Presentation 2008-06-13
Acoustic Modeling for Robust Recognition using Running Spectrum Analysis
Kazunaga OHNUKI, Wataru TAKAHASHI, Yoshikazu MIYANAGA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, new robust speech feature model is designed and applied for the continuous speech recognition system. Using this features, the noise robust continuous speech recognition can be realized. The new running spectrum analysis (RSA) method is used in order to remove un-speech components over 15Hz in modulation spectrum domain. Using RSA, speech features are emphasized for the design of tri-phone HMM. In order to show the performance of the proposed method, some comparisons with conventional ones are given in experiments.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) RSA / Robust speech recognition / Modulation spectrum / Acoustic Modeling
Paper # SIS2008-16
Date of Issue

Conference Information
Committee SIS
Conference Date 2008/6/6(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 Smart Info-Media Systems (SIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Acoustic Modeling for Robust Recognition using Running Spectrum Analysis
Sub Title (in English)
Keyword(1) RSA
Keyword(2) Robust speech recognition
Keyword(3) Modulation spectrum
Keyword(4) Acoustic Modeling
1st Author's Name Kazunaga OHNUKI
1st Author's Affiliation Kushiro National College of Technology()
2nd Author's Name Wataru TAKAHASHI
2nd Author's Affiliation Graduate School of Information Sicience and Technology, Hokkaido University
3rd Author's Name Yoshikazu MIYANAGA
3rd Author's Affiliation Graduate School of Information Sicience and Technology, Hokkaido University
Date 2008-06-13
Paper # SIS2008-16
Volume (vol) vol.108
Number (no) 86
Page pp.pp.-
#Pages 6
Date of Issue