Tracing Back Music Emotion Predictions to Sound Sources and Intuitive Perceptual Qualities

TL;DR

This paper introduces a method combining source separation and perceptual features to interpret music emotion predictions, enhancing understanding of model decisions and aiding in debugging biases.

Contribution

It presents an innovative approach that merges audioLIME with perceptual features, providing more intuitive explanations for music emotion recognition models.

Findings

Improved interpretability of emotion prediction models.

Effective debugging of biased models.

Enhanced connection between audio inputs and emotion outputs.

Abstract

Music emotion recognition is an important task in MIR (Music Information Retrieval) research. Owing to factors like the subjective nature of the task and the variation of emotional cues between musical genres, there are still significant challenges in developing reliable and generalizable models. One important step towards better models would be to understand what a model is actually learning from the data and how the prediction for a particular input is made. In previous work, we have shown how to derive explanations of model predictions in terms of spectrogram image segments that connect to the high-level emotion prediction via a layer of easily interpretable perceptual features. However, that scheme lacks intuitive musical comprehensibility at the spectrogram level. In the present work, we bridge this gap by merging audioLIME -- a source-separation based explainer -- with mid-level perceptual features, thus forming an intuitive connection chain between the input audio and the output emotion predictions. We demonstrate the usefulness of this method by applying it to debug a biased emotion prediction model.