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 2024-02-29 10:30
Multi-task learning with age information model for highly accurate elderly speech recognition.
Shine Takumi, Kinouchi Takahiro, Wakabayashi Yukoh, Kitaoka Norihide (TUT) EA2023-64 SIP2023-111 SP2023-46
Abstract (in Japanese) (See Japanese page) 
(in English) The speech recognition of the elderly is less accurate, especially in smart speaker speech recognition, due to aging-related factors such as vocal organ deterioration, increased pauses, and decreased speech rate, leading to differences in acoustic features. Challenges include difficulty in recording elderly speech and limited data availability. The aim of this study is to enhance the accuracy of elderly speech recognition. As a proposed method, we built two models: one integrating age estimation into the speech recognition model and another employing adversarial multitask learning with GAN framework. Comparing the accuracy of these models revealed improved recognition for both elderly and general adult speech, demonstrating the effectiveness of the multitasking approach.
Keyword (in Japanese) (See Japanese page) 
(in English) Elderly Speech / Speech Recognition / Multi-task Learning / Adversarial Multi-task Learning / / / /  
Reference Info. IEICE Tech. Rep., vol. 123, no. 403, SP2023-46, pp. 19-24, Feb. 2024.
Paper # SP2023-46 
Date of Issue 2024-02-22 (EA, SIP, SP) 
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 EA2023-64 SIP2023-111 SP2023-46

Conference Information
Committee SIP SP EA IPSJ-SLP  
Conference Date 2024-02-29 - 2024-03-01 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To SP 
Conference Code 2024-02-SIP-SP-EA-SLP 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Multi-task learning with age information model for highly accurate elderly speech recognition. 
Sub Title (in English)  
Keyword(1) Elderly Speech  
Keyword(2) Speech Recognition  
Keyword(3) Multi-task Learning  
Keyword(4) Adversarial Multi-task Learning  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Shine Takumi  
1st Author's Affiliation Toyohashi University of Technology (TUT)
2nd Author's Name Kinouchi Takahiro  
2nd Author's Affiliation Toyohashi University of Technology (TUT)
3rd Author's Name Wakabayashi Yukoh  
3rd Author's Affiliation Toyohashi University of Technology (TUT)
4th Author's Name Kitaoka Norihide  
4th Author's Affiliation Toyohashi University of Technology (TUT)
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 ()
Speaker Author-1 
Date Time 2024-02-29 10:30:00 
Presentation Time 20 minutes 
Registration for SP 
Paper # EA2023-64, SIP2023-111, SP2023-46 
Volume (vol) vol.123 
Number (no) no.401(EA), no.402(SIP), no.403(SP) 
Page pp.19-24 
#Pages
Date of Issue 2024-02-22 (EA, SIP, SP) 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan