A technique for controlling voice quality of synthetic speech using multiple regression HSMM

Makoto Tachibana; Takashi Nose; Junichi Yamagishi; Takao Kobayashi

A technique for controlling voice quality of synthetic speech using multiple regression HSMM

Makoto Tachibana, Takashi Nose, Junichi Yamagishi, Takao Kobayashi

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

Abstract

This paper describes a technique for controlling voice quality of synthetic speech using multiple regression hidden semi-Markov model (HSMM). In the technique, we assume that the mean vectors of output and state duration distribution of HSMM are modeled by multiple regression with a parameter vector called voice quality control vector. We first choose three features for controlling voice qualities, that is, "smooth voice - nonsmooth voice," "warm - cold," "high-pitched - low-pitched," and then we attempt to control voice quality of synthetic speech for these features. From the results of several subjective tests, we show that the proposed technique can change these features of voice quality intuitively.

Original language	English
Title of host publication	INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP
Publisher	International Speech Communication Association
Pages	2438-2441
Number of pages	4
Volume	5
ISBN (Print)	9781604234497
Publication status	Published - 2006 Jan 1
Externally published	Yes
Event	INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP - Pittsburgh, PA, United States Duration: 2006 Sept 17 → 2006 Sept 21

Other

Other	INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP
Country/Territory	United States
City	Pittsburgh, PA
Period	06/9/17 → 06/9/21

Keywords

HMM-based speech synthesis
HSMM
Multiple regression HMM
Voice quality control

ASJC Scopus subject areas

Computer Science(all)

Cite this

A technique for controlling voice quality of synthetic speech using multiple regression HSMM. / Tachibana, Makoto; Nose, Takashi; Yamagishi, Junichi et al.
INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP. Vol. 5 International Speech Communication Association, 2006. p. 2438-2441.

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

Tachibana, M, Nose, T, Yamagishi, J & Kobayashi, T 2006, A technique for controlling voice quality of synthetic speech using multiple regression HSMM. in INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP. vol. 5, International Speech Communication Association, pp. 2438-2441, INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP, Pittsburgh, PA, United States, 06/9/17.

@inproceedings{3d46aeb329ca4030a1419fe009c688bd,

title = "A technique for controlling voice quality of synthetic speech using multiple regression HSMM",

abstract = "This paper describes a technique for controlling voice quality of synthetic speech using multiple regression hidden semi-Markov model (HSMM). In the technique, we assume that the mean vectors of output and state duration distribution of HSMM are modeled by multiple regression with a parameter vector called voice quality control vector. We first choose three features for controlling voice qualities, that is, {"}smooth voice - nonsmooth voice,{"} {"}warm - cold,{"} {"}high-pitched - low-pitched,{"} and then we attempt to control voice quality of synthetic speech for these features. From the results of several subjective tests, we show that the proposed technique can change these features of voice quality intuitively.",

keywords = "HMM-based speech synthesis, HSMM, Multiple regression HMM, Voice quality control",

author = "Makoto Tachibana and Takashi Nose and Junichi Yamagishi and Takao Kobayashi",

year = "2006",

month = jan,

day = "1",

language = "English",

isbn = "9781604234497",

volume = "5",

pages = "2438--2441",

booktitle = "INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP",

publisher = "International Speech Communication Association",

note = "INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP ; Conference date: 17-09-2006 Through 21-09-2006",

}

TY - GEN

T1 - A technique for controlling voice quality of synthetic speech using multiple regression HSMM

AU - Tachibana, Makoto

AU - Nose, Takashi

AU - Yamagishi, Junichi

AU - Kobayashi, Takao

PY - 2006/1/1

Y1 - 2006/1/1

N2 - This paper describes a technique for controlling voice quality of synthetic speech using multiple regression hidden semi-Markov model (HSMM). In the technique, we assume that the mean vectors of output and state duration distribution of HSMM are modeled by multiple regression with a parameter vector called voice quality control vector. We first choose three features for controlling voice qualities, that is, "smooth voice - nonsmooth voice," "warm - cold," "high-pitched - low-pitched," and then we attempt to control voice quality of synthetic speech for these features. From the results of several subjective tests, we show that the proposed technique can change these features of voice quality intuitively.

AB - This paper describes a technique for controlling voice quality of synthetic speech using multiple regression hidden semi-Markov model (HSMM). In the technique, we assume that the mean vectors of output and state duration distribution of HSMM are modeled by multiple regression with a parameter vector called voice quality control vector. We first choose three features for controlling voice qualities, that is, "smooth voice - nonsmooth voice," "warm - cold," "high-pitched - low-pitched," and then we attempt to control voice quality of synthetic speech for these features. From the results of several subjective tests, we show that the proposed technique can change these features of voice quality intuitively.

KW - HMM-based speech synthesis

KW - HSMM

KW - Multiple regression HMM

KW - Voice quality control

UR - http://www.scopus.com/inward/record.url?scp=44949155552&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=44949155552&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:44949155552

SN - 9781604234497

VL - 5

SP - 2438

EP - 2441

BT - INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP

PB - International Speech Communication Association

T2 - INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP

Y2 - 17 September 2006 through 21 September 2006

ER -

A technique for controlling voice quality of synthetic speech using multiple regression HSMM

Abstract

Other

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this