Speaker and style adaptation using average voice model for style control in HMM-based speech synthesis

Makoto Tachibana; Shinsuke Izawa; Takashi Nose; Takao Kobayashi

doi:10.1109/ICASSP.2008.4518689

Speaker and style adaptation using average voice model for style control in HMM-based speech synthesis

Makoto Tachibana, Shinsuke Izawa, Takashi Nose, Takao Kobayashi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

18 Citations (Scopus)

Abstract

We propose a technique for synthesizing speech with desired style expressivity of an arbitrary target speaker's voice. In an MLLR-based speaker adaptation technique for multiple regression hidden semi-Markov model (MRHSMM), the quality of synthesized speech crucially depends on the initial MRHSMM trained from a certain source speaker's data and it is not always possible to synthesize natural sounding speech with a given target speaker's voice. To overcome this problem, we perform simultaneous adaptation of speaker and style from an average voice model. Experimental results show that the proposed technique provides more natural sounding speech than the conventional one with speaker adaptation only.

Original language	English
Title of host publication	2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP
Pages	4633-4636
Number of pages	4
DOIs	https://doi.org/10.1109/ICASSP.2008.4518689
Publication status	Published - 2008 Sep 16
Externally published	Yes
Event	2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP - Las Vegas, NV, United States Duration: 2008 Mar 31 → 2008 Apr 4

Other

Other	2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP
Country	United States
City	Las Vegas, NV
Period	08/3/31 → 08/4/4

Keywords

Average voice model
Expressive speech synthesis
Hidden Markov model
Speaker adaptation
Style control

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2008.4518689

Fingerprint Dive into the research topics of 'Speaker and style adaptation using average voice model for style control in HMM-based speech synthesis'. Together they form a unique fingerprint.

Cite this

Speaker and style adaptation using average voice model for style control in HMM-based speech synthesis. / Tachibana, Makoto; Izawa, Shinsuke; Nose, Takashi; Kobayashi, Takao.

2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP. 2008. p. 4633-4636 4518689.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Tachibana, M, Izawa, S, Nose, T & Kobayashi, T 2008, Speaker and style adaptation using average voice model for style control in HMM-based speech synthesis. in 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP., 4518689, pp. 4633-4636, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP, Las Vegas, NV, United States, 08/3/31. https://doi.org/10.1109/ICASSP.2008.4518689

@inproceedings{7a530761bc2e407586fa0461eafcb58f,

title = "Speaker and style adaptation using average voice model for style control in HMM-based speech synthesis",

abstract = "We propose a technique for synthesizing speech with desired style expressivity of an arbitrary target speaker's voice. In an MLLR-based speaker adaptation technique for multiple regression hidden semi-Markov model (MRHSMM), the quality of synthesized speech crucially depends on the initial MRHSMM trained from a certain source speaker's data and it is not always possible to synthesize natural sounding speech with a given target speaker's voice. To overcome this problem, we perform simultaneous adaptation of speaker and style from an average voice model. Experimental results show that the proposed technique provides more natural sounding speech than the conventional one with speaker adaptation only.",

keywords = "Average voice model, Expressive speech synthesis, Hidden Markov model, Speaker adaptation, Style control",

author = "Makoto Tachibana and Shinsuke Izawa and Takashi Nose and Takao Kobayashi",

year = "2008",

month = sep,

day = "16",

doi = "10.1109/ICASSP.2008.4518689",

language = "English",

isbn = "1424414849",

pages = "4633--4636",

booktitle = "2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP",

note = "2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP ; Conference date: 31-03-2008 Through 04-04-2008",

}

TY - GEN

T1 - Speaker and style adaptation using average voice model for style control in HMM-based speech synthesis

AU - Tachibana, Makoto

AU - Izawa, Shinsuke

AU - Nose, Takashi

AU - Kobayashi, Takao

PY - 2008/9/16

Y1 - 2008/9/16

N2 - We propose a technique for synthesizing speech with desired style expressivity of an arbitrary target speaker's voice. In an MLLR-based speaker adaptation technique for multiple regression hidden semi-Markov model (MRHSMM), the quality of synthesized speech crucially depends on the initial MRHSMM trained from a certain source speaker's data and it is not always possible to synthesize natural sounding speech with a given target speaker's voice. To overcome this problem, we perform simultaneous adaptation of speaker and style from an average voice model. Experimental results show that the proposed technique provides more natural sounding speech than the conventional one with speaker adaptation only.

AB - We propose a technique for synthesizing speech with desired style expressivity of an arbitrary target speaker's voice. In an MLLR-based speaker adaptation technique for multiple regression hidden semi-Markov model (MRHSMM), the quality of synthesized speech crucially depends on the initial MRHSMM trained from a certain source speaker's data and it is not always possible to synthesize natural sounding speech with a given target speaker's voice. To overcome this problem, we perform simultaneous adaptation of speaker and style from an average voice model. Experimental results show that the proposed technique provides more natural sounding speech than the conventional one with speaker adaptation only.

KW - Average voice model

KW - Expressive speech synthesis

KW - Hidden Markov model

KW - Speaker adaptation

KW - Style control

UR - http://www.scopus.com/inward/record.url?scp=51449098017&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51449098017&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2008.4518689

DO - 10.1109/ICASSP.2008.4518689

M3 - Conference contribution

AN - SCOPUS:51449098017

SN - 1424414849

SN - 9781424414840

SP - 4633

EP - 4636

BT - 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP

T2 - 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP

Y2 - 31 March 2008 through 4 April 2008

ER -