English

Translation-Equivariant Self-Supervised Learning for Pitch Estimation with Optimal Transport

Sound 2025-10-28 v1 Artificial Intelligence Machine Learning Audio and Speech Processing

Abstract

In this paper, we propose an Optimal Transport objective for learning one-dimensional translation-equivariant systems and demonstrate its applicability to single pitch estimation. Our method provides a theoretically grounded, more numerically stable, and simpler alternative for training state-of-the-art self-supervised pitch estimators.

Keywords

Cite

@article{arxiv.2508.01493,
  title  = {Translation-Equivariant Self-Supervised Learning for Pitch Estimation with Optimal Transport},
  author = {Bernardo Torres and Alain Riou and Gaël Richard and Geoffroy Peeters},
  journal= {arXiv preprint arXiv:2508.01493},
  year   = {2025}
}

Comments

Extended Abstracts for the Late-Breaking Demo Session of the 26th International Society for Music Information Retrieval Conference

R2 v1 2026-07-01T04:31:21.169Z