Addressing Emotion Bias in Music Emotion Recognition and Generation with Frechet Audio Distance

TL;DR

This paper investigates emotion bias in music emotion recognition and generation, proposing the use of Frechet Audio Distance with multiple encoders for more objective evaluation and introducing an improved emotional music generation method.

Contribution

It introduces a novel evaluation approach using FAD with diverse encoders and enhances EMG to produce more realistic and varied musical emotions.

Findings

FAD with multiple encoders provides more objective MER evaluation

Enhanced EMG improves emotional variability and realism

Emotion bias affects both recognition and generation tasks

Abstract

The complex nature of musical emotion introduces inherent bias in both recognition and generation, particularly when relying on a single audio encoder, emotion classifier, or evaluation metric. In this work, we conduct a study on Music Emotion Recognition (MER) and Emotional Music Generation (EMG), employing diverse audio encoders alongside Frechet Audio Distance (FAD), a reference-free evaluation metric. Our study begins with a benchmark evaluation of MER, highlighting the limitations of using a single audio encoder and the disparities observed across different measurements. We then propose assessing MER performance using FAD derived from multiple encoders to provide a more objective measure of musical emotion. Furthermore, we introduce an enhanced EMG approach designed to improve both the variability and prominence of generated musical emotion, thereby enhancing its realism. Additionally, we investigate the differences in realism between the emotions conveyed in real and synthetic music, comparing our EMG model against two baseline models. Experimental results underscore the issue of emotion bias in both MER and EMG and demonstrate the potential of using FAD and diverse audio encoders to evaluate musical emotion more objectively and effectively.