Chemnitz University of Technology Develops New App for Animal Sound Recognition

Most animal call apps require an internet connection

Previous animal call recognition apps often reached their limits in practical use. Without a stable internet connection, real-time identification in the forest or high mountains was impossible. The new app solves this problem through technology. The underlying AI models have been optimized, downsized, and accelerated to such an extent that they can now run directly on a smartphone.   

Animal species are identified in 25 languages

Birdnet live can acoustically identify nearly 10,000 animal species—more than any other app, emphasized lead developer Stefan Kahl from the Department of Media Informatics at Chemnitz University of Technology. Currently, the app’s user interface is available in seven languages. The names of the identified animal species are translated into 25 languages.

According to the university, the app is primarily aimed at nature conservation experts, researchers, students, as well as ambitious citizen scientists and amateur ornithologists. The app automatically records animal sounds with precise location and time data and stores them locally on the smartphone. The recordings can be reviewed later, exported, or shared with other experts for validation. The German Federal Environmental Foundation funded the development team.