Optimization of passive acoustic bird surveys: a global assessment of <scp>BirdNET</scp> settings

BirdNET is a popular machine learning tool for automated recognition of bird sounds. However, evidence on how to optimize its settings for accurate bird monitoring remains limited. Here, we evaluate how BirdNET settings influence model performance in identifying bird vocalizations and characterizing bird communities, using 4224 1‐min recordings from 67 recording locations worldwide. Giving equal importance to recall and precision, a low confidence score threshold (0.1–0.3) appears optimal for detecting bird vocalizations, whereas higher thresholds (around 0.5) are more suitable for characterizing bird communities. Based on our findings, we recommend increasing the Overlap parameter from its default value of 0 to 2 s, as this consistently improves BirdNET performance in detecting both bird vocalizations and species presence. The effect of the Sensitivity parameter varied across regions. However, a value of 0.5 maximizes global performance for community‐level analyses across all confidence thresholds, and a value of 1.5 generally yields better results for vocalization‐level studies, particularly at low confidence thresholds. Our findings offer practical guidance for selecting BirdNET settings in passive acoustic bird surveys, enhancing both the identification of bird vocalizations and the characterization of bird communities.

Authors

Institutions

Publication Details

Journal
Ibis
Published
2025-11-24
DOI
https://doi.org/10.1111/ibi.70013
Citations
3
Primary Topic
Animal Vocal Communication and Behavior
Type
article
Field-Weighted Citation Impact
4.92
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Optimization of passive acoustic bird surveys: a global assessment of BirdNET settings

Christos Astaras, Marina H. L. Duarte, Alberto González‐Romero, Jörn Theuerkauf et al.
3 citations
Ibis
Animal Vocal Communication and Behavior
4.92
article

Optimization of passive acoustic bird surveys: a global assessment of BirdNET settings

Christos Astaras, Marina H. L. Duarte, Alberto González‐Romero, Jörn Theuerkauf, Eladio L. García de la Morena, Alice Calvente, Aaron S. Weed, Andrea L. Aguilera, Manuel B. Morales, Juan Pablo Zurano, Cristian Pérez‐Granados, Gabriel Marcacci, Mariano J. Feldman, Estanislao Aguayo Navarrete, Ana Rainho, Andrey Atemasov, Cássio Rachid Meireles de Almeida Simões, Luc Barbaro, Julian D. Olden, Agudelo Muñoz, Fernando González‐García, Sergio Chaparro‐Herrera, Amanda K. Navine, Eben Goodale, Karl‐Ludwig Schuchmann, Anja Hutschenreiter, Igor V. Palko, Porfirio Nava, Tomasz S. Osiejuk, Louis Imbeau, Sara Bombaci, Renata S. Sousa‐Lima, Diego Llusia, Walmir da Silva, Erika Santamaría, Juán Traba, Gérard Bota, Juan A. Medina Méndez, Alan Monroy‐Ojeda, Sandra Müeller, Thiago Pinheiro, Natalia dos Santos Saturnino, Pablo Acebes, Esther Sebastián‐González, Alfredo Attisano, Diego� Varela, Alain Jacot, Yi‐Ru Cheng, Alexandre Antonelli, Samuel R. Silva, Nick A. Littlewood, Alba Márquez‐Rodríguez, Cicero Simão Lima‐Santos, Mao‐Ning Tuanmu, Ilias Karmiris, Ingrid M. D. Torres, Jon Morant, Christos Mammides, Carlos Barros de Araújo, Mauricio Díaz‐Vallejo, Marinêz Isaac Marques, Sebastian Kepfer‐Rojas, Gabriel L. M. Rosa, Facundo G. Di Sallo, Maxime Cauchoix, Anant Deshwal, Michał Budka, Elena Gangenova, Junior A. Tremblay, Oscar Humberto Marín‐Gómez, Irene Mendoza, Robert A. Briers, David Funosas, Mauro Pichorim, Miguel A. Mohedano-Muñoz, Ana Silvia de Oliveira Tissiani, Svetlana S. Gogoleva, Adrián Barrero, Raiane Vital da Paz, Mario Fernández‐Tizón, Gustavo A. Zurita, Patrick J. Hart, Carlos Salustio‐Gomes, Dorgival Diógenes Oliveira‐Júnior, Giordano Jacuzzi, Paula Lopes, N.M.D. Brown, Chao He, Dominik Arend, Manu Santa‐Cruz, Hugo Sánchez‐Mateos, Ana L. C. Catalano, Michaelangelo Pontikis, Anamaria Dal Molin, Hiếu V. Trong, Giulia Bastianelli, Pooja Panwar, José M. De la Peña‐Rubio
article en
3 citations

