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 2023-03-01 10:10
Anomalous sound detection with complex-valued hybrid neural networks considering phase variations
Shota Nishiyama, Akira Tamamori (AIT) EA2022-106 SIP2022-150 SP2022-70
Abstract (in Japanese) (See Japanese page) 
(in English) Anomalous sound detection is the task of identifying whether an incoming mechanical sound is normal or anomalous. Since anomalous sounds occur infrequently and are highly diverse, it is treated as a problem of detecting anomalous sounds from normal sounds only. The acoustic features used as input to most anomalous sound detection models are mel-spectrogram. However, the phase variation is lost when the complex-spectrogram obtained by Fourier transforming the sound waveform is converted to the mel-spectrogram. In this study, we compare anomalous sound detection methods using complex-valued neural networks and real-valued neural networks to demonstrate the usefulness of phase variation. As a result of the comparison, there existed machine sounds for which phase variation was valuable and machine sounds for which it was not valuable. In this study, we propose a complex-valued hybrid neural network that combines a complex-valued module that preserves the structure of complex values and a real-valued module that takes mel-spectrogram as input for all feature extraction operations in which complex-spectrogram can be input in order to take phase variation into account. We propose a complex-valued hybrid neural network that combines a complex-valued structure-preserving module and a real-valued module that takes the mel-spectrogram as input for all feature extraction operations. Experiments verified the effectiveness of the proposed method on anomalous sound detection for multi-channel sound in the ToyADMOS dataset. Experimental results showed that the proposed method improved the average AUC of all machine sounds by around 3% compared to both complex-valued and real-valued neural networks.
Keyword (in Japanese) (See Japanese page) 
(in English) Anomalous sound detection / complex-valued neural networks / phase variations / / / / /  
Reference Info. IEICE Tech. Rep., vol. 122, no. 387, EA2022-106, pp. 185-190, Feb. 2023.
Paper # EA2022-106 
Date of Issue 2023-02-21 (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 EA2022-106 SIP2022-150 SP2022-70

Conference Information
Committee SP IPSJ-SLP EA SIP  
Conference Date 2023-02-28 - 2023-03-01 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To EA 
Conference Code 2023-02-SP-SLP-EA-SIP 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Anomalous sound detection with complex-valued hybrid neural networks considering phase variations 
Sub Title (in English)  
Keyword(1) Anomalous sound detection  
Keyword(2) complex-valued neural networks  
Keyword(3) phase variations  
Keyword(4)  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Shota Nishiyama  
1st Author's Affiliation Aichi Institute of Technology (AIT)
2nd Author's Name Akira Tamamori  
2nd Author's Affiliation Aichi Institute of Technology (AIT)
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 2023-03-01 10:10:00 
Presentation Time 20 minutes 
Registration for EA 
Paper # EA2022-106, SIP2022-150, SP2022-70 
Volume (vol) vol.122 
Number (no) no.387(EA), no.388(SIP), no.389(SP) 
Page pp.185-190 
#Pages 6 
Date of Issue 2023-02-21 (EA, SIP, SP) 

[Return to Top Page]

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