Bioacoustic fundamental frequency estimation: a cross-species dataset and deep learning baseline

The fundamental frequency (F0) is a key parameter for characterising structures in vertebrate vocalisations, for instance defining vocal repertoires and their variations at different biological scales ( <i>e.g</i> population dialects, individual signatures). However, the task is too laborious to perform manually, and its automation is complex. Despite significant advancements in the fields of speech and music for automatic F0 estimation, similar progress in bioacoustics has been limited. To address this gap, we compile and publish a benchmark dataset of over 250,000 calls from 14 taxa, each paired with ground truth F0 values. These vocalisations range from infra-sounds to ultra-sounds, from high to low harmonicity, and some include non-linear phenomena. Testing different algorithms on these signals, we demonstrate the potential of neural networks for F0 estimation, even for taxa not seen in training, or when trained without labels. Also, to inform on the applicability of algorithms to analyse signals, we propose spectral measurements of F0 quality which correlate well with performance. While current performance results are not satisfying for all studied taxa, they suggest that deep learning could bring a more generic and reliable bioacoustic F0 tracker, helping the community to analyse vocalisations via their F0 contours.

Authors

Institutions

Publication Details

Journal
Bioacoustics
Published
2025-06-02
DOI
https://doi.org/10.1080/09524622.2025.2500380
Primary Topic
Animal Vocal Communication and Behavior
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Bioacoustic fundamental frequency estimation: a cross-species dataset and deep learning baseline

Robert C. Liu, Ricard Marxer, Marie A. Roch, Pavel Linhart et al.
Bioacoustics
Animal Vocal Communication and Behavior
article

Bioacoustic fundamental frequency estimation: a cross-species dataset and deep learning baseline

Robert C. Liu, Ricard Marxer, Marie A. Roch, Pavel Linhart, Paul Best, Marcelo Araya‐Salas, Arik Kershenbaum, Axel G. Ekström, Matthew Wijers, Adriano R. Lameira, Kenna D. S. Lehmann, Holly Root‐Gutteridge, Megan R. Warren, Ariana Strandburg‐Peshkin, Martin Šálek, Andrew Markham, Grace Smith-Vidaurre, Frants Havmand Jensen, Bárbara Freitas, Malavika Madhavan
article en

Abstract

The fundamental frequency (F0) is a key parameter for characterising structures in vertebrate vocalisations, for instance defining vocal repertoires and their variations at different biological scales ( <i>e.g</i> population dialects, individual signatures). However, the task is too laborious to perform manually, and its automation is complex. Despite significant advancements in the fields of speech and music for automatic F0 estimation, similar progress in bioacoustics has been limited. To address this gap, we compile and publish a benchmark dataset of over 250,000 calls from 14 taxa, each paired with ground truth F0 values. These vocalisations range from infra-sounds to ultra-sounds, from high to low harmonicity, and some include non-linear phenomena. Testing different algorithms on these signals, we demonstrate the potential of neural networks for F0 estimation, even for taxa not seen in training, or when trained without labels. Also, to inform on the applicability of algorithms to analyse signals, we propose spectral measurements of F0 quality which correlate well with performance. While current performance results are not satisfying for all studied taxa, they suggest that deep learning could bring a more generic and reliable bioacoustic F0 tracker, helping the community to analyse vocalisations via their F0 contours.

BioacousticsVol. 34(4)
Université Toulouse III - Paul Sabatier (FR), Consejo Superior de Investigaciones Científicas (ES), Université de Toulon (FR), Emory University (US), Emory and Henry College (US), University of Konstanz (DE), Aarhus University (DK), Czech University of Life Sciences Prague (CZ), Aix-Marseille Université (FR), University of Cambridge (GB), San Diego State University (US), Universidad de Costa Rica (CR), University of Warwick (GB), University of Oxford (GB), Max Planck Institute of Animal Behavior (DE), Museo Nacional de Ciencias Naturales (ES), Wildlife Conservation Society United Kingdom (GB), Forestry and Game Management Research Institute (CZ), Czech Academy of Sciences, Institute of Vertebrate Biology (CZ), Science Oxford (GB), University of Lincoln (GB), University of Neuchâtel (CH), Universidad Autónoma de Madrid (ES), KTH Royal Institute of Technology (SE), Michigan State University (US)
National Science Foundation, Human Frontier Science Program, Agence Nationale de la Recherche, Grantová Agentura České Republiky, Carlsbergfondet, Vetenskapsrådet, Fundação para a Ciência e a Tecnologia, Aarhus Institute of Advanced Studies, Aarhus Universitet, National Institute of Mental Health, National Institute on Deafness and Other Communication Disorders
Openalex Percentile: Top 88%
Animal Vocal Communication and Behavior
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.