T-Modules: Translation Modules for Zero-Shot Cross-Modal Machine Translation

Paul-Ambroise Duquenne, Hongyu Gong, Benoît Sagot, Holger Schwenk


Abstract
We present a new approach to perform zero-shot cross-modal transfer between speech and text for translation tasks. Multilingual speech and text are encoded in a joint fixed-size representation space. Then, we compare different approaches to decode these multimodal and multilingual fixed-size representations, enabling zero-shot translation between languages and modalities. All our models are trained without the need of cross-modal labeled translation data.Despite a fixed-size representation, we achieve very competitive results on several text and speech translation tasks. In particular, we significantly improve the state-of-the-art for zero-shot speech translation on Must-C. Incorporating a speech decoder in our framework, we introduce the first results for zero-shot direct speech-to-speech and text-to-speech translation.
Anthology ID:
2022.emnlp-main.391
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5794–5806
Language:
URL:
https://aclanthology.org/2022.emnlp-main.391
DOI:
10.18653/v1/2022.emnlp-main.391
Bibkey:
Cite (ACL):
Paul-Ambroise Duquenne, Hongyu Gong, Benoît Sagot, and Holger Schwenk. 2022. T-Modules: Translation Modules for Zero-Shot Cross-Modal Machine Translation. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 5794–5806, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
T-Modules: Translation Modules for Zero-Shot Cross-Modal Machine Translation (Duquenne et al., EMNLP 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.emnlp-main.391.pdf