Exploiting Low-Resource Code-Switching Data to Mandarin-English Speech Recognition Systems

Hou-An Lin, Chia-Ping Chen


Abstract
In this paper, we investigate how to use limited code-switching data to implement a code-switching speech recognition system. We utilize the Transformer end-to-end model to develop our code switching speech recognition system, which is trained with the Mandarin dataset and a small amount of Mandarin-English code switching dataset, as the baseline of this paper. Next, we compare the performance of systems after adding multi-task learning and transfer learning. Character Error Rate(CER) is adopted as the criterion for the system. Finally, we combined the three systems with the language model, respectively, our best result dropped to 23.9% compared with the baseline of 28.7%.
Anthology ID:
2021.rocling-1.11
Volume:
Proceedings of the 33rd Conference on Computational Linguistics and Speech Processing (ROCLING 2021)
Month:
October
Year:
2021
Address:
Taoyuan, Taiwan
Editors:
Lung-Hao Lee, Chia-Hui Chang, Kuan-Yu Chen
Venue:
ROCLING
SIG:
Publisher:
The Association for Computational Linguistics and Chinese Language Processing (ACLCLP)
Note:
Pages:
81–86
Language:
URL:
https://aclanthology.org/2021.rocling-1.11
DOI:
Bibkey:
Cite (ACL):
Hou-An Lin and Chia-Ping Chen. 2021. Exploiting Low-Resource Code-Switching Data to Mandarin-English Speech Recognition Systems. In Proceedings of the 33rd Conference on Computational Linguistics and Speech Processing (ROCLING 2021), pages 81–86, Taoyuan, Taiwan. The Association for Computational Linguistics and Chinese Language Processing (ACLCLP).
Cite (Informal):
Exploiting Low-Resource Code-Switching Data to Mandarin-English Speech Recognition Systems (Lin & Chen, ROCLING 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.rocling-1.11.pdf