Paper Abstract and Keywords |
Presentation |
2019-01-27 11:30
Multimodal Data Augmentation for Visual Speech Recognition using Deep Canonical Correlation Analysis Masaki Shimonishi, Satoshi Tamura, Satoru Hayamizu (Gifu University) SP2018-60 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
This paper proposes ta new data augmentation strategy for deep learning, in which feature vectors in one modality can be used as an additional training data set in the target modality by means of Deep Canonical Correlation Analysis (DCCA). Particularly, this work applies our scheme to visual speech recognition, i.e. lipread- ing, employing audio training feature vectors. A long short-term memory is chosen as a recognition model, which is built using visual training data and augmented data from the audio modality. We evaluated our method using an audio-visual corpus, preparing audio data by adding acoustic noises. Experimental results show applying DCCA enables us to augment training data from the other modalities, and it is turned out that we can further improve the model by implicitly utilize information and knowledge in those modalities. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
lip reading / deep learning / deep canonical correlation analysis / multimodal speech recognition / / / / |
Reference Info. |
IEICE Tech. Rep., vol. 118, no. 426, SP2018-60, pp. 41-45, Jan. 2019. |
Paper # |
SP2018-60 |
Date of Issue |
2019-01-19 (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 |
SP2018-60 |
Conference Information |
Committee |
SP |
Conference Date |
2019-01-26 - 2019-01-27 |
Place (in Japanese) |
(See Japanese page) |
Place (in English) |
Kanazawa-Harmonie |
Topics (in Japanese) |
(See Japanese page) |
Topics (in English) |
Speech Synthesis, Generation, Prosody, Emergency Broadcast, etc. |
Paper Information |
Registration To |
SP |
Conference Code |
2019-01-SP |
Language |
Japanese |
Title (in Japanese) |
(See Japanese page) |
Sub Title (in Japanese) |
(See Japanese page) |
Title (in English) |
Multimodal Data Augmentation for Visual Speech Recognition using Deep Canonical Correlation Analysis |
Sub Title (in English) |
|
Keyword(1) |
lip reading |
Keyword(2) |
deep learning |
Keyword(3) |
deep canonical correlation analysis |
Keyword(4) |
multimodal speech recognition |
Keyword(5) |
|
Keyword(6) |
|
Keyword(7) |
|
Keyword(8) |
|
1st Author's Name |
Masaki Shimonishi |
1st Author's Affiliation |
Gifu University (Gifu University) |
2nd Author's Name |
Satoshi Tamura |
2nd Author's Affiliation |
Gifu University (Gifu University) |
3rd Author's Name |
Satoru Hayamizu |
3rd Author's Affiliation |
Gifu University (Gifu University) |
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 |
() |
Speaker |
Author-1 |
Date Time |
2019-01-27 11:30:00 |
Presentation Time |
25 minutes |
Registration for |
SP |
Paper # |
SP2018-60 |
Volume (vol) |
vol.118 |
Number (no) |
no.426 |
Page |
pp.41-45 |
#Pages |
5 |
Date of Issue |
2019-01-19 (SP) |
|