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 2022-05-13 14:35
A serial anomalous sound detection method using outlier exposure based on two types of binary classification
Ibuki Kuroyanagi (Nagoya Univ.), Tomoki Hayashi (Nagoya Univ./HDL/), Kazuya Takeda, Tomoki Toda (Nagoya Univ.) EA2022-8
Abstract (in Japanese) (See Japanese page) 
(in English) Anomalous sound detection systems use only normal sound data to detect unknown, atypical sounds. Conventional methods use a serial method, a combination of outlier exposure, which classifies normal and pseudo-anomalous data and obtains embedding, and inlier modeling, which models the probability distribution of the embedding. Outlier exposure has a difficulty in training a good classifier when normal data and pseudo-anomalous data are too similar or too different. To explicitly distinguish cases where normal data and pseudo-anomalous data are too similar or too different, the proposed method performs two types of binary classification tasks when training outlier exposure. It allows more anomalous data to be detected. Evaluation results on the DCASE~2021 Task~2 dataset show that the proposed method, with a single model, outperforms the top methods that ensemble multiple models by 2.1,% in the harmonic mean of AUC and pAUC ($p=0.1$).
Keyword (in Japanese) (See Japanese page) 
(in English) anomalous sound detection / outlier exposure / inlier modeling / hypersphere / multi-task learning / / /  
Reference Info. IEICE Tech. Rep., vol. 122, no. 20, EA2022-8, pp. 35-40, May 2022.
Paper # EA2022-8 
Date of Issue 2022-05-06 (EA) 
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 EA2022-8

Conference Information
Committee EA  
Conference Date 2022-05-13 - 2022-05-13 
Place (in Japanese) (See Japanese page) 
Place (in English) Online 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To EA 
Conference Code 2022-05-EA 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) A serial anomalous sound detection method using outlier exposure based on two types of binary classification 
Sub Title (in English)  
Keyword(1) anomalous sound detection  
Keyword(2) outlier exposure  
Keyword(3) inlier modeling  
Keyword(4) hypersphere  
Keyword(5) multi-task learning  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Ibuki Kuroyanagi  
1st Author's Affiliation Nagoya University (Nagoya Univ.)
2nd Author's Name Tomoki Hayashi  
2nd Author's Affiliation Nagoya University/Human Dataware Lab. Co. Ltd. (Nagoya Univ./HDL/)
3rd Author's Name Kazuya Takeda  
3rd Author's Affiliation Nagoya University (Nagoya Univ.)
4th Author's Name Tomoki Toda  
4th Author's Affiliation Nagoya University (Nagoya Univ.)
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 2022-05-13 14:35:00 
Presentation Time 25 minutes 
Registration for EA 
Paper # EA2022-8 
Volume (vol) vol.122 
Number (no) no.20 
Page pp.35-40 
#Pages
Date of Issue 2022-05-06 (EA) 


[Return to Top Page]

[Return to IEICE Web Page]


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