A Study on a Low-Resource Speech Recognition System for Taiwan Hakka Based on Whisper and LoRA

Zheng-Ting Liu, Heng-You Wang, Yi-Xiang Liao, Zhong-Yuan Qiu, Zhao-Yi Huang


Abstract
This study presents the development of a high-performance automatic speech recognition (ASR) system for Taiwan Hakka, a low-resource language facing challenges in preservation and digitalization. We adopt OpenAI’s Whisper large-v3-taiwanese-hakka as the foundation, leveraging its advanced Transformer encoder–decoder architecture. To achieve parameter efficiency and adaptability to a new language, we employ the Low-Rank Adaptation (LoRA) fine-tuning strategy, targeting key modules including q_proj, k_proj, v_proj, out_proj, fc1, and fc2. Experimental results demonstrate that the fine-tuned model achieves strong performance on the FSR 2025 HAT-Vol2 test set, with an average character error rate (CER) of 7.07% and an average word error rate (WER) of 40.99%. Training analysis further indicates that both validation loss and error rates consistently decreased and converged, confirming that LoRA enables effective knowledge transfer to Hakka ASR without catastrophic forgetting. These findings provide an efficient and practical solution for speech recognition in low-resource languages.
Anthology ID:
2025.rocling-main.54
Volume:
Proceedings of the 37th Conference on Computational Linguistics and Speech Processing (ROCLING 2025)
Month:
November
Year:
2025
Address:
National Taiwan University, Taipei City, Taiwan
Editors:
Kai-Wei Chang, Ke-Han Lu, Chih-Kai Yang, Zhi-Rui Tam, Wen-Yu Chang, Chung-Che Wang
Venue:
ROCLING
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
459–466
Language:
URL:
https://aclanthology.org/2025.rocling-main.54/
DOI:
Bibkey:
Cite (ACL):
Zheng-Ting Liu, Heng-You Wang, Yi-Xiang Liao, Zhong-Yuan Qiu, and Zhao-Yi Huang. 2025. A Study on a Low-Resource Speech Recognition System for Taiwan Hakka Based on Whisper and LoRA. In Proceedings of the 37th Conference on Computational Linguistics and Speech Processing (ROCLING 2025), pages 459–466, National Taiwan University, Taipei City, Taiwan. Association for Computational Linguistics.
Cite (Informal):
A Study on a Low-Resource Speech Recognition System for Taiwan Hakka Based on Whisper and LoRA (Liu et al., ROCLING 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.rocling-main.54.pdf