Unsupervised Detection of Music Boundaries by Time Series Structure Features

Authors

  • Joan Serrà Artificial Intelligence Research Institute, Spanish National Research Council (IIIA-CSIC)
  • Meinard Müller Max Planck Institute for Computer Science and Saarland University
  • Peter Grosche Max Planck Institute for Computer Science and Saarland University
  • Josep Lluis Arcos Artificial Intelligence Research Institute, Spanish National Research Council (IIIA-CSIC)

DOI:

https://doi.org/10.1609/aaai.v26i1.8328

Keywords:

time series, music, information retrieval, segmentation

Abstract

Locating boundaries between coherent and/or repetitive segments of a time series is a challenging problem pervading many scientific domains. In this paper we propose an unsupervised method for boundary detection, combining three basic principles: novelty, homogeneity, and repetition. In particular, the method uses what we call structure features, a representation encapsulating both local and global properties of a time series. We demonstrate the usefulness of our approach in detecting music structure boundaries, a task that has received much attention in recent years and for which exist several benchmark datasets and publicly available annotations. We find our method to significantly outperform the best accuracies published so far. Importantly, our boundary approach is generic, thus being applicable to a wide range of time series beyond the music and audio domains.

Downloads

Published

2021-09-20

How to Cite

Serrà, J., Müller, M., Grosche, P., & Arcos, J. L. (2021). Unsupervised Detection of Music Boundaries by Time Series Structure Features. Proceedings of the AAAI Conference on Artificial Intelligence, 26(1), 1613-1619. https://doi.org/10.1609/aaai.v26i1.8328