There are tens of thousands of plants, animals and fungi in the Netherlands alone. Learning how to distinguish them can seem like a mountain climb. Apps that can identify species with Artificial Intelligence can lower this threshold enormously. For example, there is ObsIdenfity for photos of Dutch and European species. For bird sounds there is Birdnet and Merlin. However, birds are not the only animal species that make noise: many of the Dutch locusts and crickets produce noise, which differs per species. Crickets make whistling sounds, grasshoppers usually make more scratching, ticking or rattling noises.

To make learning these sounds much more accessible, Aquila Ecologie, in collaboration with the Province of Limburg, has focused on making an Android app that uses AI to distinguish the sounds. The 35 species that the app can recognize are distinguished with 90% accuracy, which is exceptionally high. When it becomes available, the app will show pictures of the species and a text with basic info when it is recognized.

With the accurate recognition of sounds, it suddenly becomes possible to analyze long recordings and recognize all types in them. All kinds of research can be done into the occurrence and activity of locusts that were previously not possible.

The accuracy of the recognition is due to an algorithm developed by Aquila Ecologie specially for this purpose. Where birds are recognized on the basis of spectrograms (visualization of the pitches over time), the algorithm of Aquila Ecologie works on the basis of the raw sound data. Developing such an algorithm is a lot of work: there are infinitely many different possibilities of different layers of processing that eventually lead to a recognition. But it is also the challenge to stack the right layers to get the best possible results.

The app for Android will probably be available around June 2024.