Abstract

BirdNET is a popular machine learning tool for automated recognition of bird sounds. However, evidence on how to optimize its settings for accurate bird monitoring remains limited. Here, we evaluate how BirdNET settings influence model performance in identifying bird vocalizations and characterizing bird communities, using 4224 1‐min recordings from 67 recording locations worldwide. Giving equal importance to recall and precision, a low confidence score threshold (0.1–0.3) appears optimal for detecting bird vocalizations, whereas higher thresholds (around 0.5) are more suitable for characterizing bird communities. Based on our findings, we recommend increasing the Overlap parameter from its default value of 0 to 2 s, as this consistently improves BirdNET performance in detecting both bird vocalizations and species presence. The effect of the Sensitivity parameter varied across regions. However, a value of 0.5 maximizes global performance for community‐level analyses across all confidence thresholds, and a value of 1.5 generally yields better results for vocalization‐level studies, particularly at low confidence thresholds. Our findings offer practical guidance for selecting BirdNET settings in passive acoustic bird surveys, enhancing both the identification of bird vocalizations and the characterization of bird communities.

IbisVol. 168(2)
Dartmouth College (US), Universidad de Valladolid (ES), Universidade Federal da Bahia (BR), Universidade Estadual de Londrina (BR), Centre National de la Recherche Scientifique (FR), University of Alicante (ES), National Park Service (US), Environment and Climate Change Canada (CA), Université Toulouse III - Paul Sabatier (FR), University of Quindío (CO), Universidad del Norte (CO), Universidad de San Carlos de Guatemala (GT), University of Freiburg (DE), Universidade Federal da Paraíba (BR), Pontifícia Universidade Católica de Minas Gerais (BR), Copenhagen Business School (DK), Université du Québec en Abitibi-Témiscamingue (CA), University of Washington (US), Institut National Polytechnique de Toulouse (FR), Bradley University (US), Edinburgh Napier University (GB), Scotland's Rural College (GB), Colorado Parks and Wildlife (US), Swiss Ornithological Institute (CH), BirdLife Australia (AU), Royal Botanic Gardens, Kew (GB), BirdLife International (KE), Portuguese Army (PT), Museum and Institute of Zoology (PL), Instituto Federal de Educação Ciência e Tecnologia da Paraíba (BR), Universidade Federal de Mato Grosso (BR), University of Cyprus (CY), Universidade Federal do Rio Grande do Norte (BR), University of Hawaii at Hilo (US), Universidad de Antioquia (CO), Frederick University (CY), Universidad de Cádiz (ES), University of Oxford (GB), Institut National de Recherche pour l'Agriculture, l'Alimentation et l'Environnement (FR), Estación Biológica de Doñana (ES), The Nature Conservancy (CO), Sustainability Institute (ZA), Institute for Biodiversity (DE), Zoological Research Museum Alexander Koenig (DE), Eurofins (Netherlands) (NL), Instituto de Investigación en Recursos Cinegéticos (ES), Loufakis (Greece) (GR), Forest Science and Technology Centre of Catalonia (ES), Centre for Research on Ecology and Forestry Applications (ES), Forest Research Institute (BG), Instituto de Recursos Naturales y Agrobiología de Sevilla (ES), Biodiversity Research Center, Academia Sinica (TW), Science Oxford (GB), Instituto de Biología Subtropical (AR), International Center for Tropical Agriculture (VN), Universidad Autónoma Latinoamericana (CO), Instituto de Ecología (MX), Génétique Physiologie et Systèmes d'Elevage (FR), Centro Universitário de João Pessoa (BR), International Union for Conservation of Nature (CH), Hospital Virgen del Puerto (ES), Station d’Écologie Théorique et Expérimentale (FR), Bronberg Observatory (ZA), Imperial College London (GB), University of Salford (GB), Adam Mickiewicz University in Poznań (PL), Universidad Autónoma de Madrid (ES), Xi’an Jiaotong-Liverpool University (CN), Universidad de Sevilla (ES), V. N. Karazin Kharkiv National University (UA), Universidad Politécnica de Madrid (ES), University of Gothenburg (SE), Universidad Autónoma del Estado de Morelos (MX), Colorado State University (US), Polish Academy of Sciences (PL)
Openalex Percentile: Top 5%
Animal Vocal Communication and Behavior
4.92
DOISourceOpenAlex
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.