Flexible Control in Symbolic Music Generation via Musical Metadata

TL;DR

This paper presents a flexible symbolic music generation method using an autoregressive model that incorporates musical metadata, allowing for controllable and high-quality music synthesis with demonstrated effectiveness and user control.

Contribution

Introduces a novel autoregressive model that enables flexible control in symbolic music generation by using randomly dropped musical metadata tokens during training.

Findings

Model achieves high musical fidelity and diversity.

Enhanced controllability over music generation.

Outperforms other models in subjective quality tests.

Abstract

In this work, we introduce the demonstration of symbolic music generation, focusing on providing short musical motifs that serve as the central theme of the narrative. For the generation, we adopt an autoregressive model which takes musical metadata as inputs and generates 4 bars of multitrack MIDI sequences. During training, we randomly drop tokens from the musical metadata to guarantee flexible control. It provides users with the freedom to select input types while maintaining generative performance, enabling greater flexibility in music composition. We validate the effectiveness of the strategy through experiments in terms of model capacity, musical fidelity, diversity, and controllability. Additionally, we scale up the model and compare it with other music generation model through a subjective test. Our results indicate its superiority in both control and music quality. We provide a URL link https://www.youtube.com/watch?v=-0drPrFJdMQ to our demonstration video.