Composer Vector: Style-steering Symbolic Music Generation in a Latent Space

TL;DR

Composer Vector introduces a latent space steering method for symbolic music generation, enabling flexible, style-specific, and blended outputs without retraining, thus enhancing creative control.

Contribution

It presents a novel inference-time approach for style control in symbolic music generation by manipulating the model's latent space, avoiding the need for large labeled datasets or retraining.

Findings

Effectively guides music generation toward target styles.

Enables smooth and interpretable style control via a continuous coefficient.

Allows seamless fusion of multiple styles within a unified latent space.

Abstract

Symbolic music generation has made significant progress, yet achieving fine-grained and flexible control over composer style remains challenging. Existing training-based methods for composer style conditioning depend on large labeled datasets. Besides, these methods typically support only single-composer generation at a time, limiting their applicability to more creative or blended scenarios. In this work, we propose Composer Vector, an inference-time steering method that operates directly in the model's latent space to control composer style without retraining. Through experiments on multiple symbolic music generation models, we show that Composer Vector effectively guides generations toward target composer styles, enabling smooth and interpretable control through a continuous steering coefficient. It also enables seamless fusion of multiple styles within a unified latent space framework. Overall, our work demonstrates that simple latent space steering provides a practical and general mechanism for controllable symbolic music generation, enabling more flexible and interactive creative workflows. Code and Demo are available here: https://github.com/JiangXunyi/Composer-Vector and https://jiangxunyi.github.io/composervector.github.io/