Multilingual Speech Translation from Efficient Finetuning of Pretrained Models

Xian Li, Changhan Wang, Yun Tang, Chau Tran, Yuqing Tang, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli


Abstract
We present a simple yet effective approach to build multilingual speech-to-text (ST) translation through efficient transfer learning from a pretrained speech encoder and text decoder. Our key finding is that a minimalistic LNA (LayerNorm and Attention) finetuning can achieve zero-shot crosslingual and cross-modality transfer ability by only finetuning 10 50% of the pretrained parameters. This effectively leverages large pretrained models at low training cost such as wav2vec 2.0 for acoustic modeling, and mBART for multilingual text generation. This sets a new state-of-the-art for 36 translation directions (and surpassing cascaded ST for 26 of them) on the large-scale multilingual ST benchmark CoVoST 2 (+6.4 BLEU on average for En-X directions and +6.7 BLEU for X-En directions). Our approach demonstrates strong zero-shot performance in a many-to-many multilingual model (+5.6 BLEU on average across 28 non-English directions), making it an appealing approach for attaining high-quality speech translation with improved parameter and data efficiency.
Anthology ID:
2021.acl-long.68
Volume:
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
Month:
August
Year:
2021
Address:
Online
Editors:
Chengqing Zong, Fei Xia, Wenjie Li, Roberto Navigli
Venues:
ACL | IJCNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
827–838
Language:
URL:
https://aclanthology.org/2021.acl-long.68
DOI:
10.18653/v1/2021.acl-long.68
Bibkey:
Cite (ACL):
Xian Li, Changhan Wang, Yun Tang, Chau Tran, Yuqing Tang, Juan Pino, Alexei Baevski, Alexis Conneau, and Michael Auli. 2021. Multilingual Speech Translation from Efficient Finetuning of Pretrained Models. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 827–838, Online. Association for Computational Linguistics.
Cite (Informal):
Multilingual Speech Translation from Efficient Finetuning of Pretrained Models (Li et al., ACL-IJCNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.acl-long.68.pdf
Video:
 https://aclanthology.org/2021.acl-long.68.mp4
Data
CoVoST2Europarl-ST