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 2025-03-04 11:05
[Poster Presentation] Improvement and Evaluation of End of Utterance Time Estimation Method for Spoken Dialog Systems
Takanori Kanai, Yukoh Wakabayashi (TUT), Ryota Nishimura (Tokushima Univ.), Norihide Kitaoka (TUT) EA2024-139 SIP2024-174 SP2024-80
Abstract (in Japanese) (See Japanese page) 
(in English) To achieve natural interaction in spoken dialog systems, accurate turn-taking prediction is essential.
Consequently, many turn-taking prediction models have been proposed, enabling smooth responses without unnatural pauses. However, in actual dialogs, responses do not always occur immediately after the end of a conversation partner's utterance, because sometimes humans intentionally pause before responding, or their response may slightly overlap the end of the previous speaker's utterance.Therefore, we have proposed a method that estimates in advance when a speaker's utterance will end in order to enable the system to speak at more natural timing.A previous study proposed a method in which a Long Short-Term Memory (LSTM) classifier estimates the time until the end of an utterance as a classification problem, using the latent variables of a self-supervised learning model (SSL) fine-tuned for Automatic speech recognition (ASR) as input.However, it has fixed the duration of the input signal used for estimation at 3 seconds in its experiment, without examining its optimal length.To improve estimation, we conducted experiments by increasing and decreasing the duration of the input signal used for the estimation.The results showed that accuracy was improved by using utterance from the beginning to the current time, instead of using a fixed length of the input signal.The observation of actual utterance end time estimation estimated the time shorter in places where an utterance could naturally end, even in the middle of utterance.
Keyword (in Japanese) (See Japanese page) 
(in English) spoken dialog system / turn-taking / utterance-final timing prediction / / / / /  
Reference Info. IEICE Tech. Rep., vol. 124, no. 391, SP2024-80, pp. 375-380, March 2025.
Paper # SP2024-80 
Date of Issue 2025-02-23 (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 EA2024-139 SIP2024-174 SP2024-80

Conference Information
Committee EA SIP SP IPSJ-SLP  
Conference Date 2025-03-02 - 2025-03-04 
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 2025-03-EA-SIP-SP-SLP 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Improvement and Evaluation of End of Utterance Time Estimation Method for Spoken Dialog Systems 
Sub Title (in English)  
Keyword(1) spoken dialog system  
Keyword(2) turn-taking  
Keyword(3) utterance-final timing prediction  
Keyword(4)  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Takanori Kanai  
1st Author's Affiliation Toyohashi University of Technology (TUT)
2nd Author's Name Yukoh Wakabayashi  
2nd Author's Affiliation Toyohashi University of Technology (TUT)
3rd Author's Name Ryota Nishimura  
3rd Author's Affiliation Tokushima University (Tokushima Univ.)
4th Author's Name Norihide Kitaoka  
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 ()
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 2025-03-04 11:05:00 
Presentation Time 80 minutes 
Registration for SP 
Paper # EA2024-139, SIP2024-174, SP2024-80 
Volume (vol) vol.124 
Number (no) no.389(EA), no.390(SIP), no.391(SP) 
Page pp.375-380 
#Pages 6 
Date of Issue 2025-02-23 (EA, SIP, SP) 

[Return to Top Page]

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