{
 "id": "mythicinfinity/libritts_r",
 "owner": "mythicinfinity",
 "name": "libritts_r",
 "downloads": 15164,
 "likes": 48,
 "last_modified": "2024-02-09T21:20:19.000Z",
 "stated_licence": "cc-by-4.0",
 "stated_from": "tag",
 "licence_name": null,
 "licence_link": null,
 "file_licence": null,
 "licence_files": [],
 "sources": [],
 "gated": false,
 "tasks": [
  "text-to-speech"
 ],
 "citation": "@ARTICLE{Koizumi2023-hs,\n  title         = \"{LibriTTS-R}: A restored multi-speaker text-to-speech corpus\",\n  author        = \"Koizumi, Yuma and Zen, Heiga and Karita, Shigeki and Ding,\n                   Yifan and Yatabe, Kohei and Morioka, Nobuyuki and Bacchiani,\n                   Michiel and Zhang, Yu and Han, Wei and Bapna, Ankur\",\n  abstract      = \"This paper introduces a new speech dataset called\n                   ``LibriTTS-R'' designed for text-to-speech (TTS) use. It is\n                   derived by applying speech restoration to the LibriTTS\n                   corpus, which consists of 585 hours of speech data at 24 kHz\n                   sampling rate from 2,456 speakers and the corresponding\n                   texts. The constituent samples of LibriTTS-R are identical\n                   to those of LibriTTS, with only the sound quality improved.\n                   Experimental results show that the LibriTTS-R ground-truth\n                   samples showed significantly improved sound quality compared\n                   to those in LibriTTS. In addition, neural end-to-end TTS\n                   trained with LibriTTS-R achieved speech naturalness on par\n                   with that of the ground-truth samples. The corpus is freely\n                   available for download from\n                   \\textbackslashurl\\{http://www.openslr.org/141/\\}.\",\n  month         =  may,\n  year          =  2023,\n  copyright     = \"http://creativecommons.org/licenses/by-nc-nd/4.0/\",",
 "signal": "permissive",
 "reason": "The declared licence (cc-by-4.0) is classified as permissive."
}