This paper describes our work in participation of the IWSLT-2021 offline speech translation task. Our system was built in a cascade form, including a speaker diarization module, an Automatic Speech Recognition (ASR) module and a Machine Translation (MT) module. We directly use the LIUM SpkDiarization tool as the diarization module. The ASR module is trained with three ASR datasets from different sources, by multi-source training, using a modified Transformer encoder. The MT module is pretrained on the large-scale WMT news translation dataset and fine-tuned on the TED corpus. Our method achieves 24.6 BLEU score on the 2021 test set.
@article{arxiv.2108.03845,
title = {The HW-TSC's Offline Speech Translation Systems for IWSLT 2021 Evaluation},
author = {Minghan Wang and Yuxia Wang and Chang Su and Jiaxin Guo and Yingtao Zhang and Yujia Liu and Min Zhang and Shimin Tao and Xingshan Zeng and Liangyou Li and Hao Yang and Ying Qin},
journal= {arXiv preprint arXiv:2108.03845},
year = {2021}
}