Understanding Self-Attention of Self-Supervised Audio Transformers

06/05/2020
by   Shu-wen Yang, et al.
0

Self-supervised Audio Transformers (SAT) enable great success in many downstream speech applications like ASR, but how they work has not been widely explored yet. In this work, we present multiple strategies for the analysis of attention mechanisms in SAT. We categorize attentions into explainable categories, where we discover each category possesses its own unique functionality. We provide a visualization tool for understanding multi-head self-attention, importance ranking strategies for identifying critical attention, and attention refinement techniques to improve model performance.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset