Presentation 2006/1/20
A Robust Voice Activity Detector for Embedded Capture Device
David COURNAPEAU, Kenji MASE, Yasuyuki SUMI, Tomoji TORIYAMA, Kiyoshi KOGURE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper addresses the problem of segmenting audio data in utterances for the purpose of multi-modal analysis of social interactions. We propose a method for robust speech detection in natural, noisy environments for embedded devices. Preliminary results suggest that it can provide good results under different type of natural noises, even for low signal-noise ratio and low sampling rate. As the algorithm is based on features which have a low computational cost, and has a latency below one second, it is suitable for real-time applications.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) speech detection / real-time / correlogram / tracking
Paper # SP2005-151
Date of Issue

Conference Information
Committee SP
Conference Date 2006/1/20(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 ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Robust Voice Activity Detector for Embedded Capture Device
Sub Title (in English)
Keyword(1) speech detection
Keyword(2) real-time
Keyword(3) correlogram
Keyword(4) tracking
1st Author's Name David COURNAPEAU
1st Author's Affiliation ATR Media Information Science()
2nd Author's Name Kenji MASE
2nd Author's Affiliation Nagoya University, Graduate School of Information Science:Nagoya University, Information Technology Center:ATR Media Information Science
3rd Author's Name Yasuyuki SUMI
3rd Author's Affiliation ATR Media Information Science:Kyoto University, Graduate School of Informatics
4th Author's Name Tomoji TORIYAMA
4th Author's Affiliation ATR Media Information Science
5th Author's Name Kiyoshi KOGURE
5th Author's Affiliation ATR Media Information Science
Date 2006/1/20
Paper # SP2005-151
Volume (vol) vol.105
Number (no) 572
Page pp.pp.-
#Pages 6
Date of Issue