Behavior of Keyword Spotting Networks Under Noisy Conditions

TL;DR

This paper compares various keyword spotting networks under noisy conditions and introduces adaptive batch normalization to enhance their robustness when noise levels are unknown during training.

Contribution

It provides an extensive comparison of KWS models under noise and proposes adaptive batch normalization to improve performance in high noise scenarios.

Findings

Performance deteriorates under high noise conditions

Adaptive batch normalization improves robustness of KWS networks

High noise characterization guides future model development

Abstract

Keyword spotting (KWS) is becoming a ubiquitous need with the advancement in artificial intelligence and smart devices. Recent work in this field have focused on several different architectures to achieve good results on datasets with low to moderate noise. However, the performance of these models deteriorates under high noise conditions as shown by our experiments. In our paper, we present an extensive comparison between state-of-the-art KWS networks under various noisy conditions. We also suggest adaptive batch normalization as a technique to improve the performance of the networks when the noise files are unknown during the training phase. The results of such high noise characterization enable future work in developing models that perform better in the aforementioned conditions.