D3PIA: A Discrete Denoising Diffusion Model for Piano Accompaniment Generation From Lead sheet

2026-02-03

AI Summary

This work proposes a piano accompaniment generation method based on a discrete denoising diffusion mechanism, aiming to automatically synthesize structurally coherent and musically expressive piano accompaniments from lead sheets comprising melody and chord sequences. It introduces the first application of discrete diffusion models to symbolic music accompaniment generation, representing music in piano-roll format and employing a local alignment strategy combined with neighborhood attention mechanisms to effectively capture local dependencies between melody and harmony. Experimental results on the POP909 dataset demonstrate that the proposed model adheres more faithfully to chord constraints according to objective metrics and achieves superior musical coherence and expressiveness compared to both continuous diffusion models and Transformer-based baselines in subjective listening evaluations.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Generating piano accompaniments in the symbolic music domain is a challenging task that requires producing a complete piece of piano music from given melody and chord constraints, such as those provided by a lead sheet. In this paper, we propose a discrete diffusion-based piano accompaniment generation model, D3PIA, leveraging local alignment between lead sheet and accompaniment in piano-roll representation. D3PIA incorporates Neighborhood Attention (NA) to both encode the lead sheet and condition it for predicting note states in the piano accompaniment. This design enhances local contextual modeling by efficiently attending to nearby melody and chord conditions. We evaluate our model using the POP909 dataset, a widely used benchmark for piano accompaniment generation. Objective evaluation results demonstrate that D3PIA preserves chord conditions more faithfully compared to continuous diffusion-based and Transformer-based baselines. Furthermore, a subjective listening test indicates that D3PIA generates more musically coherent accompaniments than the comparison models.

Problem

• piano accompaniment generation
• lead sheet
• symbolic music
• melody and chord constraints

Innovation

• Discrete Diffusion
• Piano Accompaniment Generation
• Neighborhood Attention
• Lead Sheet
• Symbolic Music