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 2022-10-21 15:25
Features and Deep Learning Models Suitable for Speech Source Discrimination Method in Plural Voice User Interfaces Environment
Kengo Maeda, Takahiro Yoshida (TUS) PRMU2022-27
Abstract (in Japanese) (See Japanese page) 
(in English) Under the situation that plural devices equipped with a voice user interface exist in the user’s environment in the near future, technology to discriminate between directly uttered speech by the user and playbacked speech from a device will be necessary for each device to work correctly. Therefore, our previous study proposed a speech source discrimination method using a Convolutional Neural Network (CNN) and Mel-Frequency Cepstral Coefficients (MFCC). However, features and deep learning models suitable for the speech source discrimination method have not been researched in previous studies. Therefore, in this study, we compared and evaluated several features and deep learning models by their speech source discrimination accuracy to investigate suitable features and deep learning models for the speech source discrimination method. From the experimental results, we confirmed that the rich feature that includes the fine structure of the spectrum is effective for the speech source discrimination method, since the discrimination accuracy of MFCC improves as the number of dimensions increase. We also confirmed that using a pre-learned model with re-learning by fine-tuning is also effective for the speech source discrimination method.
Keyword (in Japanese) (See Japanese page) 
(in English) Voice user interface / Speech source discrimination / Convolutional neural network / Fine-tuning / / / /  
Reference Info. IEICE Tech. Rep., vol. 122, no. 223, PRMU2022-27, pp. 29-34, Oct. 2022.
Paper # PRMU2022-27 
Date of Issue 2022-10-14 (PRMU) 
ISSN 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 PRMU2022-27

Conference Information
Committee PRMU  
Conference Date 2022-10-21 - 2022-10-22 
Place (in Japanese) (See Japanese page) 
Place (in English) Miraikan - The National Museum of Emerging Science and Innovation 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Recognition and understanding related to people 
Paper Information
Registration To PRMU 
Conference Code 2022-10-PRMU 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Features and Deep Learning Models Suitable for Speech Source Discrimination Method in Plural Voice User Interfaces Environment 
Sub Title (in English)  
Keyword(1) Voice user interface  
Keyword(2) Speech source discrimination  
Keyword(3) Convolutional neural network  
Keyword(4) Fine-tuning  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Kengo Maeda  
1st Author's Affiliation Tokyo University of Science (TUS)
2nd Author's Name Takahiro Yoshida  
2nd Author's Affiliation Tokyo University of Science (TUS)
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 2022-10-21 15:25:00 
Presentation Time 15 minutes 
Registration for PRMU 
Paper # PRMU2022-27 
Volume (vol) vol.122 
Number (no) no.223 
Page pp.29-34 
#Pages 6 
Date of Issue 2022-10-14 (PRMU) 

[Return to Top Page]

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