An Excitation Model for HMM-Based Speech Synthesis Based on Residual Modeling

Abstract

This paper describes a trainable excitation approach to eliminate the unnaturalness of HMM-based speech synthesizers. During the waveform generation part, mixed excitation is constructed by state-dependent filtering of pulse trains and white noise sequences. In the training part, filters and pulse trains are jointly optimized through a procedure which resembles analysis-bysynthesis speech coding algorithms, where likelihood maximization of residual signals (derived from the same database which is used to train the HMM-based synthesizer) is pursued. Preliminary results show that the novel excitation model in question eliminates the unnaturalness of synthesized speech, being comparable in quality to the the best approaches thus far reported to eradicate the buzziness of HMM-based synthesizers.