You are here

BirdCLEF++

Tentative Timeline

- March 11, 2026 - Start Date.

- May 27, 2026 - Entry Deadline. You must accept the competition rules before this date to compete.

- May 27, 2026 - Team Merger Deadline. This is the last day participants may join or merge teams.

- June 3, 2026 - Final Submission Deadline.

All deadlines are at 11:59 PM UTC on the corresponding day unless otherwise noted. The competition organizers reserve the right to update the contest timeline if they deem it necessary.

Motivation

How do you protect an ecosystem you can’t fully see? One way is to listen.

This competition involves building models that automatically identify wildlife species from their vocalizations in audio recordings collected across the Pantanal wetlands. This work will support more reliable biodiversity monitoring in one of the world’s most diverse and threatened ecosystems.

Understanding how ecological communities respond to environmental change and restoration efforts is a central challenge in conservation science. The Pantanal — a wetland spanning 150,000+ km² across Brazil and neighboring countries — is home to over 650 bird species plus countless other animals, yet much of it remains unmonitored. Seasonal flooding, wildfires, agricultural expansion, and climate change make regular fieldwork challenging.

Task Description

Conventional biodiversity monitoring across vast, remote regions is expensive and logistically demanding. To help address these challenges, a growing network of 1,000 acoustic recorders is being deployed across the Pantanal, running continuously to capture wildlife sounds across different habitats and seasons. Continuous audio recording allows researchers to capture multi-species soundscapes over extended periods, providing a community-level perspective on biodiversity dynamics. But the sheer volume of audio is too large to review manually, and labeled species data is limited.

This competition focuses on the development of machine learning models that identify wildlife species from passive acoustic monitoring (PAM). Proposed approaches should work across different habitats, withstand the constraints of messy, field-collected data, and support evidence-based conservation decisions. Successful solutions will help advance biodiversity monitoring in the last wild places on Earth, including research initiatives in the Pantanal wetlands of Brazil.

Listening carefully, and at scale, may be one of the most effective tools available to protect this landscape.

Kaggle competition

This competition will be held entirely on Kaggle: https://www.kaggle.com/competitions/birdclef-2026/overview

Publication Track

We encourage participants to submit a working note write-up of their approach to the CLEF conference. Organizers will award $5,000 in prize money ($2,500 each) for the two best working note submissions.

Submission dates are:

- June 17, 2026 - Working note submission deadline
- June 24, 2026 - Notification of acceptance
- July 6, 2026 - Camera-ready submission deadline

Additional information on the submission process will be posted ahead of time on the discussion forum.

Acknowledgements

The development of the competition dataset was supported by the Bezos Earth Fund AI for Climate and Nature Grand Challenge.

Compiling this extensive dataset was a major undertaking, and we are very thankful to the many domain experts who helped to collect and manually annotate the data for this competition. Specifically, we would like to thank (institutions and individual contributors in alphabetic order):

Chemnitz University of Technology
Stefan Kahl, Mario Lasseck, and Maximilian Eibl

Google Deepmind
Tom Denton

iNaturalist
Grant van Horn

Instituto Homem Pantaneiro
Wener Hugo Arruda Moreno

Instituto Nacional de Pesquisa do Pantanal (INPP)
Carolline Zatta Fieker, Karl-L. Schuchmann, Kirk Thiago Pedroso Azevedo, Lucas Korzune Sampaio Teles, Marinez Isaac Marques and Matheus Gonçalves dos Reis

K. Lisa Yang Center for Conservation Bioacoustics
Stefan Kahl, Larissa Sugai and Holger Klinck

LifeCLEF
Alexis Joly and Henning Müller

Sauá Consultoria Ambiental
Carolina Martins Garcia

Universidade Federal de Mato Grosso do Sul (UFMS)
Alyson Vieira de Melo, Daiene Louveira Hokama Sousa, José Luiz Massao Moreira Sugai, João Emílio de Almeida Júnior, Liliana Piatti, Mariana Motti Barbosa, Matheus de Oliveira Neves, Priscila do Nascimento Lopes and Ryan Christopher Kridler

Xeno-canto
Willem-Pier Vellinga, Bob Planqué