Diffusion-based Symbolic Music Generation with Structured State Space Models

TL;DR

This paper introduces SMDIM, a diffusion-based symbolic music generation model that combines Structured State Space Models and a novel MFA block to achieve scalable, efficient, and high-quality long-sequence music synthesis.

Contribution

The paper presents a new diffusion architecture integrating SSMs and MFA blocks, enabling efficient long-sequence symbolic music generation with improved quality and scalability.

Findings

Outperforms state-of-the-art models in quality and efficiency

Achieves near-linear complexity for long sequences

Successfully models traditional Chinese folk music datasets

Abstract

Recent advancements in diffusion models have significantly improved symbolic music generation. However, most approaches rely on transformer-based architectures with self-attention mechanisms, which are constrained by quadratic computational complexity, limiting scalability for long sequences. To address this, we propose Symbolic Music Diffusion with Mamba (SMDIM), a novel diffusion-based architecture integrating Structured State Space Models (SSMs) for efficient global context modeling and the Mamba-FeedForward-Attention Block (MFA) for precise local detail preservation. The MFA Block combines the linear complexity of Mamba layers, the non-linear refinement of FeedForward layers, and the fine-grained precision of self-attention mechanisms, achieving a balance between scalability and musical expressiveness. SMDIM achieves near-linear complexity, making it highly efficient for long-sequence tasks. Evaluated on diverse datasets, including FolkDB, a collection of traditional Chinese folk music that represents an underexplored domain in symbolic music generation, SMDIM outperforms state-of-the-art models in both generation quality and computational efficiency. Beyond symbolic music, SMDIM's architectural design demonstrates adaptability to a broad range of long-sequence generation tasks, offering a scalable and efficient solution for coherent sequence modeling.