692e4084f1e443af94075d8171d41b0c5dc0252eb1be0a077c60e3"
      }
    },
    {
      "id": "libri-01-6930-75918-0000",
      "dataset": "LibriSpeech",
      "subset": "test-clean",
      "speech_kind": "real_human_read_english",
      "category": null,
      "duration_seconds": 3.505,
      "reference_words": 8,
      "source_id": "librispeech",
      "source_item": {
        "subset": "test-clean",
        "utterance_id": "6930-75918-0000",
        "selection_offset": 0
      },
      "audio": {
        "path": "audio/libri-01-6930-75918-0000.wav",
        "sha256": "f8911490351969b8f60260a462b2ef9d4f85231d1bbd404bf33267bfb743ce76"
      },
      "reference": {
        "path": "reference/libri-01-6930-75918-0000.txt",
        "sha256": "f0d15ba54e0099c3dcb9f74a37fcb00175badbc257e4b314baedcbfa2192b7ba"
      }
    },
    {
      "id": "libri-02-7729-102255-0039",
      "dataset": "LibriSpeech",
      "subset": "test-clean",
      "speech_kind": "real_human_read_english",
      "category": null,
      "duration_seconds": 6.815,
      "reference_words": 18,
      "source_id": "librispeech",
      "source_item": {
        "subset": "test-clean",
        "utterance_id": "7729-102255-0039",
        "selection_offset": 300
      },
      "audio": {
        "path": "audio/libri-02-7729-102255-0039.wav",
        "sha256": "45f308beabcf33b7f46406656ce65b2f2be6ac10c480a8aae6a1a57599fd3d39"
      },
      "reference": {
        "path": "reference/libri-02-7729-102255-0039.txt",
        "sha256": "3a2c1b09b6022afdf68a79cb88fec09fe548df1504767593c3b035cd156030c0"
      }
    },
    {
      "id": "libri-03-8463-294828-0023",
      "dataset": "LibriSpeech",
      "subset": "test-clean",
      "speech_kind": "real_human_read_english",
      "category": null,
      "duration_seconds": 4.745,
      "reference_words": 13,
      "source_id": "librispeech",
      "source_item": {
        "subset": "test-clean",
        "utterance_id": "8463-294828-0023",
        "selection_offset": 600
      },
      "audio": {
        "path": "audio/libri-03-8463-294828-0023.wav",
        "sha256": "e24aee48f1e8deb1fba761218cc790e6a00381d9b0ea85f142f012be46d4886d"
      },
      "reference": {
        "path": "reference/libri-03-8463-294828-0023.txt",
        "sha256": "c9980767e5c3d38580390f22cd230b7b02b8ab3ccce497fb638da3d0ee4d2c99"
      }
    },
    {
      "id": "libri-04-5142-33396-0002",
      "dataset": "LibriSpeech",
      "subset": "test-clean",
      "speech_kind": "real_human_read_english",
      "category": null,
      "duration_seconds": 3.67,
      "reference_words": 12,
      "source_id": "librispeech",
      "source_item": {
        "subset": "test-clean",
        "utterance_id": "5142-33396-0002",
        "selection_offset": 900
      },
      "audio": {
        "path": "audio/libri-04-5142-33396-0002.wav",
        "sha256": "eb837b6e1f0dd7680cdf28f595987d1b5eaf0be991c81380418e3b2f0eb79d12"
      },
      "reference": {
        "path": "reference/libri-04-5142-33396-0002.txt",
        "sha256": "f2f40b45ae5bedfe2fa96fef4fa1aa73248f269adbeebe2e8b3f3459815b87fa"
      }
    },
    {
      "id": "libri-05-4446-2275-0005",
      "dataset": "LibriSpeech",
      "subset": "test-clean",
      "speech_kind": "real_human_read_english",
      "category": null,
      "duration_seconds": 4.445,
      "reference_words": 19,
      "source_id": "librispeech",
      "source_item": {
        "subset": "test-clean",
        "utterance_id": "4446-2275-0005",
        "selection_offset": 1200
      },
      "audio": {
        "path": "audio/libri-05-4446-2275-0005.wav",
