Deep audio embeddings for vocalisation clustering
PLOS ONE · 2023
DOI: 10.1371/journal.pone.0283396 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, 2023. Verfasst von Paul Best und 3 weiteren Autor:innen. Beteiligte Einrichtung: Université de Toulon, Aix Marseille Univ, CNRS, LIS, Toulon, France. Hauptthemen laut MEDLINE: Algorithms, Machine Learning. Die Arbeit wurde bisher 34-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundação Grupo Boticário de Proteção à Natureza, John D. and Catherine T. MacArthur Foundation, Society for Marine Mammalogy.
Abstract (Originalsprache)
The study of non-human animals’ communication systems generally relies on the transcription of vocal sequences using a finite set of discrete units. This set is referred to as a vocal repertoire, which is specific to a species or a sub-group of a species. When conducted by human experts, the formal description of vocal repertoires can be laborious and/or biased. This motivates computerised assistance for this procedure, for which machine learning algorithms represent a good opportunity. Unsupervised clustering algorithms are suited for grouping close points together, provided a relevant representation. This paper therefore studies a new method for encoding vocalisations, allowing for automatic clustering to alleviate vocal repertoire characterisation. Borrowing from deep representation learning, we use a convolutional auto-encoder network to learn an abstract representation of vocalisations. We report on the quality of the learnt representation, as well as of state of the art methods, by quantifying their agreement with expert labelled vocalisation types from 8 datasets of other studies across 6 species (birds and marine mammals). With this benchmark, we demonstrate that using auto-encoders improves the relevance of vocalisation representation which serves repertoire characterisation using a very limited number of settings. We also publish a Python package for the bioacoustic community to train their own vocalisation auto-encoders or use a pretrained encoder to browse vocal repertoires and ease unit wise annotation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Université de Toulon, Aix Marseille Univ, CNRS, LIS, Toulon, France.
ORCID-Profile: Paul Best, Ricard Marxer
Förderung
- Fundação Grupo Boticário de Proteção à Natureza
- John D. and Catherine T. MacArthur Foundation
- Society for Marine Mammalogy
- Agence Nationale de la Recherche
Zitieren
APA
Best, P., Paris, S., Glotin, H., & Marxer, R. (2023). Deep audio embeddings for vocalisation clustering. PLOS ONE. https://doi.org/10.1371/journal.pone.0283396
Vancouver
Best P, Paris S, Glotin H, Marxer R. Deep audio embeddings for vocalisation clustering. PLOS ONE. 2023. doi:10.1371/journal.pone.0283396
BibTeX
@article{Best2023,
author = {Best, Paul and Paris, Sébastien and Glotin, Hervé and Marxer, Ricard},
title = {Deep audio embeddings for vocalisation clustering},
journal = {PLOS ONE},
year = {2023},
doi = {10.1371/journal.pone.0283396}
}
EUVIMED