DinoSR: Self-Distillation and Online Clustering for Self-supervised Speech Representation Learning
This work addresses speech representation learning for applications in AI and speech processing, offering a novel integration of methods but is incremental in building upon existing self-supervised techniques.
The paper tackled the problem of self-supervised speech representation learning by introducing DinoSR, which combines masked language modeling, self-distillation, and online clustering, resulting in a model that surpasses previous state-of-the-art performance in several downstream tasks.
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.