A newly developed deep-learning AI tool, that generates life-like bird vocalisations to train bird identification tools, is helping ecologists to monitor rare species.
In 2023, identifying birds via AI software – such as Cornell's Merlin app – has never been easier. However, when it comes to some rarer species, there are few samples of their vocalisations, and this prompted researchers at the University of Moncton, Canada, to develop ECOGEN.
ECOGEN is a first-of-its-kind deep-learning tool that can generate lifelike bird sounds to enhance the samples of underrepresented species. These realistic sounds created by AI can then be used to train audio identification tools used in ecological monitoring.
ECOGEN could be used to help Critically Endangered species such as Regent Honeyeater (via Wikimedia).
Researchers found that adding artificial birdsong samples generated by ECOGEN to a birdsong identifier improved the bird song classification accuracy by 12% on average. These findings have been presented in a paper in Methods in Ecology and Evolution.
The researchers say that creating lifelike bird songs in this way can contribute to the conservation of endangered species and also provide insight into their vocalisations, behaviours and habitat preferences.
Furthermore, ECOGEN could be used to help conserve extremely rare species, such as the Critically Endangered Regent Honeyeater, where young individuals are unable to learn their species' songs because there aren't enough adult birds to learn from.
Christin, S, Guei, A-C, Hervet, É, & Lecomte, N. 2023. ECOGEN: Bird sounds generation using deep learning. Methods in Ecology and Evolution. DOI: https://doi.org/10.1111/2041-210X.14239