First International Conference on Spoken Language Processing (ICSLP 90)

Kobe, Japan
November 18-22, 1990

Speech Synthesis using Superposition of Sinusoidal Waves Generated by Synchronized Oscillators

Katsuhiko Shirai, Y. Sato, K. Hashimoto

Department of Electrical Engineering, Waseda University, Tokyo, Japan

In this paper, a new speech synthesis method is proposed. The voiced speech has the line spectrum structure and is represented by the superposition of sinusoidal waves which are generated by a group of mutually synchronized oscillators. This method has some features as follows. (1) Voiced and unvoiced sounds can be generated in a same framework to operate sinusoidal oscillators in parallel. (2) Since the phase and power information of each sinusoidal wave can be easily controlled, if necessary, periodic waveforms in the voiced sounds can be precisely reproduced in the time domain. (3) The pitch frequency and phoneme duration can be easily changed without much degradation of original sound quality.

Full Paper

Bibliographic reference.  Shirai, Katsuhiko / Sato, Y. / Hashimoto, K. (1990): "Speech synthesis using superposition of sinusoidal waves generated by synchronized oscillators", In ICSLP-1990, 345-348.