Skip to yearly menu bar Skip to main content


DinoSR: Self-Distillation and Online Clustering for Self-supervised Speech Representation Learning

Alexander Liu · Heng-Jui Chang · Michael Auli · Wei-Ning Hsu · Jim Glass

Great Hall & Hall B1+B2 (level 1) #510
[ ]
Thu 14 Dec 3 p.m. PST — 5 p.m. PST


In this paper, we introduce self-distillation and online clustering for self-supervised speech representation learning (DinoSR) which combines masked language modeling, self-distillation, and online clustering. We show that these concepts complement each other and result in a strong representation learning model for speech. DinoSR first extracts contextualized embeddings from the input audio with a teacher network, then runs an online clustering system on the embeddings to yield a machine-discovered phone inventory, and finally uses the discretized tokens to guide a student network. We show that DinoSR surpasses previous state-of-the-art performance in several downstream tasks, and provide a detailed analysis of the model and the learned discrete units.

Chat is not available.