We present a method for introducing a text encoder into pre-trained end-to-end speech translation systems. It enhances the ability of adapting one modality (i.e., source-language speech) to another (i.e., source-language text). Thus, the speech translation model can learn from both unlabeled and labeled data, especially when the source-language text data is abundant. Beyond this, we present a denoising method to build a robust text encoder that can deal with both normal and noisy text data. Our system sets new state-of-the-arts on the MuST-C En-De, En-Fr, and LibriSpeech En-Fr tasks.
@article{arxiv.2212.01778,
title = {Improving End-to-end Speech Translation by Leveraging Auxiliary Speech and Text Data},
author = {Yuhao Zhang and Chen Xu and Bojie Hu and Chunliang Zhang and Tong Xiao and Jingbo Zhu},
journal= {arXiv preprint arXiv:2212.01778},
year = {2022}
}