Soft Dynamic Time Warping for Multi-Pitch Estimation and Beyond

Krause M, Weiß C, Müller M (2023)

Publication Type: Conference contribution

Publication year: 2023

Journal

2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Publisher: Institute of Electrical and Electronics Engineers Inc.

Book Volume: 2023-June

Conference Proceedings Title: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Event location: Rhodes Island

ISBN: 9781728163277

DOI: 10.1109/ICASSP49357.2023.10095907

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.

Authors with CRIS profile

Michael Krause International Audio Laboratories Erlangen (AudioLabs) Christof Weiß International Audio Laboratories Erlangen (AudioLabs) Meinard Müller International Audio Laboratories Erlangen (AudioLabs)

How to cite

APA:

Krause, M., Weiß, C., & Müller, M. (2023). Soft Dynamic Time Warping for Multi-Pitch Estimation and Beyond. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Rhodes Island, GR: Institute of Electrical and Electronics Engineers Inc..

MLA:

Krause, Michael, Christof Weiß, and Meinard Müller. "Soft Dynamic Time Warping for Multi-Pitch Estimation and Beyond." Proceedings of the 48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023, Rhodes Island Institute of Electrical and Electronics Engineers Inc., 2023.

BibTeX: Download