Adapting Offline Speech Translation Models for Streaming with Future-Aware Distillation and Inference
Biao Fu, Minpeng Liao, Kai Fan, Zhongqiang Huang, Boxing Chen, Yidong Chen, Xiaodong Shi
Abstract
A popular approach to streaming speech translation is to employ a single offline model with a wait-k policy to support different latency requirements, which is simpler than training multiple online models with different latency constraints. However, there is a mismatch problem in using a model trained with complete utterances for streaming inference with partial input. We demonstrate that speech representations extracted at the end of a streaming input are significantly different from those extracted from a complete utterance. To address this issue, we propose a new approach called Future-Aware Streaming Translation (FAST) that adapts an offline ST model for streaming input. FAST includes a Future-Aware Inference (FAI) strategy that incorporates future context through a trainable masked embedding, and a Future-Aware Distillation (FAD) framework that transfers future context from an approximation of full speech to streaming input. Our experiments on the MuST-C EnDe, EnEs, and EnFr benchmarks show that FAST achieves better trade-offs between translation quality and latency than strong baselines. Extensive analyses suggest that our methods effectively alleviate the aforementioned mismatch problem between offline training and online inference.- Anthology ID:
- 2023.emnlp-main.1033
- Volume:
- Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 16600–16619
- Language:
- URL:
- https://aclanthology.org/2023.emnlp-main.1033
- DOI:
- 10.18653/v1/2023.emnlp-main.1033
- Cite (ACL):
- Biao Fu, Minpeng Liao, Kai Fan, Zhongqiang Huang, Boxing Chen, Yidong Chen, and Xiaodong Shi. 2023. Adapting Offline Speech Translation Models for Streaming with Future-Aware Distillation and Inference. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 16600–16619, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Adapting Offline Speech Translation Models for Streaming with Future-Aware Distillation and Inference (Fu et al., EMNLP 2023)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2023.emnlp-main.1033.pdf