overview
What is BirdNET?
BirdNET is an AI-powered bioacoustics tool developed by the K. Lisa Yang Center for Conservation Bioacoustics at the Cornell Lab of Ornithology and Chemnitz University of Technology that enables researchers, conservationists, and citizen scientists to identify over 6,000 bird species from audio recordings. It processes raw acoustic data using deep learning algorithms, transforming soundscapes into standardized feature representations to isolate and identify subtle vocal signatures. The model is trained on thousands of hours of curated bird vocalizations from public collections like Xeno-canto and Macaulay Library. BirdNET provides open-source tools and models for large-scale biodiversity monitoring, supporting both mobile applications for real-time identification and command-line interfaces for high-throughput data processing. Recent developments include BirdNET-Cloud (July 2026) for always-on listening stations and BirdNET Live App (August 2026) for on-device, offline identification.
