Interpreting deep urban sound classification using Layer-wise Relevance Propagation

TL;DR

This paper develops an explainable AI framework using Layer-wise Relevance Propagation to interpret deep neural networks for urban sound classification, aiding applications like assisting hearing-impaired drivers.

Contribution

It introduces a method to interpret deep urban sound classifiers using relevance propagation on different audio representations, enhancing model transparency.

Findings

High relevance frequency content indicates discriminative features.

Layer-wise relevance helps justify model decisions.

The framework improves understanding of urban sound classification.

Abstract

After constructing a deep neural network for urban sound classification, this work focuses on the sensitive application of assisting drivers suffering from hearing loss. As such, clear etiology justifying and interpreting model predictions comprise a strong requirement. To this end, we used two different representations of audio signals, i.e. Mel and constant-Q spectrograms, while the decisions made by the deep neural network are explained via layer-wise relevance propagation. At the same time, frequency content assigned with high relevance in both feature sets, indicates extremely discriminative information characterizing the present classification task. Overall, we present an explainable AI framework for understanding deep urban sound classification.