SKILL: Similarity-aware Knowledge distILLation for Speech Self-Supervised Learning

arXiv:2402.16830v1 Announce Type: cross
Abstract: Self-supervised learning (SSL) has achieved remarkable success across various speech-processing tasks. To enhance its efficiency, previous works often leverage the use of compression techniques. A notable recent attempt is DPHuBERT, which applies joint knowledge distillation (KD) and structured pruning to learn a significantly smaller SSL model. In this paper, we contribute to this research domain by introducing SKILL, a novel method that conducts distillation across groups of layers instead of distilling individual arbitrarily selected layers …

abstract arxiv compression cs.cl cs.lg cs.sd distillation eess.as efficiency knowledge learn processing pruning self-supervised learning speech ssl success supervised learning tasks type