BirdNET – Bird song identification
BirdNET is the result of a joint effort of Cornell Ornithology Laboratory (USA) andChemnitz Technological University (Germany). The BirdNET research project uses artificial intelligence and neural networks to identify nearly 1,000 of the most common species in Europe and North America.
How can computers learn to recognize birds from sounds? Research is mainly focused on the detection and classification of bird sounds using machine learning, to help experts and citizen scientists in their work in monitoring and protecting birds.
Characteristics :
– Record a bird song with the microphone of your Android mobile device.
– Wait, and see if BirdNET correctly identifies the species of bird.
– Response time may be affected by the need to transfer audio recordings to BirdNET servers for the purpose of processing files.
– The quality of the recording influences the results and will vary depending on your device.
– An external microphone will probably improve the recording quality.
A live demo, for analysis of audio recordings, is available at https://birdnet.cornell.edu/live