BirdNET, a free machine-learning tool that can identify more than 3,000 birds through sound alone, generates reliable scientific data and makes it easy for people to contribute citizen science data on birds by simply recording sounds.
Article published on June 28eIn the open access journal Biology Plus By Connor Wood and colleagues at the K Center. Users simply listen to birds and tap the app to record. BirdNET uses artificial intelligence to automatically identify species by sound and capture the recording for use in research.
“Our guiding principles were that we needed a precise algorithm and a simple user interface,” said study co-author Stefan Kahl of the Yang Center at Cornell Lab, who led the study. Otherwise, users will not return to the application. The results have exceeded expectations: Since its launch in 2018, more than 2.2 million people have contributed data.
To test whether the app can generate reliable scientific data, the authors selected four test cases in which traditional research has already provided strong answers. Their results show that data from BirdNET accurately reproduced known patterns of North American song dialects and songbirds and mapped the migrations of birds across the two continents.
Verifying the reliability of the app’s data for research purposes was the first step in what they hope will be a long-term global research effort — not just for birds, but ultimately for all wildlife and even entire soundscapes. The data used in the four test cases is publicly available, and the authors are working to unlock the dataset.
“The most exciting part of this work is how easily people can get involved in bird research and conservation,” Wood adds. “You don’t need to know anything about birds, you just need a smartphone, and the BirdNET app can provide you and the research team with a prediction of the bird you’ve heard. This has led to a huge worldwide sharing, resulting in an enormous wealth of data. It is Truly a testament to the enthusiasm for birds that unites people from all walks of life.
BirdNET is part of the Cornell Lab of Ornithology’s suite of tools, including educational app Merlin Bird ID and citizen science apps eBird, NestWatch and Project FeederWatch, which together have produced more than one billion bird views, sounds, and images from participants around the world for use in science and conservation .
Material provided by OLP. Note: Content can be modified according to style and length.
“Music guru. Incurable web practitioner. Thinker. Lifelong zombie junkie. Tv buff. Typical organizer. Evil beer scholar.”