Mandarin-English Code-Switching Speech Recognition System for Specific Domain

Chung-Pu Chiou, Hou-An Lin, Chia-Ping Chen

Abstract
This paper will introduce the use of Automatic Speech Recognition (ASR) technology to process speech content with specific domain. We will use the Conformer end-to-end model as the system architecture, and use pure Chinese data for initial training. Next, use the transfer learning technology to fine-tune the system with Mandarin-English code-switching data. Finally, use the Mandarin-English code-switching data with a specific domain makes the final fine-tuning of the model so that it can achieve a certain effect on speech recognition in a specific domain. Experiments with different fine-tuning methods reduce the final error rate from 82.0% to 34.8%.
Anthology ID:
2022.rocling-1.25
Volume:
Proceedings of the 34th Conference on Computational Linguistics and Speech Processing (ROCLING 2022)
Month:
November
Year:
2022
Address:
Taipei, Taiwan
Editors:
Yung-Chun Chang, Yi-Chin Huang
Venue:
ROCLING
SIG:
Publisher:
The Association for Computational Linguistics and Chinese Language Processing (ACLCLP)
Note:
Pages:
200–204
Language:
Chinese
URL:
https://aclanthology.org/2022.rocling-1.25
DOI:
Bibkey:
Cite (ACL):
Chung-Pu Chiou, Hou-An Lin, and Chia-Ping Chen. 2022. Mandarin-English Code-Switching Speech Recognition System for Specific Domain. In Proceedings of the 34th Conference on Computational Linguistics and Speech Processing (ROCLING 2022), pages 200–204, Taipei, Taiwan. The Association for Computational Linguistics and Chinese Language Processing (ACLCLP).
Cite (Informal):
Mandarin-English Code-Switching Speech Recognition System for Specific Domain (Chiou et al., ROCLING 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/add_acl24_videos/2022.rocling-1.25.pdf
Data
AISHELL-1, AISHELL-2
PDF Search