Symbolic Music Generation with Diffusion Models

Score-based generative models and diffusion probabilistic models have been successful at generating high-quality samples in continuous domains such as images and audio. However, due to their Langevin-inspired sampling mechanisms, their application to discrete and sequential data has been limited. In...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-11
Main Authors: Mittal, Gautam, Engel, Jesse, Hawthorne, Curtis, Simon, Ian
Format: Article
Language:English
Subjects:
Autoregressive models
Diffusion
Domains
Iterative methods
Music
Probabilistic models
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Score-based generative models and diffusion probabilistic models have been successful at generating high-quality samples in continuous domains such as images and audio. However, due to their Langevin-inspired sampling mechanisms, their application to discrete and sequential data has been limited. In this work, we present a technique for training diffusion models on sequential data by parameterizing the discrete domain in the continuous latent space of a pre-trained variational autoencoder. Our method is non-autoregressive and learns to generate sequences of latent embeddings through the reverse process and offers parallel generation with a constant number of iterative refinement steps. We apply this technique to modeling symbolic music and show strong unconditional generation and post-hoc conditional infilling results compared to autoregressive language models operating over the same continuous embeddings.
ISSN:2331-8422