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 2026-03-01 10:45
Cross-Lingual Dementia Detection from Speech Using DANN
Ryosuke Oba, Mondher Bouazizi, Tomoaki Ohtsuki (Keio Univ.) SeMI2025-67
Abstract (in Japanese) (See Japanese page) 
(in English) Dementia is an umbrella term for conditions in which declines in cognitive function and behavioral symptoms significantly impair daily living. With rapid population aging, dementia has become a major global public health challenge. In recent years, speech-based dementia detection has attracted attention as a low-cost and scalable method. However, most well-established large-scale datasets are limited to languages such as English and Chinese, and large datasets are not sufficiently available for less-resourced languages. One way to address this issue is to transfer models trained on well-established large-scale datasets to other languages. Yet differences in phonological and prosodic systems across languages can cause domain shift, leading to performance degradation in the target language.
In this study, we evaluated cross-lingual dementia detection using a domain adaptation method, the Domain-Adversarial Neural Network (DANN). We used Wav2vec2.0 embeddings and the eGeMAPS acoustic feature set as input features. Experiments on the English, Chinese, Spanish, Greek datasets showed that, under multiple settings, DANN achieved better dementia detection performance in terms of Area Under the Curve (AUC) than a simple multilayer perceptron (MLP) baseline for transfer learning. SHapley Additive exPlanations (SHAP) analysis further indicated that disfluency features were among the most influential factors for dementia detection. Overall, these results suggest that DANN is effective for cross-lingual dementia detection.
Keyword (in Japanese) (See Japanese page) 
(in English) Dementia detection / Domain-Adversarial Neural Network / Domain adaptation / Machine learning / / / /  
Reference Info. IEICE Tech. Rep., vol. 125, no. 367, SeMI2025-67, pp. 5-10, March 2026.
Paper # SeMI2025-67 
Date of Issue 2026-02-22 (SeMI) 
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 SeMI2025-67

Conference Information
Committee SeMI IPSJ-MBL IPSJ-UBI  
Conference Date 2026-03-01 - 2026-03-02 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To SeMI 
Conference Code 2026-03-SeMI-MBL-UBI 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Cross-Lingual Dementia Detection from Speech Using DANN 
Sub Title (in English)  
Keyword(1) Dementia detection  
Keyword(2) Domain-Adversarial Neural Network  
Keyword(3) Domain adaptation  
Keyword(4) Machine learning  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Ryosuke Oba  
1st Author's Affiliation Keio University Graduate School (Keio Univ.)
2nd Author's Name Mondher Bouazizi  
2nd Author's Affiliation Keio University (Keio Univ.)
3rd Author's Name Tomoaki Ohtsuki  
3rd Author's Affiliation Keio University (Keio Univ.)
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 2026-03-01 10:45:00 
Presentation Time 25 minutes 
Registration for SeMI 
Paper # SeMI2025-67 
Volume (vol) vol.125 
Number (no) no.367 
Page pp.5-10 
#Pages 6 
Date of Issue 2026-02-22 (SeMI) 

[Return to Top Page]

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