openpecha/tibetan-voice-benchmark

这是一个藏语语音识别数据集，包含有声书、儿童语音、历史、藏语电影、自然语音、新闻、播客和藏语教学等类别的语音数据。所有转录文本都经过至少一个人的额外审核。数据集还包括转录质量、字符长度和音频长度的相关信息。

This is a Tibetan Speech-To-Text dataset containing voice data from categories such as audio books, childrens speech, history, Tibetan movies, natural speech, news, podcasts, and Tibetan teachings. All transcription texts have been reviewed by at least one additional person. The dataset also includes information related to transcription quality, character length, and audio length.