An excitation model for HMM-based speech synthesis based on residual modeling

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.

[1]  Wai C. Chu,et al.  Speech Coding Algorithms , 2003 .

[2]  Heiga Zen,et al.  Details of the Nitech HMM-Based Speech Synthesis System for the Blizzard Challenge 2005 , 2007, IEICE Trans. Inf. Syst..

[3]  T.H. Crystal,et al.  Linear prediction of speech , 1977, Proceedings of the IEEE.

[4]  Chuan Yi Tang,et al.  A 2.|E|-Bit Distributed Algorithm for the Directed Euler Trail Problem , 1993, Inf. Process. Lett..

[5]  Sherif Abdou,et al.  Improving Arabic HMM based speech synthesis quality , 2006, INTERSPEECH.

[6]  Takao Kobayashi,et al.  Speaking style adaptation using context clustering decision tree for HMM-based speech synthesis , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[7]  Minsoo Hahn,et al.  Two-Band Excitation for HMM-Based Speech Synthesis , 2007, IEICE Trans. Inf. Syst..

[8]  P. Welch.  The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms , 1967 .

[9]  Takehiko Kagoshima,et al.  Analytic generation of synthesis units by closed loop training for totally speaker driven text to speech system (TOS drive TTS) , 1998, ICSLP.

[10]  Keiichi Tokuda,et al.  Mixed excitation for HMM-based speech synthesis , 2001, INTERSPEECH.

[11]  Keiichi Tokuda,et al.  Simultaneous modeling of spectrum, pitch and duration in HMM-based speech synthesis , 1999, EUROSPEECH.

[12]  H. Zen,et al.  An HMM-based speech synthesis system applied to English , 2002, Proceedings of 2002 IEEE Workshop on Speech Synthesis, 2002..

[13]  Hideki Kawahara,et al.  Restructuring speech representations using a pitch-adaptive time-frequency smoothing and an instantaneous-frequency-based F0 extraction: Possible role of a repetitive structure in sounds , 1999, Speech Commun..

[14]  Keiichi Tokuda,et al.  An adaptive algorithm for mel-cepstral analysis of speech , 1992, [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[15]  Eric Moulines,et al.  Pitch-synchronous waveform processing techniques for text-to-speech synthesis using diphones , 1989, Speech Commun..

[16]  Antonio Bonafonte,et al.  Residual Conversion Versus Prediction on Voice Morphing Systems , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.