Presentation 2004/12/14
A Study on Speech Recognition Using Syllable Duration Ratio Model
Masahide ARIU, Takashi MASUKO, Shinichi TANAKA, Akinori KAWAMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes a new duration model for speech recognition. Syllable duration ratio model is defined as a probability function for a duration ratio of neighboring syllables. The syllable duration ratio model can add duration information to speech recognition system, and is less subject to speech rate. The modeling technique is evaluated.in two tasks of speech recognition. In the experiments, the proposed model improves the speech recognition, although the speech rates of test data are different from that of training data.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Syllable / Duration / Speech rate / N-Best
Paper # NLC2004-73,SP2004-113
Date of Issue

Conference Information
Committee NLC
Conference Date 2004/12/14(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Study on Speech Recognition Using Syllable Duration Ratio Model
Sub Title (in English)
Keyword(1) Syllable
Keyword(2) Duration
Keyword(3) Speech rate
Keyword(4) N-Best
1st Author's Name Masahide ARIU
1st Author's Affiliation Corporate Research & Development Center, Toshiba Corporation()
2nd Author's Name Takashi MASUKO
2nd Author's Affiliation Corporate Research & Development Center, Toshiba Corporation
3rd Author's Name Shinichi TANAKA
3rd Author's Affiliation Corporate Research & Development Center, Toshiba Corporation
4th Author's Name Akinori KAWAMURA
4th Author's Affiliation Corporate Research & Development Center, Toshiba Corporation
Date 2004/12/14
Paper # NLC2004-73,SP2004-113
Volume (vol) vol.104
Number (no) 539
Page pp.pp.-
#Pages 6
Date of Issue