1 paper · 1 filter
Kele Xu, Yulu Fang, Boda Zhou +6
This paper examines audio self-supervised learning (SSL) through the alignment between pretraining objectives, architectural inductive biases, and downstream applications. Rather t…