Presentation 1998/9/11
State Duration Modeling for HMM-Based Speech Synthesis
Takayoshi Yoshimura, Keiichi Tokuda, Takashi Masuko, Takao Kobayashi, Tadashi Kitamura,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a new approach to state duration modeling for HMM-based speech synthesis. A set of state durations of each phoneme HMM is modeled by a multi-dimensional Gaussian distribution. Duration models are clustered using a decision tree based context clustering technique. In the synthesis stage, state durations are determined by maximizing the state duration probability. In this paper, we take account of contextual factors such as mora count, stress and part-of-speech in addition to current, preceding and succeeding phonemes. Experimental results show that the synthetic speech has a good quality with natural timing.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) HMM / text-to-speech synthesis / state duration model / mel-cepstrum / context clustering
Paper # DSP98-85,SP98-64
Date of Issue

Conference Information
Committee DSP
Conference Date 1998/9/11(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Digital Signal Processing (DSP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) State Duration Modeling for HMM-Based Speech Synthesis
Sub Title (in English)
Keyword(1) HMM
Keyword(2) text-to-speech synthesis
Keyword(3) state duration model
Keyword(4) mel-cepstrum
Keyword(5) context clustering
1st Author's Name Takayoshi Yoshimura
1st Author's Affiliation Department of Computer Science, Nagoya Inst.of Tech.()
2nd Author's Name Keiichi Tokuda
2nd Author's Affiliation Department of Computer Science, Nagoya Inst.of Tech.
3rd Author's Name Takashi Masuko
3rd Author's Affiliation Precision and Intelligence Lab., Tokyo Inst.of Tech.
4th Author's Name Takao Kobayashi
4th Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Inst.of Tech.
5th Author's Name Tadashi Kitamura
5th Author's Affiliation Department of Computer Science, Nagoya Inst.of Tech.
Date 1998/9/11
Paper # DSP98-85,SP98-64
Volume (vol) vol.98
Number (no) 262
Page pp.pp.-
#Pages 6
Date of Issue