We’ve created Bird@Edge, an Edge AI system for recognizing bird species in audio recordings to support real-time biodiversity monitoring. Bird@Edge is based on embedded edge devices operating in a distributed system to enable eﬀicient, continuous evaluation of soundscapes recorded in forests.
There are multiple ESP32-based microphones (called Bird@Edge Mics) stream audio to a local Bird@Edge Station, on which bird species recognition is performed. The results of several Bird@Edge Stations are transmitted to a backend cloud for further analysis, e.g., by biodiversity researchers.
To recognize bird species in soundscapes, a deep neural network based on the EﬀicientNet-B3 architecture is trained and optimized for execution on embedded edge devices and deployed on a NVIDIA Jetson Nano board using the DeepStream SDK.
We have some ressources available online if you are interested:
Down below are some pictures of the devices we’ve build for outdoor usage.
We’d be happy if you’d try our systems yourself!
Bird@Edge Mic including solar panel (left), Bird@Edge Station including battery box (right).
Close-up of the Bird@Edge Mic’s internals.