A speech parameter generation algorithm using local variance for HMM-based speech synthesis

Vataya Chunwijitra; Takashi Nose; Takao Kobayashi

A speech parameter generation algorithm using local variance for HMM-based speech synthesis

Vataya Chunwijitra, Takashi Nose, Takao Kobayashi

ENG - Communications Engineering

Tokyo Institute of Technology

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

2 Citations (Scopus)

Abstract

This paper proposes a parameter generation algorithm using lo-cal variance (LV) constraint of spectral parameter trajectory for HMM-based speech synthesis. In the parameter generation pro-cess, we take account of both the HMM likelihood of speech feature vectors and a likelihood for LVs. To model LV precisely, we use dynamic features of LV with context-dependent HMMs. The objective experimental results show that the proposed tech-nique can generate a better spectral trajectory in terms of the spectral and LV distortions than a conventional technique with global variance (GV) constraint. The subjective experimental results also show that the proposed technique significantly im-prove the reproducibility of the synthetic speech than the con-ventional one.

Original language	English
Title of host publication	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012
Pages	1150-1153
Number of pages	4
Publication status	Published - 2012
Event	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012 - Portland, OR, United States Duration: 2012 Sept 9 → 2012 Sept 13

Publication series

Name	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012
Volume	2

Conference

Conference	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012
Country/Territory	United States
City	Portland, OR
Period	12/9/9 → 12/9/13

Keywords

HMM-based speech synthesis
Local variance
Over-smoothing problem
Speech parameter generation

Cite this

Chunwijitra, V., Nose, T., & Kobayashi, T. (2012). A speech parameter generation algorithm using local variance for HMM-based speech synthesis. In 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012 (pp. 1150-1153). (13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012; Vol. 2).

Chunwijitra, Vataya ; Nose, Takashi ; Kobayashi, Takao. / A speech parameter generation algorithm using local variance for HMM-based speech synthesis. 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012. 2012. pp. 1150-1153 (13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012).

@inproceedings{2ab216554cfa45be8dfef95b1e1d3f84,

title = "A speech parameter generation algorithm using local variance for HMM-based speech synthesis",

abstract = "This paper proposes a parameter generation algorithm using lo-cal variance (LV) constraint of spectral parameter trajectory for HMM-based speech synthesis. In the parameter generation pro-cess, we take account of both the HMM likelihood of speech feature vectors and a likelihood for LVs. To model LV precisely, we use dynamic features of LV with context-dependent HMMs. The objective experimental results show that the proposed tech-nique can generate a better spectral trajectory in terms of the spectral and LV distortions than a conventional technique with global variance (GV) constraint. The subjective experimental results also show that the proposed technique significantly im-prove the reproducibility of the synthetic speech than the con-ventional one.",

keywords = "HMM-based speech synthesis, Local variance, Over-smoothing problem, Speech parameter generation",

author = "Vataya Chunwijitra and Takashi Nose and Takao Kobayashi",

year = "2012",

language = "English",

isbn = "9781622767595",

series = "13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012",

pages = "1150--1153",

booktitle = "13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012",

note = "13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012 ; Conference date: 09-09-2012 Through 13-09-2012",

}

Chunwijitra, V, Nose, T & Kobayashi, T 2012, A speech parameter generation algorithm using local variance for HMM-based speech synthesis. in 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012. 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, vol. 2, pp. 1150-1153, 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, Portland, OR, United States, 12/9/9.

A speech parameter generation algorithm using local variance for HMM-based speech synthesis. / Chunwijitra, Vataya; Nose, Takashi; Kobayashi, Takao.
13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012. 2012. p. 1150-1153 (13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012; Vol. 2).

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

TY - GEN

T1 - A speech parameter generation algorithm using local variance for HMM-based speech synthesis

AU - Chunwijitra, Vataya

AU - Nose, Takashi

AU - Kobayashi, Takao

PY - 2012

Y1 - 2012

N2 - This paper proposes a parameter generation algorithm using lo-cal variance (LV) constraint of spectral parameter trajectory for HMM-based speech synthesis. In the parameter generation pro-cess, we take account of both the HMM likelihood of speech feature vectors and a likelihood for LVs. To model LV precisely, we use dynamic features of LV with context-dependent HMMs. The objective experimental results show that the proposed tech-nique can generate a better spectral trajectory in terms of the spectral and LV distortions than a conventional technique with global variance (GV) constraint. The subjective experimental results also show that the proposed technique significantly im-prove the reproducibility of the synthetic speech than the con-ventional one.

AB - This paper proposes a parameter generation algorithm using lo-cal variance (LV) constraint of spectral parameter trajectory for HMM-based speech synthesis. In the parameter generation pro-cess, we take account of both the HMM likelihood of speech feature vectors and a likelihood for LVs. To model LV precisely, we use dynamic features of LV with context-dependent HMMs. The objective experimental results show that the proposed tech-nique can generate a better spectral trajectory in terms of the spectral and LV distortions than a conventional technique with global variance (GV) constraint. The subjective experimental results also show that the proposed technique significantly im-prove the reproducibility of the synthetic speech than the con-ventional one.

KW - HMM-based speech synthesis

KW - Local variance

KW - Over-smoothing problem

KW - Speech parameter generation

UR - http://www.scopus.com/inward/record.url?scp=84878412344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878412344&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84878412344

SN - 9781622767595

T3 - 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012

SP - 1150

EP - 1153

BT - 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012

T2 - 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012

Y2 - 9 September 2012 through 13 September 2012

ER -

A speech parameter generation algorithm using local variance for HMM-based speech synthesis

Abstract

Publication series

Conference

Keywords

Other files and links

Fingerprint

Cite this