BIT-Xiaomi’s System for AutoSimTrans 2022
Mengge Liu, Xiang Li, Bao Chen, Yanzhi Tian, Tianwei Lan, Silin Li, Yuhang Guo, Jian Luan, Bin Wang
Abstract
This system paper describes the BIT-Xiaomi simultaneous translation system for Autosimtrans 2022 simultaneous translation challenge. We participated in three tracks: the Zh-En text-to-text track, the Zh-En audio-to-text track and the En-Es test-to-text track. In our system, wait-k is employed to train prefix-to-prefix translation models. We integrate streaming chunking to detect boundaries as the source streaming read in. We further improve our system with data selection, data-augmentation and R-drop training methods. Results show that our wait-k implementation outperforms organizer’s baseline by 8 BLEU score at most, and our proposed streaming chunking method further improves about 2 BLEU in low latency regime.- Anthology ID:
- 2022.autosimtrans-1.6
- Volume:
- Proceedings of the Third Workshop on Automatic Simultaneous Translation
- Month:
- July
- Year:
- 2022
- Address:
- Online
- Venue:
- AutoSimTrans
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 34–42
- Language:
- URL:
- https://aclanthology.org/2022.autosimtrans-1.6
- DOI:
- 10.18653/v1/2022.autosimtrans-1.6
- Cite (ACL):
- Mengge Liu, Xiang Li, Bao Chen, Yanzhi Tian, Tianwei Lan, Silin Li, Yuhang Guo, Jian Luan, and Bin Wang. 2022. BIT-Xiaomi’s System for AutoSimTrans 2022. In Proceedings of the Third Workshop on Automatic Simultaneous Translation, pages 34–42, Online. Association for Computational Linguistics.
- Cite (Informal):
- BIT-Xiaomi’s System for AutoSimTrans 2022 (Liu et al., AutoSimTrans 2022)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2022.autosimtrans-1.6.pdf
- Data
- BSTC