Deep learning for passive acoustic monitoring: how to study changing
phenology in remote areas
- Sylvain Christin,
- Éric Hervet,
- Paul Smith,
- Ray Alisauskas,
- Dominique Berteaux,
- Glen Brown,
- Kyle Elliott,
- Jannik Hansen,
- Sandra Lai,
- Jean-François Lamarre,
- Richard Lanctot,
- Christopher Latty,
- Audrey Le Pogam,
- Douglas MacNearney,
- Vijay Patil,
- Jennie Rausch,
- Sarah Saalfeld,
- Niels Schmidt,
- Andrew Tam,
- François Vézina,
- Øystein Varpe,
- Paul Woodard,
- Glenn Yannic,
- Nicolas Lecomte
Paul Smith
Environment and Climate Change Canada National Wildlife Research Centre
Author ProfileDouglas MacNearney
Environment and Climate Change Canada National Wildlife Research Centre
Author ProfileJennie Rausch
Environment and Climate Change Canada Canadian Wildlife Service
Author ProfileSarah Saalfeld
US Fish and Wildlife Service Alaska Region
Author ProfilePaul Woodard
Environment and Climate Change Canada Canadian Wildlife Service
Author ProfileAbstract
Understanding how species adjust to seasonality is fundamental in
ecology, especially with rapidly increasing global air temperatures.
Bioacoustic monitoring offers promise for tracking shifts in seasonal
timing of vocal species, as recent automated sound recorders enable
large-scale and long-term data collection. Yet, analyzing vast datasets
necessitates automation and innovative detection methods. Here, we
introduce BioSoundNet, a deep learning model designed for bird
vocalization detection. Trained on field data and open-access databases,
BioSoundNet achieved AUC scores of 0.88-0.93 and average precisions of
0.87-0.97 across five datasets spanning various ecosystems, and
effectively captured the temporal patterns of avian acoustic activity at
different time scales. Our findings underline the importance of
evaluating models in ecological contexts and to address the potential
consequences of missing detections. Operating efficiently on standard
computers, BioSoundNet is a robust tool for automated bird vocalization
detection, providing a valuable resource for ecological phenology
studies and acoustic dataset analysis.09 Nov 2023Submitted to Ecology Letters 10 Nov 2023Submission Checks Completed
10 Nov 2023Assigned to Editor
10 Nov 2023Review(s) Completed, Editorial Evaluation Pending
20 Nov 2023Reviewer(s) Assigned