IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
... (for ESS/CS/ES/ISS)
Tech. Rep. Archives
... (for ES)
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2014-05-24 11:30
[研究紹介] A spectrogram-patch-input DNN model for detection and classification of acoustic events robust to speech overlapping scenarios
Miquel Espi, Masakiyo Fujimoto, Yotaro Kubo, Tomohiro Nakatani (NTT)
Abstract (in Japanese) (See Japanese page) 
(in English) This paper presents an acoustic event detection and classification method that learns features from spectrogram patches (i.e. concatenation of a certain number of consecutive spectrum frames) in an unsupervised manner, and integrates effortlessly within the deep neural network framework. Most AED use-cases happen in scenarios where speech overlaps with acoustic events, and while derived features (e.g. MFCCs, Mel-filter-banks) have traditionally characterized well the spectrum of speech, they are too dense and centered on specific frequencies to be used with non-speech tasks. Results show that the proposed model based on spectrogram-patch out-performs those based on derived features, as well as previous AED works.
Keyword (in Japanese) (See Japanese page) 
(in English) acoustic event detection / deep neural network / spectrogram patch / source sparsity / / / /  
Reference Info. IEICE Tech. Rep., vol. 114, no. 52, SP2014-17, pp. 171-176, May 2014.
Paper # SP2014-17 
Date of Issue 2014-05-17 (SP) 
ISSN Print edition: ISSN 0913-5685  Online edition: ISSN 2432-6380

Conference Information
Committee SP IPSJ-MUS  
Conference Date 2014-05-24 - 2014-05-25 
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 2014-05-SP-MUS 
Language English (Japanese title is available) 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) [研究紹介] A spectrogram-patch-input DNN model for detection and classification of acoustic events robust to speech overlapping scenarios 
Sub Title (in English)  
Keyword(1) acoustic event detection  
Keyword(2) deep neural network  
Keyword(3) spectrogram patch  
Keyword(4) source sparsity  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Miquel Espi  
1st Author's Affiliation Nippon Telegraph and Telephone Corporation (NTT)
2nd Author's Name Masakiyo Fujimoto  
2nd Author's Affiliation Nippon Telegraph and Telephone Corporation (NTT)
3rd Author's Name Yotaro Kubo  
3rd Author's Affiliation Nippon Telegraph and Telephone Corporation (NTT)
4th Author's Name Tomohiro Nakatani  
4th Author's Affiliation Nippon Telegraph and Telephone Corporation (NTT)
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
Date Time 2014-05-24 11:30:00 
Presentation Time 240 
Registration for SP 
Paper # IEICE-SP2014-17 
Volume (vol) IEICE-114 
Number (no) no.52 
Page pp.171-176 
#Pages IEICE-6 
Date of Issue IEICE-SP-2014-05-17 


[Return to Top Page]

[Return to IEICE Web Page]


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