Detailed description of triphone model using SSS-free algorithm

Motoyuki Suzuki; Daisuke Honma; Akinori Ito; Shozo Makino

Detailed description of triphone model using SSS-free algorithm

Motoyuki Suzuki, Daisuke Honma, Akinori Ito, Shozo Makino

ENG - Communications Engineering

Research output: Contribution to journal › Conference article › peer-review

Abstract

The triphone model is frequently used as an acoustic model. It is effective for modeling phonetic variations caused by coarticulation. However, it is known that acoustic features of phonemes are also affected by other factors such as speaking style and speaking speed. In this paper, a new acoustic model is proposed. All training data which have the same phoneme context are automatically clustered into several clusters based on acoustic similarity, and a "sub-triphones" is trained using training data corresponding to a cluster. In experiments, the sub-triphone model achieved about 5% higher phoneme accuracy than the triphone model.

Original language	English
Pages (from-to)	1399-1402
Number of pages	4
Journal	Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Publication status	Published - 2009 Nov 26
Event	10th Annual Conference of the International Speech Communication Association, INTERSPEECH 2009 - Brighton, United Kingdom Duration: 2009 Sept 6 → 2009 Sept 10

Keywords

HMnet
SSS-free
Sub-triphone model
Triphone

ASJC Scopus subject areas

Human-Computer Interaction
Signal Processing
Software
Sensory Systems

Cite this

@article{a5e3d1966e40476f967a6f71897c0dde,

title = "Detailed description of triphone model using SSS-free algorithm",

abstract = "The triphone model is frequently used as an acoustic model. It is effective for modeling phonetic variations caused by coarticulation. However, it is known that acoustic features of phonemes are also affected by other factors such as speaking style and speaking speed. In this paper, a new acoustic model is proposed. All training data which have the same phoneme context are automatically clustered into several clusters based on acoustic similarity, and a {"}sub-triphones{"} is trained using training data corresponding to a cluster. In experiments, the sub-triphone model achieved about 5% higher phoneme accuracy than the triphone model.",

keywords = "HMnet, SSS-free, Sub-triphone model, Triphone",

author = "Motoyuki Suzuki and Daisuke Honma and Akinori Ito and Shozo Makino",

year = "2009",

month = nov,

day = "26",

language = "English",

pages = "1399--1402",

journal = "Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH",

issn = "2308-457X",

note = "10th Annual Conference of the International Speech Communication Association, INTERSPEECH 2009 ; Conference date: 06-09-2009 Through 10-09-2009",

}

TY - JOUR

T1 - Detailed description of triphone model using SSS-free algorithm

AU - Suzuki, Motoyuki

AU - Honma, Daisuke

AU - Ito, Akinori

AU - Makino, Shozo

PY - 2009/11/26

Y1 - 2009/11/26

N2 - The triphone model is frequently used as an acoustic model. It is effective for modeling phonetic variations caused by coarticulation. However, it is known that acoustic features of phonemes are also affected by other factors such as speaking style and speaking speed. In this paper, a new acoustic model is proposed. All training data which have the same phoneme context are automatically clustered into several clusters based on acoustic similarity, and a "sub-triphones" is trained using training data corresponding to a cluster. In experiments, the sub-triphone model achieved about 5% higher phoneme accuracy than the triphone model.

AB - The triphone model is frequently used as an acoustic model. It is effective for modeling phonetic variations caused by coarticulation. However, it is known that acoustic features of phonemes are also affected by other factors such as speaking style and speaking speed. In this paper, a new acoustic model is proposed. All training data which have the same phoneme context are automatically clustered into several clusters based on acoustic similarity, and a "sub-triphones" is trained using training data corresponding to a cluster. In experiments, the sub-triphone model achieved about 5% higher phoneme accuracy than the triphone model.

KW - HMnet

KW - SSS-free

KW - Sub-triphone model

KW - Triphone

UR - http://www.scopus.com/inward/record.url?scp=70450158870&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70450158870&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:70450158870

SN - 2308-457X

SP - 1399

EP - 1402

JO - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

JF - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

T2 - 10th Annual Conference of the International Speech Communication Association, INTERSPEECH 2009

Y2 - 6 September 2009 through 10 September 2009

ER -

Detailed description of triphone model using SSS-free algorithm

Abstract

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this