Symbolic music generation conditioned on continuous-valued emotions

TL;DR

This paper introduces a transformer-based model for generating multi-instrument symbolic music conditioned on continuous emotion labels, supported by a new dataset and quantitative evaluations showing improved performance over existing methods.

Contribution

It presents a novel method for emotion-conditioned music generation using continuous valence-arousal labels and provides a large-scale paired dataset.

Findings

Outperforms control token conditioning in note prediction accuracy

Achieves better emotion regulation in generated music

Demonstrates effective continuous emotion conditioning

Abstract

In this paper we present a new approach for the generation of multi-instrument symbolic music driven by musical emotion. The principal novelty of our approach centres on conditioning a state-of-the-art transformer based on continuous-valued valence and arousal labels. In addition, we provide a new large-scale dataset of symbolic music paired with emotion labels in terms of valence and arousal. We evaluate our approach in a quantitative manner in two ways, first by measuring its note prediction accuracy, and second via a regression task in the valence-arousal plane. Our results demonstrate that our proposed approaches outperform conditioning using control tokens which is representative of the current state of the art.