In the recent trend of semi-supervised speech recognition, both self-supervised representation learning and pseudo-labeling have shown promising results. In this paper, we propose a novel approach to combine their ideas for end-to-end speech recognition model. Without any extra loss function, we utilize the Gradient Mask to optimize the model when training on pseudo-label. This method forces the speech recognition model to predict from the masked input to learn strong acoustic representation and make training robust to label noise. In our semi-supervised experiments, the method can improve the model performance when training on pseudo-label and our method achieved competitive results comparing with other semi-supervised approaches on the Librispeech 100 hours experiments.
@article{arxiv.2110.04056,
title = {Improving Pseudo-label Training For End-to-end Speech Recognition Using Gradient Mask},
author = {Shaoshi Ling and Chen Shen and Meng Cai and Zejun Ma},
journal= {arXiv preprint arXiv:2110.04056},
year = {2021}
}