How can computers learn to recognize birds from sounds? The BirdNET research project uses artificial intelligence
and neural networks to train computers to identify nearly 1,000 of the most common species of North America and Europe.
Help us test this prototype app by recording a file using the internal microphone of your Android device and seeing if
BirdNET correctly identifies the probable bird species present in your recording.
Please note: Response time is affected by the need to transfer the audio recordings to our servers in order to process the files. Recording quality may affect results, and will vary depending on your device. External microphones will probably increase the recording quality.
BirdNET is a joint project of the Cornell Lab of Ornithology and Chemnitz University of Technology.
Seems to work well, first IDs spot on, but sometimes is busy, and won't submit recording. It also won't accept input from external or earlier recordings, which means many opportunities are missed. When there's a lot of different bird song it can be difficult to select the required area of the spectrogram. Annoying that it starts recording when the app is opened, it doesn't wait for the MIC to be pressed as in the tutorial.
Fabulous job. Not just another mickey mouse app able to identify sparrows. For example, I saw a bird which was either a Cirl Bunting or a Serin and this app confirmed the latter. Top job!
The specifications on this app or amazing and I have not found any fault in any of the analyzed acordes I have wished to identify the birds on this is just afraid of me the app for any birds lover
Tweet chirp humm grackle
One of the best apps ever! For bird lovers is a must, specially if you want them identify the bastard that keeps singing at 6 a.m!