Analysis and synthesis of the four tones in connected speech of the standard Chinese based on a command-response model

The Chinese language is a typical tone language in which a syllable possesses several tone types and thus can represent different morphemes. While these tone types have rather clear manifestations in the fundamental frequency contour in isolated syllables, they vary considerably in connected speech due to the influences of such factors as tones of adjacent syllables, syntactic and pragmatic information of the whole utterance. This paper describes the results of analysis of F0 contours of the Standard Chinese using a command-response model, and shows that systematic relationships exit between the timing of the tone commands and the “vowel plus coda” part of a syllable. The results are then used to derive rules for tone command generation in speech synthesis. The validity of the rules has been confirmed by the naturalness of prosody of synthetic speech.