| 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 | 2023-06-24 13:50 Domain adaptation of speech recognition models based on self-supervised learning using target domain speech Takahiro Kinouchi (TUT), Atsunori Ogawa (NTT), Yuko Wakabayashi, Norihide Kitaoka (TUT)  SP2023-19 |
|
| Abstract | (in Japanese) | (See Japanese page) |
| (in English) | In this study, we propose a domain adaptation method using only speech data in the target domain without using transcribed text data in the target domain based on a speech recognition model that has been pre-trained in the source domain. Speech recognition is used in various services and businesses, and it is known that the accuracy of speech recognition in each of these domains depends on the amount of speech data in that domain. Generally, it is desirable to train or fine-tune speech recognition models from scratch using a large amount of speech data and transcribed text data to build highly accurate models. However, preparing such data is expensive and difficult every time a model is built in each domain. Therefore, we focused on the fact that it is relatively inexpensive to prepare only audio data. Under these conditions, we developed an Encoder-Decoder speech recognition model using a Wav2Vec2.0 model as the Encoder, which was pre-trained with a large amount of target-domain speech only, and a large corpus of fine-tuned transcriptions in the non-target domain. We propose adapting an Encoder-Decoder type speech recognition model to the target domain by fine-tuning it with a large corpus of transcriptions in the off-target domain. The proposed method consists of three steps: 1) additional pre-training of wav2vec 2.0, 2) fine-tuning of wav2vec 2.0, and 3) building a Joint CTC/Transformer model with wav2vec 2.0 as the Encoder. This method improved the character error rate by approximately 3.8 pts compared to the case where the Encoder was not pre-trained in the target domain for the target domain evaluation data. | |
| Keyword | (in Japanese) | (See Japanese page) |
| (in English) | wav2vec 2.0 / domain adaptation / end-to-end speech recognition / Encoder-Decoder model / / / / | |
| Reference Info. | IEICE Tech. Rep., vol. 123, no. 88, SP2023-19, pp. 91-96, June 2023. | |
| Paper # | SP2023-19 | |
| Date of Issue | 2023-06-16 (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 | SP2023-19 |
|
| Conference Information | |
| Committee | SP IPSJ-MUS IPSJ-SLP |
| Conference Date | 2023-06-23 - 2023-06-24 |
| 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 | 2023-06-SP-MUS-SLP |
| Language | Japanese |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Domain adaptation of speech recognition models based on self-supervised learning using target domain speech |
| Sub Title (in English) | |
| Keyword(1) | wav2vec 2.0 |
| Keyword(2) | domain adaptation |
| Keyword(3) | end-to-end speech recognition |
| Keyword(4) | Encoder-Decoder model |
| Keyword(5) | |
| Keyword(6) | |
| Keyword(7) | |
| Keyword(8) | |
| 1st Author's Name | Takahiro Kinouchi |
| 1st Author's Affiliation | Toyohashi University of Technology (TUT) |
| 2nd Author's Name | Atsunori Ogawa |
| 2nd Author's Affiliation | NIPPON TELEGRAPH AND TELEPHONE CORPORATION (NTT) |
| 3rd Author's Name | Yuko Wakabayashi |
| 3rd Author's Affiliation | Toyohashi University of Technology (TUT) |
| 4th Author's Name | Norihide Kitaoka |
| 4th Author's Affiliation | Toyohashi University of Technology (TUT) |
| 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 | () |
| Speaker | Author-1 |
| Date Time | 2023-06-24 13:50:00 |
| Presentation Time | 140 minutes |
| Registration for | SP |
| Paper # | SP2023-19 |
| Volume (vol) | vol.123 |
| Number (no) | no.88 |
| Page | pp.91-96 |
| #Pages | 6 |
| Date of Issue | 2023-06-16 (SP) |
[Return to Top Page]