Listenable Maps for Audio Classifiers
This addresses the problem of interpretability for audio classifiers, which is incremental as it builds on existing posthoc interpretation techniques.
The paper tackles the challenge of interpreting complex deep learning models for audio signals by introducing Listenable Maps for Audio Classifiers (L-MAC), a posthoc method that generates binary masks to highlight relevant audio portions, resulting in more faithful interpretations than existing methods and user preference in evaluations.
Despite the impressive performance of deep learning models across diverse tasks, their complexity poses challenges for interpretation. This challenge is particularly evident for audio signals, where conveying interpretations becomes inherently difficult. To address this issue, we introduce Listenable Maps for Audio Classifiers (L-MAC), a posthoc interpretation method that generates faithful and listenable interpretations. L-MAC utilizes a decoder on top of a pretrained classifier to generate binary masks that highlight relevant portions of the input audio. We train the decoder with a loss function that maximizes the confidence of the classifier decision on the masked-in portion of the audio while minimizing the probability of model output for the masked-out portion. Quantitative evaluations on both in-domain and out-of-domain data demonstrate that L-MAC consistently produces more faithful interpretations than several gradient and masking-based methodologies. Furthermore, a user study confirms that, on average, users prefer the interpretations generated by the proposed technique.