ComposerX: Multi-Agent Symbolic Music Composition with LLMs

TL;DR

ComposerX introduces a multi-agent framework leveraging GPT-4's reasoning to improve symbolic music composition, producing coherent, melody-rich polyphonic pieces aligned with user instructions.

Contribution

It is the first to utilize a multi-agent approach with LLMs for symbolic music composition, enhancing quality and coherence over traditional single-agent methods.

Findings

Multi-agent approach significantly improves music quality.

ComposerX produces coherent and captivating polyphonic music.

Framework adheres well to user instructions.

Abstract

Music composition represents the creative side of humanity, and itself is a complex task that requires abilities to understand and generate information with long dependency and harmony constraints. While demonstrating impressive capabilities in STEM subjects, current LLMs easily fail in this task, generating ill-written music even when equipped with modern techniques like In-Context-Learning and Chain-of-Thoughts. To further explore and enhance LLMs' potential in music composition by leveraging their reasoning ability and the large knowledge base in music history and theory, we propose ComposerX, an agent-based symbolic music generation framework. We find that applying a multi-agent approach significantly improves the music composition quality of GPT-4. The results demonstrate that ComposerX is capable of producing coherent polyphonic music compositions with captivating melodies, while adhering to user instructions.