Multilingual Speech Translation with Unified Transformer: Huawei Noah’s Ark Lab at IWSLT 2021

Xingshan Zeng, Liangyou Li, Qun Liu

Abstract
This paper describes the system submitted to the IWSLT 2021 Multilingual Speech Translation (MultiST) task from Huawei Noah’s Ark Lab. We use a unified transformer architecture for our MultiST model, so that the data from different modalities (i.e., speech and text) and different tasks (i.e., Speech Recognition, Machine Translation, and Speech Translation) can be exploited to enhance the model’s ability. Specifically, speech and text inputs are firstly fed to different feature extractors to extract acoustic and textual features, respectively. Then, these features are processed by a shared encoder–decoder architecture. We apply several training techniques to improve the performance, including multi-task learning, task-level curriculum learning, data augmentation, etc. Our final system achieves significantly better results than bilingual baselines on supervised language pairs and yields reasonable results on zero-shot language pairs.
Anthology ID:
2021.iwslt-1.17
Volume:
Proceedings of the 18th International Conference on Spoken Language Translation (IWSLT 2021)
Month:
August
Year:
2021
Address:
Bangkok, Thailand (online)
Editors:
Marcello Federico, Alex Waibel, Marta R. Costa-jussà, Jan Niehues, Sebastian Stuker, Elizabeth Salesky
Venue:
IWSLT
SIG:
SIGSLT
Publisher:
Association for Computational Linguistics
Note:
Pages:
149–153
Language:
URL:
https://aclanthology.org/2021.iwslt-1.17
DOI:
10.18653/v1/2021.iwslt-1.17
Bibkey:
Cite (ACL):
Xingshan Zeng, Liangyou Li, and Qun Liu. 2021. Multilingual Speech Translation with Unified Transformer: Huawei Noah’s Ark Lab at IWSLT 2021. In Proceedings of the 18th International Conference on Spoken Language Translation (IWSLT 2021), pages 149–153, Bangkok, Thailand (online). Association for Computational Linguistics.
Cite (Informal):
Multilingual Speech Translation with Unified Transformer: Huawei Noah’s Ark Lab at IWSLT 2021 (Zeng et al., IWSLT 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/emnlp-22-attachments/2021.iwslt-1.17.pdf
PDF Search