ETRW on Dialogue and Prosody September 1-3, 1999 Veldhoven, The Netherlands

Speech Rate Control for Dialogue Speech Synthesis Based on the Prosodic Structures

Hiromichi Kawanami, Keikichi Hirose

Dept. of Information and Communication Engineering, Univ. of Tokyo, Japan

For the purpose of constructing prosodic rules for dialogue speech synthesis a comparative study of speech rate was conducted between dialogue speech and read speech. Based on the generation modeling of FO contour we can define four prosodic units such as prosodic sentence, prosodic phrase, and so on. Speech rate was analyzed with respect te these units especially for the prosodic phrase. Dialogue speech rate starts with a siower value than that of read speech. Then it gradually increases and after passing through the middle of the phrase decreases. Based on the results rules for speech rate control which used reduction rate template for mora, were developed for dialogne speech synthesis. A hearing test showed that speech synthesized by the developed rules sounded more dialogue-like.

Full Paper (PDF)   Zipped Postscript

Bibliographic reference.  Kawanami, Hiromichi / Hirose, Keikichi (1999): "Speech rate control for dialogue speech synthesis based on the prosodic structures", In DIAPRO-1999, 59-64 .