deHuBERT: Disentangling Noise in a Self-supervised Model for Robust Speech Recognition

Autor:	Ng, Dianwen, Zhang, Ruixi, Yip, Jia Qi, Yang, Zhao, Ni, Jinjie, Zhang, Chong, Ma, Yukun, Ni, Chongjia, Chng, Eng Siong, Ma, Bin
Rok vydání:	2023
Předmět:	Computer Science - Sound Electrical Engineering and Systems Science - Audio and Speech Processing
Druh dokumentu:	Working Paper
Popis:	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. Comment: Accepted by ICASSP 2023
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2302.14597 Zobrazit plný text záznamu View this record from Arxiv