Presentation 2002/1/18
PHONOBEST : a noise-robust automatic speech recognition system with phonetic estimation process based on "expectation"
Ikuyo KATSUSE, Yoshimori SUGANO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a new method, PHONOBEST (PHOnetic kNOwledge Biased ESTimation) for speech recognition in the presence of non-stationary and unpredictable noise by extending PreFEst (Predominant-FO Estimation Method) that was developed to estimate melody and bass lines from music signals. The proposed method does not need to know about the characteristics of noises in advance and does no even estimate them in its process. A small set of evaluations demonstrate the feasibility of the method by showing good performance even when the background noise is real and non-stationary noise and SNR is 10 dB.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) non-stationary noise / automatic speech recognition / phonetic knowledge / top-down process
Paper #
Date of Issue

Conference Information
Committee SP
Conference Date 2002/1/18(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 Speech (SP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) PHONOBEST : a noise-robust automatic speech recognition system with phonetic estimation process based on "expectation"
Sub Title (in English)
Keyword(1) non-stationary noise
Keyword(2) automatic speech recognition
Keyword(3) phonetic knowledge
Keyword(4) top-down process
1st Author's Name Ikuyo KATSUSE
1st Author's Affiliation Kyushu Institute of Design:Institute of Systems $amp; Information Technologies()
2nd Author's Name Yoshimori SUGANO
2nd Author's Affiliation Kyushu Institute of Design:Institute of Systems $amp; Information Technologies
Date 2002/1/18
Paper #
Volume (vol) vol.101
Number (no) 604
Page pp.pp.-
#Pages 8
Date of Issue