Corpus-based synthesis of Mandarin speech with F0 contours generated by superposing tone components on rule-generated phrase components

Abstract

Mandarin speech synthesis was conducted by generating prosodic features by the proposed method and segmental features by HMM-based method. The proposed method generates sentence fundamental frequency (F0) contours by representing them as a superposition of tone components on phrase components. The tone components are realized by concatenating their fragments at tone nuclei predicted by a corpus-based method, while the phrase components are generated by rules under the generation process model (F0 model) framework. The method includes prediction of phoneme/pause durations in a statistical method as the first step. Through a listening test on the quality of synthetic speech, it was shown that a better quality was obtainable by the method as compared to that by the full HMM-based method. It was also shown that a better quality is obtainable as compared to the case of generating F0 contours without super-positional scheme.