The Research And Implementation Of Chao Shan Dialect Text To Speech System

Today , information technology highly develops, the information forms are many and varied, the humanity's ability to control information under computer's help is also in enhancement. Under this kind of situation, The humanity longs for the using of pronunciation to communicate with computer, realizing intelligent communication between person and the computer, maximum limit displays the computer's formidable operation ability to serve the human society. Therefore, computer speech synthesizing becomes an important research topic of the modern computer science. At present, text to speech conversion (TTS for short) is the main research stage of computer speech synthesizing, It processes the written text and convert into fluently , intelligible speech waveform. At present TTS has been applied in different occasions, such as reading the homepage and e-mails, collating the manuscripts, man-machine conversation, information query, etc.In this paper, a Chao Shan Dialect text to speech system using PSOLA method is presented. The paper has analyzed the difference between tonic language and intonation language, and has made improvement to PSOLA according to the tonic language characteristic. According to the characteristic of Chao Shan Dialect, the system uses speech wavform concatenation method. The author also make beneficial attempt in the research about text analyzing and the result is comparatively satisfying. Chao Shan Dialect is one of the eight major Chinese dialects. It's the oldest and most special dialect. Many people are using it, so our system will surely improve the Chao Shan Dialect information processing.