sindhuhegde/multivsr
收藏Hugging Face2025-04-10 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/sindhuhegde/multivsr
下载链接
链接失效反馈官方服务:
资源简介:
MultiVSR是一个大规模多语言唇读数据集,包含12,000小时视频,涵盖英语及12种非英语语言,共有约123K个YouTube视频的1.6M个视频片段。数据集具有丰富的说话人和语言多样性。
MultiVSR is a large-scale multilingual lip-reading dataset containing 12,000 hours of video footage, covering English and 12 non-English languages, with approximately 1.6M video clips across 123K YouTube videos. The dataset exhibits a high diversity in terms of speakers and languages.
提供机构:
sindhuhegde



