LMAC-TD: Producing Time Domain Explanations for Audio Classifiers

TL;DR

LMAC-TD is a novel post-hoc explanation method for audio classifiers that generates time domain explanations, improving audio quality without losing faithfulness, by building upon L-MAC and integrating SepFormer.

Contribution

It introduces LMAC-TD, a new approach that produces listenable, faithful explanations directly in the time domain for audio classifiers, enhancing explanation quality.

Findings

Significantly improves audio quality of explanations

Maintains faithfulness of explanations

Builds upon L-MAC with SepFormer integration

Abstract

Neural networks are typically black-boxes that remain opaque with regards to their decision mechanisms. Several works in the literature have proposed post-hoc explanation methods to alleviate this issue. This paper proposes LMAC-TD, a post-hoc explanation method that trains a decoder to produce explanations directly in the time domain. This methodology builds upon the foundation of L-MAC, Listenable Maps for Audio Classifiers, a method that produces faithful and listenable explanations. We incorporate SepFormer, a popular transformer-based time-domain source separation architecture. We show through a user study that LMAC-TD significantly improves the audio quality of the produced explanations while not sacrificing from faithfulness.