This R&D project presents the Nocturnal Bird Migration (NBM) dataset associated machine learning model.
The dataset is a collection of precise time and frequency annotations13,359 annotated vocalizations from 117 species of the Western Palearctic, gathered by dozens of bird enthusiasts across France, enabling novel downstream acoustic analysis.
The two-stage object detection model shows competitive accuracy on the 45 main species of the dataset with state-of-the-art systems trained on much larger audio collections.
All data and code are made openly available.
More information: https://arxiv.org/abs/2412.03633
Website
External Resources
Conservation Challenges
Testing locations
WILDLABS Content
Have you created any WILDLABS discussions, resources or events relevant to this content? Please connect them to the Inventory!
Tto link relevant WILDLABS content to appear on Inventory pages, simply edit or create your post on WILDLABS, find the 'Related Inventory Content' field in the editing form, and then search for the names of related Products, Projects or Organisations on the Inventory.
Once added, the related Inventory content will appear at the bottom of your WILDLABS post and vice versa!
See page 12 of our FAQ document to learn how to link your posts to The Inventory!
Gallery
Contacts
Claim this R&D project
Are you involved with NBM - A project to automate the tracking of nocturnal bird migration in Europe? If you'd like to claim this R&D project, let us know.
Flag this R&D project
If you'd like to report an issue with this R&D project or profile, please let us know.