{
  "dataset": "google/fleurs",
  "revision": "70bb2e84b976b7e960aa89f1c648e09c59f894dd",
  "config": "cmn_hans_cn",
  "split": "test",
  "source": "https://huggingface.co/datasets/google/fleurs/resolve/70bb2e84b976b7e960aa89f1c648e09c59f894dd/data/cmn_hans_cn/audio/test.tar.gz",
  "referenceSource": "https://huggingface.co/datasets/google/fleurs/resolve/70bb2e84b976b7e960aa89f1c648e09c59f894dd/data/cmn_hans_cn/test.tsv",
  "license": "CC-BY-4.0",
  "durationSeconds": 126.94000000000001,
  "selection": "First archive entries with matching test.tsv reference until at least 120 seconds; concatenated without gaps, encoded to 64 kbit/s MP3.",
  "clips": [
    {
      "filename": "10026684690566417990.wav",
      "id": "1906",
      "reference": "“这并不是告别。这是一个篇章的结束，也是新篇章的开始。”",
      "startSeconds": 0,
      "durationSeconds": 10.38,
      "sha256": "a72c0b59fdba80850552af7991de07f9f4940846d2d70f0890eeb153c40f164b"
    },
    {
      "filename": "10040380210557600780.wav",
      "id": "2006",
      "reference": "钙、钾等元素属于金属，银和金等元素当然也是金属。",
      "startSeconds": 10.38,
      "durationSeconds": 8.28,
      "sha256": "b39d2c84aefada280b73a9ab439726d370d45e5a75bd05f0a90153288ea4561b"
    },
    {
      "filename": "10048525650290665384.wav",
      "id": "1883",
      "reference": "桥下垂直净空 15 米。该项目于 2011 年 8 月完工，但直到 2017 年 3 月才开始通车。",
      "startSeconds": 18.66,
      "durationSeconds": 13.86,
      "sha256": "aeb95366f4907daf8797d66797c87f6e002f10704e78e87a0c4b4521756a1616"
    },
    {
      "filename": "10053956375630517392.wav",
      "id": "1852",
      "reference": "适当使用博客，可以使学生变得更善于分析和进行思辨；通过积极回应网络材料，学生们可以在他人文章的上下文语境中找到自己的立场，并能够针对特定问题提出自己的观点（Oravec, 2002）。\"",
      "startSeconds": 32.52,
      "durationSeconds": 15.42,
      "sha256": "d1274fd5e5ce0029473faa0df42bd714094428b853029d6f6c64ffd3f36a8afe"
    },
    {
      "filename": "10056338555786085046.wav",
      "id": "1734",
      "reference": "科学家们可以得出结论，暗物质对其他暗物质的影响方式与普通物质相同。",
      "startSeconds": 47.94,
      "durationSeconds": 12.9,
      "sha256": "7cc0e10d7c4ad3297d1803ac968c996a747117edd75a6cec59e485a93bbe0fe9"
    },
    {
      "filename": "10073099671796350432.wav",
      "id": "1890",
      "reference": "大多数现代科研望远镜都是巨型设施，位于大气条件优良的偏远地区。",
      "startSeconds": 60.84,
      "durationSeconds": 13.74,
      "sha256": "e07c342ae53173bec1c3d5f534073fa6368713f76e7b5ab9497b015b1020167f"
    },
    {
      "filename": "10081186303620187437.wav",
      "id": "1721",
      "reference": "1963 年大坝建成后，季节性洪水被控制住了，沉积物不再冲散到河流里。",
      "startSeconds": 74.58,
      "durationSeconds": 15.18,
      "sha256": "379999ca5e729d35f81b0d62087e78f23481e0eb47bf8aca2c66208e90a072ce"
    },
    {
      "filename": "10104380129516909856.wav",
      "id": "1869",
      "reference": "它的长下颚上布满了 70 多颗剃刀般锋利的牙齿，上颚上还有一排。这意味着任何与它相遇的东西都无路可逃。",
      "startSeconds": 89.76,
      "durationSeconds": 15.46,
      "sha256": "cd509460412479ebc989b962938881b47aad636eca388e890e3fd219e999b60d"
    },
    {
      "filename": "10112798408996336578.wav",
      "id": "1805",
      "reference": "Scotturb 403 路公共汽车定期发车前往辛特拉 (Sintra)，在罗卡角停靠。",
      "startSeconds": 105.22,
      "durationSeconds": 7.62,
      "sha256": "714a1785c636060d6a8f284abcf786c876f70b07e0a8ea966c476abf1e890fe7"
    },
    {
      "filename": "10134195680525011353.wav",
      "id": "1965",
      "reference": "这里几乎都是沙滩，游泳很安全，大部分地方都有新西兰圣诞树的树荫。",
      "startSeconds": 112.84,
      "durationSeconds": 6.84,
      "sha256": "52cd9b630b984ed486d69f8ee8dd841a57f875c752e1af411e0a39dabd3e00cb"
    },
    {
      "filename": "10142207404499844309.wav",
      "id": "1903",
      "reference": "当所有可用的资源在一个组织的职能部门之间得到有效利用时，创造力和独创性就会发生。",
      "startSeconds": 119.68,
      "durationSeconds": 7.26,
      "sha256": "cd570cba752dd4214a9c6ec050759172932e4bc4df18ed2a2bb0d8a9e30172db"
    }
  ],
  "code": "zh",
  "status": "completed",
  "processedAt": "2026-10-12T13:18:35.446Z",
  "provider": "AssemblyAI",
  "settings": {
    "language_code": "zh",
    "language_detection": false,
    "punctuate": true,
    "speech_models": [
      "universal-3-5-pro",
      "universal-2"
    ],
    "format_text": true,
    "speaker_labels": true
  },
  "scope": "Engine-only test using FastScribe normalization; concatenated human read speech, not a browser workflow or a conversational benchmark.",
  "attribution": "FLEURS: Few-shot Learning Evaluation of Universal Representations of Speech. Alexis Conneau et al., 2022. https://arxiv.org/abs/2205.12446",
  "licenseUrl": "https://creativecommons.org/licenses/by/4.0/",
  "files": {
    "chinese.mp3": "a52eb07095d2a182b348052c498ade647460f024c07c3af6dc4ff348b6ba4c5f",
    "chinese-reference.txt": "1fc2a133ab9be53e04f977f22e1d87e597756908033c1a51384c1634d6a4c2cf",
    "chinese-transcript.txt": "dc92ee1dd2b4789f09f611660aafae26871590941cc03c3877ddbae01a30a13a",
    "chinese-subtitles.srt": "2a6fef7115c47109ba41b54a08e24a5c2fbf5efeda3b7da867e8ebbdd8211285",
    "chinese.vtt": "d9ba4f1977ac9aae3829fa08c331023c8f12c952000fd441c80bfba000327476",
    "chinese-result.json": "f2700b92c2d893b42334718af78f02a41af632a53674e2d0d2336ebc1f7fa71a"
  }
}
