Submission of USTC’s System for the IWSLT 2023 - Offline Speech Translation Track

Xinyuan Zhou, Jianwei Cui, Zhongyi Ye, Yichi Wang, Luzhen Xu, Hanyi Zhang, Weitai Zhang, Lirong Dai


Abstract
This paper describes the submissions of the research group USTC-NELSLIP to the 2023 IWSLT Offline Speech Translation competition, which involves translating spoken English into written Chinese. We utilize both cascaded models and end-to-end models for this task. To improve the performance of the cascaded models, we introduce Whisper to reduce errors in the intermediate source language text, achieving a significant improvement in ASR recognition performance. For end-to-end models, we propose Stacked Acoustic-and-Textual En- coding extension (SATE-ex), which feeds the output of the acoustic decoder into the textual decoder for information fusion and to prevent error propagation. Additionally, we improve the performance of the end-to-end system in translating speech by combining the SATE-ex model with the encoder-decoder model through ensembling.
Anthology ID:
2023.iwslt-1.15
Volume:
Proceedings of the 20th International Conference on Spoken Language Translation (IWSLT 2023)
Month:
July
Year:
2023
Address:
Toronto, Canada (in-person and online)
Editors:
Elizabeth Salesky, Marcello Federico, Marine Carpuat
Venue:
IWSLT
SIG:
SIGSLT
Publisher:
Association for Computational Linguistics
Note:
Pages:
194–201
Language:
URL:
https://aclanthology.org/2023.iwslt-1.15
DOI:
10.18653/v1/2023.iwslt-1.15
Bibkey:
Cite (ACL):
Xinyuan Zhou, Jianwei Cui, Zhongyi Ye, Yichi Wang, Luzhen Xu, Hanyi Zhang, Weitai Zhang, and Lirong Dai. 2023. Submission of USTC’s System for the IWSLT 2023 - Offline Speech Translation Track. In Proceedings of the 20th International Conference on Spoken Language Translation (IWSLT 2023), pages 194–201, Toronto, Canada (in-person and online). Association for Computational Linguistics.
Cite (Informal):
Submission of USTC’s System for the IWSLT 2023 - Offline Speech Translation Track (Zhou et al., IWSLT 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-acl-2023-videos/2023.iwslt-1.15.pdf