{
 "id": "nvidia/preference-asr-bench",
 "owner": "nvidia",
 "name": "preference-asr-bench",
 "downloads": 137,
 "likes": 7,
 "last_modified": "2026-09-21T21:07:52.000Z",
 "stated_licence": "cc-by-nc-4.0",
 "stated_from": "tag",
 "licence_name": null,
 "licence_link": null,
 "file_licence": null,
 "licence_files": [],
 "sources": [],
 "gated": false,
 "tasks": [
  "automatic-speech-recognition"
 ],
 "citation": "This dataset is a **mixed-license** collection. The annotations (`instruction`\nand `preference_text`) are released under **CC BY-NC 4.0**. The audio and the\noriginal transcripts retain their per-corpus licenses — this repo does **not**\nrelicense them:\n\n| Corpus | License | Audio redistribution | Attribution |\n|---|---|---|---|\n| LibriSpeech | CC BY 4.0 (reader 1259 / Elizabeth Kletts excluded) | ✅ | Panayotov et al., 2015 |\n| AMI | CC BY 4.0 | ✅ | Carletta et al., 2005 |\n| Common Voice | CC0 1.0 (transcripts from HuggingFace / ESB, not the Mozilla Data Collective platform) | ❌ text-only | Ardila et al., 2020 |\n| VoxPopuli | CC0 (European Parliament) | ✅ (keep EP source notice) | Wang et al., 2021 |",
 "signal": "restrictive",
 "reason": "The declared licence (cc-by-nc-4.0) is classified as having use restrictions.",
 "revision": "8eed54122f1da067b586c435dd1151cbb9ae569d",
 "stated_evidence": "https://huggingface.co/datasets/nvidia/preference-asr-bench/blob/8eed54122f1da067b586c435dd1151cbb9ae569d/README.md",
 "file_evidence": "https://huggingface.co/datasets/nvidia/preference-asr-bench/tree/8eed54122f1da067b586c435dd1151cbb9ae569d",
 "card_nc": false
}