Presentation 2004/8/12
A Study on MLLR-based Style Adaptation in HSMM-based Speech Synthesis
Junichi YAMAGISHI, Makoto TACHIBANA, Takashi MASUKO, Takao KOBAYASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes an extension of maximum likelihood linear regression (MLLR) to hidden semi-Markov model (HSMM). The HSMM-based MLLR technique can estimate regression matrices for affine transform of mean vecotrs of output and state duration distributions which maximize likelihood of adaptation data using EM algorithm. In this study, we apply the adaptation technique to speaker adaptation and style adaptation in an HSMM-based speech synthesis. From the results of subjective tests, we show that the HSMM-based MLLR technique can adapt spectrum and prosodic features of synthesized speech more effectively than conventional HMM-based MLLR technique.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Hidden serni-Markov model / HMM-based speech synthesis / MLLR adaptation
Paper # SP2004-49
Date of Issue

Conference Information
Committee SP
Conference Date 2004/8/12(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 Speech (SP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Study on MLLR-based Style Adaptation in HSMM-based Speech Synthesis
Sub Title (in English)
Keyword(1) Hidden serni-Markov model
Keyword(2) HMM-based speech synthesis
Keyword(3) MLLR adaptation
1st Author's Name Junichi YAMAGISHI
1st Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology()
2nd Author's Name Makoto TACHIBANA
2nd Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology
3rd Author's Name Takashi MASUKO
3rd Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology:Presently with Corporate Research & Development Center, Toshiba Corporation
4th Author's Name Takao KOBAYASHI
4th Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology
Date 2004/8/12
Paper # SP2004-49
Volume (vol) vol.104
Number (no) 252
Page pp.pp.-
#Pages 6
Date of Issue