Soft Dynamic Time Warping for Multi-Pitch Estimation and Beyond

Michael Krause, Christof Weiß, Meinard Müller
4/11/2023
cs.SDcs.LGeess.AS

No official code found. Best implementation: groupmm/softdtw_for_mpe (4 stars, MIT, last commit 2023-07-19).

Abstract

Many tasks in music information retrieval (MIR) involve weakly aligned data, where exact temporal correspondences are unknown. The connectionist temporal classification (CTC) loss is a standard technique to learn feature representations based on weakly aligned training data. However, CTC is limited to discrete-valued target sequences and can be difficult to extend to multi-label problems. In this article, we show how soft dynamic time warping (SoftDTW), a differentiable variant of classical DTW, can be used as an alternative to CTC. Using multi-pitch estimation as an example scenario, we show that SoftDTW yields results on par with a state-of-the-art multi-label extension of CTC. In addition to being more elegant in terms of its algorithmic formulation, SoftDTW naturally extends to real-valued target sequences.

View on arXivView PDF

Code Implementations(1)

groupmm/softdtw_for_mpeCommunity61%
MIT

Code and trained models for the paper "Soft Dynamic Time Warping for Multi-Pitch Estimation and Beyond", ICASSP 2023

41PythonJul 19, 20233 years agoMIT

Cite this paper

@article{krause2023soft,
  title  = {Soft Dynamic Time Warping for Multi-Pitch Estimation and Beyond},
  author = {Michael Krause and Christof Weiß and Meinard Müller},
  year   = {2023},
  eprint = {2304.05032},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2304.05032v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2304.05032.svg)](https://codeofpaper.com/paper/2304.05032)

Discussion