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 2019-03-15 10:25
Neural Language Models based on Conditional Hierarchical Recurrent Encoder-Decoder for Multi-Party Conversational Speech Recognition
Ryo Masumura, Tomohiro Tanaka, Atsushi Ando, Takanobu Oba, Yushi Aono (NTT) EA2018-131 SIP2018-137 SP2018-93
Abstract (in Japanese) (See Japanese page) 
(in English) This paper presents fully neural network based language models (LMs) that can leverage long-range conversational contexts beyond utterance boundaries for enhancing automatic speech recognition performance of role play dialogues such as contact center dialogues or service center dialogues. The proposed methods called role play dialogue aware LMs are composed by extending hierarchical recurrent encoder-decoder modeling so as to handle speaker role information in the role play dialogues. This enables us to leverage sequential conversational contexts from start-of-conversation to a current word in a current utterance for estimating the generative probability of a next word. Experiments using contact center dialogue data sets demonstrate the effectiveness of the proposed method in terms of perplexity and word error rate.
Keyword (in Japanese) (See Japanese page) 
(in English) multi-party conversational automatic speech recognition / role play dialogue aware language models / conditional hierarchical recurrent encoder-decoder / contact center dialogues / / / /  
Reference Info. IEICE Tech. Rep., vol. 118, no. 497, SP2018-93, pp. 191-196, March 2019.
Paper # SP2018-93 
Date of Issue 2019-03-07 (EA, SIP, SP) 
ISSN Print edition: ISSN 0913-5685    Online edition: ISSN 2432-6380
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 EA2018-131 SIP2018-137 SP2018-93

Conference Information
Committee EA SIP SP  
Conference Date 2019-03-14 - 2019-03-15 
Place (in Japanese) (See Japanese page) 
Place (in English) i+Land nagasaki (Nagasaki-shi) 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Engineering/Electro Acoustics, Signal Processing, Speech, and Related Topics 
Paper Information
Registration To SP 
Conference Code 2019-03-EA-SIP-SP 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Neural Language Models based on Conditional Hierarchical Recurrent Encoder-Decoder for Multi-Party Conversational Speech Recognition 
Sub Title (in English)  
Keyword(1) multi-party conversational automatic speech recognition  
Keyword(2) role play dialogue aware language models  
Keyword(3) conditional hierarchical recurrent encoder-decoder  
Keyword(4) contact center dialogues  
1st Author's Name Ryo Masumura  
1st Author's Affiliation NTT Corporation (NTT)
2nd Author's Name Tomohiro Tanaka  
2nd Author's Affiliation NTT Corporation (NTT)
3rd Author's Name Atsushi Ando  
3rd Author's Affiliation NTT Corporation (NTT)
4th Author's Name Takanobu Oba  
4th Author's Affiliation NTT Corporation (NTT)
5th Author's Name Yushi Aono  
5th Author's Affiliation NTT Corporation (NTT)
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 2019-03-15 10:25:00 
Presentation Time 25 minutes 
Registration for SP 
Paper # EA2018-131, SIP2018-137, SP2018-93 
Volume (vol) vol.118 
Number (no) no.495(EA), no.496(SIP), no.497(SP) 
Page pp.191-196 
Date of Issue 2019-03-07 (EA, SIP, SP) 

[Return to Top Page]

[Return to IEICE Web Page]

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