This repo contains model checkpoints for DPHuBERT, a task-agnostic compression method for speech SSL based on joint distillation and structured pruning.
Yifan Peng, Yui Sudo, Shakeel Muhammad, and Shinji Watanabe, “DPHuBERT: Joint Distillation and Pruning of Self-Supervised Speech Models,” in Proc. INTERSPEECH, 2023.