【摘要】 本文提出了一种汉语语音合成的新方法。利用小波变换检测语音信号的声门闭合时刻 (GCI) ,利用语音信号的 GCI进行基音同步和样本选择的多脉冲线性预测分析 ,将得到的语音参数保存到语音库中。通过改变相应的语音参数可以灵活地调节音节的时长、基音频率和音强。本方法得到的语音参数比传统方法得到的参数更加精确 ,合成语音清晰、自然、并且大大降低了语音库的存储量 ,非常适合于小型微型计算机系统。更多还原

【Abstract】 This paper introduced a new method to synthesis Chinese speech. Using Wavelet Transform to detect the Glottal Closure Instant (GCI), proposed pitch-synchronization and sample selection LP analysis based on the GCI information, the obtained parameters would be saved in the speech database. The duration and fundamental frequency can be changed by adjusted the speech parameters. The speech parameters obtained by proposed method are more accuracy than the traditional method does, and the synthesized speech is clear and natural. on the other hand, the storage of the speech database is highly decreased, so the proposed method is suitable for the mini-micro computer system.更多还原