English

A Mamba-Based Model for Automatic Chord Recognition

Sound 2026-01-06 v1

Abstract

In this work, we propose a new efficient solution, which is a Mamba-based model named BMACE (Bidirectional Mamba-based network, for Automatic Chord Estimation), which utilizes selective structured state-space models in a bidirectional Mamba layer to effectively model temporal dependencies. Our model achieves high prediction performance comparable to state-of-the-art models, with the advantage of requiring fewer parameters and lower computational resources

Keywords

Cite

@article{arxiv.2601.02101,
  title  = {A Mamba-Based Model for Automatic Chord Recognition},
  author = {Chunyu Yuan and Johanna Devaney},
  journal= {arXiv preprint arXiv:2601.02101},
  year   = {2026}
}

Comments

International Society of Music Information Retrieval, Late-Breaking Demo 2024

R2 v1 2026-07-01T08:50:51.093Z