English

Exploring Softly Masked Language Modelling for Controllable Symbolic Music Generation

Sound 2023-05-12 v2 Machine Learning Audio and Speech Processing

Abstract

This document presents some early explorations of applying Softly Masked Language Modelling (SMLM) to symbolic music generation. SMLM can be seen as a generalisation of masked language modelling (MLM), where instead of each element of the input set being either known or unknown, each element can be known, unknown or partly known. We demonstrate some results of applying SMLM to constrained symbolic music generation using a transformer encoder architecture. Several audio examples are available at https://erl-j.github.io/smlm-web-supplement/

Keywords

Cite

@article{arxiv.2305.03530,
  title  = {Exploring Softly Masked Language Modelling for Controllable Symbolic Music Generation},
  author = {Nicolas Jonason and Bob L. T. Sturm},
  journal= {arXiv preprint arXiv:2305.03530},
  year   = {2023}
}

Comments

Version 1.1

R2 v1 2026-06-28T10:26:54.303Z