English

Fast-HuBERT: An Efficient Training Framework for Self-Supervised Speech Representation Learning

Computation and Language 2023-10-02 v2 Artificial Intelligence Machine Learning Sound Audio and Speech Processing

Abstract

Recent years have witnessed significant advancements in self-supervised learning (SSL) methods for speech-processing tasks. Various speech-based SSL models have been developed and present promising performance on a range of downstream tasks including speech recognition. However, existing speech-based SSL models face a common dilemma in terms of computational cost, which might hinder their potential application and in-depth academic research. To address this issue, we first analyze the computational cost of different modules during HuBERT pre-training and then introduce a stack of efficiency optimizations, which is named Fast-HuBERT in this paper. The proposed Fast-HuBERT can be trained in 1.1 days with 8 V100 GPUs on the Librispeech 960h benchmark, without performance degradation, resulting in a 5.2x speedup, compared to the original implementation. Moreover, we explore two well-studied techniques in the Fast-HuBERT and demonstrate consistent improvements as reported in previous work.

Keywords

Cite

@article{arxiv.2309.13860,
  title  = {Fast-HuBERT: An Efficient Training Framework for Self-Supervised Speech Representation Learning},
  author = {Guanrou Yang and Ziyang Ma and Zhisheng Zheng and Yakun Song and Zhikang Niu and Xie Chen},
  journal= {arXiv preprint arXiv:2309.13860},
  year   = {2023}
}
R2 v1 2026-06-28T12:31:07.309Z