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 2018-03-20 13:55
DNN prefiltering for enhancement of voice recognition in noise environment
Jun Takahashi, Kentaro Murase (Fujitsu Labs.) EA2017-170 SIP2017-179 SP2017-153
Abstract (in Japanese) (See Japanese page) 
(in English) In this paper, we applied convolutional denoising autoencoder (CDAE) as the prefilter of voice recognition and evaluated recognition accuracy of under the noise environoment such as BGM/environmental sound.
To improve the voice recognition accuracy under the noise environment, prefilter requires not only noise reduction but also speech feature restoration.
We trained CDAE which is the model that the noise signal converts into the speech signal.
And, for restoration of speech features, we applied U-Net structure which connects same layer of multiple stacked autoencoder.
Furthermore, considering further improvement of speech features, generative adversarial networks (GAN) are added to the model.
We evaluated the voice recognition accuracy under the noise environment using prefilter that trained the voice of 20 people and 5 patterns of BGM/environmental sound.
In the result, we confirmed that the speech recognition accuracy improved more than 10 points under the SNR 10dB.
Keyword (in Japanese) (See Japanese page) 
(in English) Deep Neural Network / Autoencoder / Voice recognition / Generative Adversarial Network / Denoising / / /  
Reference Info. IEICE Tech. Rep., vol. 117, no. 517, SP2017-153, pp. 373-378, March 2018.
Paper # SP2017-153 
Date of Issue 2018-03-12 (EA, SIP, 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 EA2017-170 SIP2017-179 SP2017-153

Conference Information
Committee SIP EA SP MI  
Conference Date 2018-03-19 - 2018-03-20 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English) Speech, Engineering/Electro Acoustics, Signal Processing, and Related Topics [SIP, EA, SP]/ Medical Image Engineering, Analysis, Recognition, etc. [MI] 
Paper Information
Registration To SP 
Conference Code 2018-03-SIP-EA-SP-MI 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) DNN prefiltering for enhancement of voice recognition in noise environment 
Sub Title (in English)  
Keyword(1) Deep Neural Network  
Keyword(2) Autoencoder  
Keyword(3) Voice recognition  
Keyword(4) Generative Adversarial Network  
Keyword(5) Denoising  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Jun Takahashi  
1st Author's Affiliation Fujitsu Laboratories LTD. (Fujitsu Labs.)
2nd Author's Name Kentaro Murase  
2nd Author's Affiliation Fujitsu Laboratories LTD. (Fujitsu Labs.)
3rd Author's Name  
3rd Author's Affiliation ()
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 2018-03-20 13:55:00 
Presentation Time 25 minutes 
Registration for SP 
Paper # EA2017-170, SIP2017-179, SP2017-153 
Volume (vol) vol.117 
Number (no) no.515(EA), no.516(SIP), no.517(SP) 
Page pp.373-378 
#Pages 6 
Date of Issue 2018-03-12 (EA, SIP, SP) 

[Return to Top Page]

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