语音生成的逆向问题及其在语音合成中的应用

A new method to synthesize formant targeted sounds based on speech production model and RTLA articulatory synthesis model is studied. A method of using codebook and dynamic interpolation constraint to solve the inverse speech problem. The synthesis model is implemented with scattering process derived from reflection-type line analog of vocal tract system according to the acoustic model of speech production. The vocal-tract area function which controls the synthesis model is derived from the first three formant trajectories using the inverse solution of speech production. The proposed method not only gives good naturalness and dynamic smoothness, but also is capable to control or modify speech timbres easily and flexibly. Furthermore, it needs less number of control parameters and very low system sampling rate.

研究语音生成的逆向问题，基于语音声学，运用声道扰动理论，由语音信号特征确定动态声道形状。研究基于发音模型的语音合成方法，把逆向解得的声道参数作为合成器控制参数，用少量参数合成出音色可控制的语音。为提高合成语音的动态过渡和自然度及文语转换质量提供新的方法和途径。对发展多媒体通信和信息高速公路中的语音技术有科学及实用价值。