Direct Speech-to-speech translation (S2ST) has drawn more and more attention recently. The task is very challenging due to data scarcity and complex speech-to-speech mapping. In this paper, we report our recent achievements in S2ST. Firstly, we build a S2ST Transformer baseline which outperforms the original Translatotron. Secondly, we utilize the external data by pseudo-labeling and obtain a new state-of-the-art result on the Fisher English-to-Spanish test set. Indeed, we exploit the pseudo data with a combination of popular techniques which are not trivial when applied to S2ST. Moreover, we evaluate our approach on both syntactically similar (Spanish-English) and distant (English-Chinese) language pairs. Our implementation is available at https://github.com/fengpeng-yue/speech-to-speech-translation.
@article{arxiv.2205.08993,
title = {Leveraging Pseudo-labeled Data to Improve Direct Speech-to-Speech Translation},
author = {Qianqian Dong and Fengpeng Yue and Tom Ko and Mingxuan Wang and Qibing Bai and Yu Zhang},
journal= {arXiv preprint arXiv:2205.08993},
year = {2022}
}