deHuBERT: Disentangling Noise in a Self-supervised Model for Robust Speech Recognition
Abstract
Existing self-supervised pre-trained speech models have offered an effective way to leverage massive unannotated corpora to build good automatic speech recognition (ASR). However, many current models are trained on a clean corpus from a single source, which tends to do poorly when noise is present during testing. Nonetheless, it is crucial to overcome the adverse influence of noise for real-world applications. In this work, we propose a novel training framework, called deHuBERT, for noise reduction encoding inspired by H. Barlow's redundancy-reduction principle. The new framework improves the HuBERT training algorithm by introducing auxiliary losses that drive the self- and cross-correlation matrix between pairwise noise-distorted embeddings towards identity matrix. This encourages the model to produce noise-agnostic speech representations. With this method, we report improved robustness in noisy environments, including unseen noises, without impairing the performance on the clean set.
Cite
@article{arxiv.2302.14597,
title = {deHuBERT: Disentangling Noise in a Self-supervised Model for Robust Speech Recognition},
author = {Dianwen Ng and Ruixi Zhang and Jia Qi Yip and Zhao Yang and Jinjie Ni and Chong Zhang and Yukun Ma and Chongjia Ni and Eng Siong Chng and Bin Ma},
journal= {arXiv preprint arXiv:2302.14597},
year = {2023}
}
Comments
Accepted by ICASSP 2023