Non-Linear Speech Processing (NOLISP 03)

May 20-23, 2003
Le Croisic, France

Towards an Oscillator-plus-Noise Model for Speech Synthesis

Erhard Rank (1), Gernot Kubin (2)

(1) Institute of Communications and Radio-Frequency Engineering, Vienna University of Technology, Vienna, Austria
(2) Institute of Communications and Wave Propagation, Graz University of Technology, Graz, Austria

The autonomous oscillator model for speech synthesis is augmented by a nonlinear predic- tor to regenerate the modulated noiselike signal component of speech signals. The resulting `oscillator-plus-noise' model in combination with vocal tract modeling by linear prediction is able to regenerate the spectral content of stationary wide-band vowel signals with high fidelity. For adequate modeling of voiced fricatives the model is further extended by a second linear prediction path. With one and the same model not only sustained voiced and mixed excitation phonemes, but also unvoiced sounds can be regenerated faithfully.

Full Paper

Bibliographic reference.  Rank, Erhard / Kubin, Gernot (2003): "Towards an oscillator-plus-noise model for speech synthesis", In NOLISP-2003, paper 027.