On the process of coarticulation for a CELP-based Chinese text-to-speech system using LSP frequencies

Jau Hung Chen; Chung Hsien Wu

On the process of coarticulation for a CELP-based Chinese text-to-speech system using LSP frequencies

Department of Computer Science and Information Engineering

Research output: Contribution to conference › Paper › peer-review

Abstract

This study proposes a novel approach based on Bayesian networks and the LSP frequencies to generate syllable prosody and the coarticulation between two concatenated syllables respectively. The Bayesian network is employed to model the relation between the prosodic information and the linguistic features. Given a Chinese character sequence, the Bayesian network can provide appropriate prosodic information, including pitch contour, syllable intensity, syllable duration and pause duration. Furthermore, the coarticulation is generated by adjusting the LSP frequencies in a CELP-based synthesizer. The synthesized speech is tested on twenty subjects. The test results indicate that the average correct rate is 95.8% for intelligibility, and the mean opinion score (MOS) is 3.2 for naturalness.

Original language	English
Pages	37-40
Number of pages	4
Publication status	Published - 1996 Dec 1
Event	Proceedings of the 1996 IEEE Region 10 TENCON - Digital Signal Processing Applications Conference. Part 2 (of 2) - Perth, Aust Duration: 1996 Nov 26 → 1996 Nov 29

Other

Other	Proceedings of the 1996 IEEE Region 10 TENCON - Digital Signal Processing Applications Conference. Part 2 (of 2)
City	Perth, Aust
Period	96-11-26 → 96-11-29

All Science Journal Classification (ASJC) codes

Computer Science Applications
Electrical and Electronic Engineering

Cite this

@conference{2a8bb5b53807460d90e53f9a1d5b0a11,

title = "On the process of coarticulation for a CELP-based Chinese text-to-speech system using LSP frequencies",

abstract = "This study proposes a novel approach based on Bayesian networks and the LSP frequencies to generate syllable prosody and the coarticulation between two concatenated syllables respectively. The Bayesian network is employed to model the relation between the prosodic information and the linguistic features. Given a Chinese character sequence, the Bayesian network can provide appropriate prosodic information, including pitch contour, syllable intensity, syllable duration and pause duration. Furthermore, the coarticulation is generated by adjusting the LSP frequencies in a CELP-based synthesizer. The synthesized speech is tested on twenty subjects. The test results indicate that the average correct rate is 95.8% for intelligibility, and the mean opinion score (MOS) is 3.2 for naturalness.",

author = "Chen, {Jau Hung} and Wu, {Chung Hsien}",

year = "1996",

month = dec,

day = "1",

language = "English",

pages = "37--40",

note = "Proceedings of the 1996 IEEE Region 10 TENCON - Digital Signal Processing Applications Conference. Part 2 (of 2) ; Conference date: 26-11-1996 Through 29-11-1996",

}

TY - CONF

T1 - On the process of coarticulation for a CELP-based Chinese text-to-speech system using LSP frequencies

AU - Chen, Jau Hung

AU - Wu, Chung Hsien

PY - 1996/12/1

Y1 - 1996/12/1

N2 - This study proposes a novel approach based on Bayesian networks and the LSP frequencies to generate syllable prosody and the coarticulation between two concatenated syllables respectively. The Bayesian network is employed to model the relation between the prosodic information and the linguistic features. Given a Chinese character sequence, the Bayesian network can provide appropriate prosodic information, including pitch contour, syllable intensity, syllable duration and pause duration. Furthermore, the coarticulation is generated by adjusting the LSP frequencies in a CELP-based synthesizer. The synthesized speech is tested on twenty subjects. The test results indicate that the average correct rate is 95.8% for intelligibility, and the mean opinion score (MOS) is 3.2 for naturalness.

AB - This study proposes a novel approach based on Bayesian networks and the LSP frequencies to generate syllable prosody and the coarticulation between two concatenated syllables respectively. The Bayesian network is employed to model the relation between the prosodic information and the linguistic features. Given a Chinese character sequence, the Bayesian network can provide appropriate prosodic information, including pitch contour, syllable intensity, syllable duration and pause duration. Furthermore, the coarticulation is generated by adjusting the LSP frequencies in a CELP-based synthesizer. The synthesized speech is tested on twenty subjects. The test results indicate that the average correct rate is 95.8% for intelligibility, and the mean opinion score (MOS) is 3.2 for naturalness.

UR - http://www.scopus.com/inward/record.url?scp=0030311837&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030311837&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0030311837

SP - 37

EP - 40

T2 - Proceedings of the 1996 IEEE Region 10 TENCON - Digital Signal Processing Applications Conference. Part 2 (of 2)

Y2 - 26 November 1996 through 29 November 1996

ER -

On the process of coarticulation for a CELP-based Chinese text-to-speech system using LSP frequencies

Abstract

Other

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this