The Influence of Context on Tibetan Lhasa Speech Synthesis
This paper studies the influence factor on HMM-based Tibetan Lhasa speech synthesis.In order to find the key factor which makes the most contribution to improve the synthesized Tibetan Lhasa speech,we synthesize Tibetan Lhasa speech by different context labeling and different number of training sentences with different speech synthesis unit,respectively.We build two Tibetan Lhasa speech corpora which respectively containing 800 sentences and 2000 sentences.The context labeling was built by manual work.One of the speech synthesis units is the initial and the final,another is syllable.The result shows that no matter what speech synthesis unit,the quality of synthesized speech by detailed contexts with 700 training sentences is higher than by rough contexts with 1600 training sentences.It means that in order to improve the quality of synthesized speech,the better way is to build a more detail context labeling than record more sentence.
HMM-based Tibetan Lhasa speech synthesis context number of training sentences speech synthesis unit
Shipeng Xu Hongzhi Yu Guanyu Li
Northwest University for Nationalities Key Laboratory of National language intelligent processing,Gansu Province Lanzhou,China
国际会议
重庆
英文
625-629
2017-03-25（万方平台首次上网日期，不代表论文的发表时间）