纳西语语音合成的初步研究

A Preliminary Study of Nakhi Speech Synthesis

本文初步建立了一个纳西语按规则语音合成系统该系统采用共振峰语音合成技术，以３５０个左右的无调单音节和４种声调模式为合成基元．该系统不仅适用于纳西语文一语转换系统，而且也是一个研究纳西语语音的有力工具．

In this paper,a preliminary system of Naxi speech synthesis-by-rule wasdeveloped.In the system,a new parallel formant synthesizer was be made and about 350syllables and 4 tones were selected as the synthesis units.The system is applicable not onlyin text-to-speech conversion for Naxi,but in studying phonetic feature in Naxi.