Presentation 2001/12/13
A method for improving the accuracy of dynamic features calculation in speech recognition
Takeshi S. KOBAYAKAWA, Hiroyuki SEGI, Atsushi MATSUI, Kazuo ONOE, Shoe SATO, TORU Imai, Akio ANDO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A new method for estimating dynamic features used in speech recognition is proposed. The proposed method involves:1)increasing time resolution of static features used for dynamic feature estimation, 2)relaxing the restriction that estimation span must be an integer multiple of frame shift, and making the estimation span arbitrary. Evaluating the proposed method with a broadcast news testset, word accuracy improved by selecting appropriate estimation span. Error reduction rate in word accuracy was up to 24% relatively in noisy news speech. The overall error reduction rate was 6.5% relatively.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Speech Recognition / Dynamic Features / Regression Coefficient
Paper # NLC2001-49,SP2001-84
Date of Issue

Conference Information
Committee NLC
Conference Date 2001/12/13(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 method for improving the accuracy of dynamic features calculation in speech recognition
Sub Title (in English)
Keyword(1) Speech Recognition
Keyword(2) Dynamic Features
Keyword(3) Regression Coefficient
1st Author's Name Takeshi S. KOBAYAKAWA
1st Author's Affiliation NHK Science & Technical Research Labs.()
2nd Author's Name Hiroyuki SEGI
2nd Author's Affiliation NHK Science & Technical Research Labs.
3rd Author's Name Atsushi MATSUI
3rd Author's Affiliation NHK Science & Technical Research Labs.
4th Author's Name Kazuo ONOE
4th Author's Affiliation NHK Science & Technical Research Labs.
5th Author's Name Shoe SATO
5th Author's Affiliation NHK Science & Technical Research Labs.
6th Author's Name TORU Imai
6th Author's Affiliation NHK Science & Technical Research Labs.
7th Author's Name Akio ANDO
7th Author's Affiliation NHK Science & Technical Research Labs.
Date 2001/12/13
Paper # NLC2001-49,SP2001-84
Volume (vol) vol.101
Number (no) 520
Page pp.pp.-
#Pages 6
Date of Issue