AI identifies bird species via bioacoustic analysis.

By listening to animal calls, the system can measure biodiversity more efficiently than traditional methods.

 Conifer, Plant, Tree, Vegetation, Yew, Land, Nature, Outdoors, Woodland, Garden

Researchers at the University of Würzburg have developed a computer-based system that can identify bird species in rainforests by analysing their calls. This breakthrough in bioacoustic analysis has the potential to revolutionize the measurement of biodiversity, making the process more efficient and accurate.

Traditionally, ecologists measured biodiversity by searching for tracks or spoor in the undergrowth, a time-consuming and expert-dependent method. The researchers aimed to automate this process using artificial intelligence (AI). They collected sound recordings from 43 sites in the Ecuadorian rainforest, ranging from old-growth forests to areas cleared for agriculture. The longer a plot had been free from agricultural activity, the greater the biodiversity it hosted.

Initially, an expert manually identified the species based on the recorded calls, creating a species list. Then, the researchers fed the recordings into AI models trained using sound samples from other parts of Ecuador. The models successfully identified 75 bird species solely from their calls. Remarkably, the AI tools demonstrated accuracy comparable to human experts.

The researchers also captured night-flying insects using light-traps and conducted DNA analysis to identify them. They found that the diversity of noisy animals served as a reliable proxy for the diversity of quieter ones.

Beyond ecological applications, this computer-based bioacoustic analysis could be valuable for companies engaged in forest restoration projects. Firms like L’Oréal and Shell, responding to customer demands, have invested in such initiatives globally. The automated approach developed by the researchers could provide a standardized method for monitoring and assessing the effectiveness of these projects.

Source: The Economist