Design and Building of a Speech Synthesis Database for Lhasa Tibetan
To meet the need of Tibetan speech synthesis, this paper has designed rules of corpus selection and optimized the selected corpus. It puts forward a prosodic annotation system suitable for Lhasa Tibetan, and has successfully built a speech synthesis corpus of Lhasa Tibetan based on the method of Chinese speech corpus construction but with Tibetan features. The corpus contains a text corpus, a speech corpus and an annotation corpus. Such a corpus is highly comprehensive and scientific, includes almost all the segmental and suprasegmentai information of Lhasa Tibetan, and will greatly promote the development of Tibetan prosody research and its speech synthesis system.
Lhasa Tibetan speech synthesis optimization prosodic annotation corpus
Jing Shi Qi Chen Yonghong Li Hongzhi Yu
Key Lab of Chinas National Linguistic Information Technology, Northwest University for Nationalities, Lanzhou, 730030, China
国际会议
2010 International Conference on Image Analysis and Signal Processing(2010 图像分析与信号处理国际会议 IASP 10)
厦门
英文
398-403
2010-04-12(万方平台首次上网日期,不代表论文的发表时间)