        "sha256": "8bdb27fafb31775dc338f2d74d8a3d08577df32d5965d2b94785bf3499048d62"
      },
      "reference": {
        "path": "reference/libri-05-4446-2275-0005.txt",
        "sha256": "626999ede865ad756b450d4ff4901206777dbea9702f15d900b9a3cae6421b44"
      }
    },
    {
      "id": "libri-06-4507-16021-0033",
      "dataset": "LibriSpeech",
      "subset": "test-clean",
      "speech_kind": "real_human_read_english",
      "category": null,
      "duration_seconds": 5.545,
      "reference_words": 15,
      "source_id": "librispeech",
      "source_item": {
        "subset": "test-clean",
        "utterance_id": "4507-16021-0033",
        "selection_offset": 1500
      },
      "audio": {
        "path": "audio/libri-06-4507-16021-0033.wav",
        "sha256": "8f6b3aaaeaf2ed58a9af8f5851685da05269bb1a14b4e7828217fc1d939710dd"
      },
      "reference": {
        "path": "reference/libri-06-4507-16021-0033.txt",
        "sha256": "18f0af4ae5c5e49da5bc493a307b58434b99c0dbc6b324072361573d4e8578ac"
      }
    },
    {
      "id": "libri-07-1188-133604-0029",
      "dataset": "LibriSpeech",
      "subset": "test-clean",
      "speech_kind": "real_human_read_english",
      "category": null,
      "duration_seconds": 3.705,
      "reference_words": 9,
      "source_id": "librispeech",
      "source_item": {
        "subset": "test-clean",
        "utterance_id": "1188-133604-0029",
        "selection_offset": 1800
      },
      "audio": {
        "path": "audio/libri-07-1188-133604-0029.wav",
        "sha256": "24838058d3936360f7b19f3e0a325910307fc7b64b3c16ec2f82fa399c01dc6c"
      },
      "reference": {
        "path": "reference/libri-07-1188-133604-0029.txt",
        "sha256": "9f6cea59b9d8540c140f22d3e304c4f18fd150b0d5de5f800e71e84f232efb2f"
      }
    },
    {
      "id": "libri-08-4970-29093-0007",
      "dataset": "LibriSpeech",
      "subset": "test-clean",
      "speech_kind": "real_human_read_english",
      "category": null,
      "duration_seconds": 6.995,
      "reference_words": 18,
      "source_id": "librispeech",
      "source_item": {
        "subset": "test-clean",
        "utterance_id": "4970-29093-0007",
        "selection_offset": 2100
      },
      "audio": {
        "path": "audio/libri-08-4970-29093-0007.wav",
        "sha256": "c5665b0560b8497006085fdd5a38adc9d04c5b2022057179548a5a35b615fe14"
      },
      "reference": {
        "path": "reference/libri-08-4970-29093-0007.txt",
        "sha256": "bce28008cdf11460657d8e6a4f49d5bf897a3170fddaa1a58b5d11d1b4856429"
      }
    },
    {
      "id": "libri-09-237-134493-0012",
      "dataset": "LibriSpeech",
      "subset": "test-clean",
      "speech_kind": "real_human_read_english",
      "category": null,
      "duration_seconds": 3.66,
      "reference_words": 13,
      "source_id": "librispeech",
      "source_item": {
        "subset": "test-clean",
        "utterance_id": "237-134493-0012",
        "selection_offset": 2400
      },
      "audio": {
        "path": "audio/libri-09-237-134493-0012.wav",
        "sha256": "82c7bda534a25ddf8433bee6b653578c29fbba4f88192535b9521fc4b336ebdc"
      },
      "reference": {
        "path": "reference/libri-09-237-134493-0012.txt",
        "sha256": "a4f679b87b825b32e4a69f5dc27c62b92421efd7f542935f0014081e1523f1fa"
      }
    }
  ],
  "summary": {
    "samples": 27,
    "reference_words": 349,
    "duration_seconds": 153.841,
    "real_human_samples": 27,
    "unique_audio_hashes": 27,
    "excluded_controlled_samples": 10,
    "exclusion_reason": "Apple System Voice output is not licensed for public/commercial redistribution."
  }
}
