Information: Join today and make your research activities more affordable! Technical workshop participation fees and annual registration fees are available at member rates.
Notice: [Important] Announcement of Changes to Registration Fee Payment and Manuscript Upload Procedures for IEICE Technical Meetings
IEICE Technical Committee Submission System
Conference Paper's Information		 Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2010-06-18 10:00
Noise-robust speech recognition decoder using speech/non-speech confidence measures
Tasuku Oonishi (Tokyo Inst. of Tech.), Koji Iwano (Tokyo City Univ.), Sadaoki Furui (Tokyo Inst. of Tech.) SP2010-30
Abstract (in Japanese) (See Japanese page) 
(in English) Abstract In a speech recognition system a Voice Activity Detection(VAD) is a crucial component for maintaining accuracy. This paper proposes an approach that uses speech/non-speech confidence measures to adjust the score of the recognition hypotheses. In order to achieve good search performance, it is important to properly adapt the GMMs for input utterances and environmental noise. This paper also proposes an unsupervised on-line GMM adaptation method based on MAP estimation. Robustness of the proposed method is further improved by weighting updating parameters of GMMs according to the confidence measure for the adaptation data and adaptation speed is largely accelerated by caching statistical values to adapt GMMs. Experimental results on Drivers’ Japanese Speech Corpus in a Car Environment(DJSC) show that our approach can improve the accuracy significantly as compared with typical front-end based VAD methods. The adaptation method significantly improves the word accuracy. Moreover, the weighting method improves the robustness of the unsupervised adaptation and the cache method largely accelerates the decoding process. Consequently, the proposed adaptive decoding method significantly improves word accuracy under noise with only minor increase in computational cost.
Keyword (in Japanese) (See Japanese page) 
(in English) speech recognition / voice activity detection / gaussian mixture model adaptation / / / / /  
Reference Info. IEICE Tech. Rep., vol. 110, no. 81, SP2010-30, pp. 49-54, June 2010.
Paper # SP2010-30 
Date of Issue 2010-06-10 (SP) 
ISSN Print edition: ISSN 0913-5685    Online edition: ISSN 2432-6380
Copyright
and
reproduction		 All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034)
Download PDF SP2010-30

Conference Information
Committee SP  
Conference Date 2010-06-17 - 2010-06-18 
Place (in Japanese) (See Japanese page) 
Place (in English) Kyushu University 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Speech/Language/Acoustic Education, etc. 
Paper Information
Registration To SP 
Conference Code 2010-06-SP 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Noise-robust speech recognition decoder using speech/non-speech confidence measures 
Sub Title (in English)  
Keyword(1) speech recognition  
Keyword(2) voice activity detection  
Keyword(3) gaussian mixture model adaptation  
Keyword(4)  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Tasuku Oonishi  
1st Author's Affiliation Tokyo Institute of Technology (Tokyo Inst. of Tech.)
2nd Author's Name Koji Iwano  
2nd Author's Affiliation Tokyo City University (Tokyo City Univ.)
3rd Author's Name Sadaoki Furui  
3rd Author's Affiliation Tokyo Institute of Technology (Tokyo Inst. of Tech.)
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
21st Author's Name  
21st Author's Affiliation ()
22nd Author's Name  
22nd Author's Affiliation ()
23rd Author's Name  
23rd Author's Affiliation ()
24th Author's Name  
24th Author's Affiliation ()
25th Author's Name  
25th Author's Affiliation ()
26th Author's Name / /
26th Author's Affiliation ()
()
27th Author's Name / /
27th Author's Affiliation ()
()
28th Author's Name / /
28th Author's Affiliation ()
()
29th Author's Name / /
29th Author's Affiliation ()
()
30th Author's Name / /
30th Author's Affiliation ()
()
31st Author's Name / /
31st Author's Affiliation ()
()
32nd Author's Name / /
32nd Author's Affiliation ()
()
33rd Author's Name / /
33rd Author's Affiliation ()
()
34th Author's Name / /
34th Author's Affiliation ()
()
35th Author's Name / /
35th Author's Affiliation ()
()
36th Author's Name / /
36th Author's Affiliation ()
()
Speaker Author-1 
Date Time 2010-06-18 10:00:00 
Presentation Time 30 minutes 
Registration for SP 
Paper # SP2010-30 
Volume (vol) vol.110 
Number (no) no.81 
Page pp.49-54 
#Pages 6 
Date of Issue 2010-06-10 (SP) 

[Return to Top Page]

[Return to IEICE Web Page]
The Institute of Electronics, Information and Communication Engineers (IEICE), Japan