New deep learning AI tool helps ecologists monitor rare birds through their songs

Medium close-up photo of a bird on a rock (a Lapland longspur in white, grey and beige, with red and black spots on the head) with its beak opened (singing or chirping). Bird and rock occupy the right half of the image while the rest of it shows a homogeneous cloudy sky in lavender hue (suggesting sunrise or sunset).
(Photo credit: Nicolas Lecomte)

Researchers at the Université de Moncton have developed a new deep learning artificial intelligence tool that generates life-like birdsongs to train bird identification tools, helping ecologists to monitor rare species in the wild. The findings are presented in the British Ecological Society journal Methods in Ecology and Evolution.

Identifying common bird species through their song has never been easier, with numerous phone apps and software available to both ecologists and the public. But what if the identification software has never heard a particular bird before, or only has a small sample of recordings to reference? This is a problem facing ecologists and conservationists monitoring some of the world’s rarest birds.

To overcome this problem, researchers at the Université de Moncton (Canada) have developed ECOGEN, a first-of-its kind deep learning tool that can generate life-like bird sounds to enhance the samples of underrepresented species. These can then be used to train audio identification tools used in ecological monitoring, which often have disproportionately more information on common species.

The work is part of the MSc thesis of Axel-Christian Guei, who is supervised by professors Nicolas Lecomte and Éric Hervet. Sylvain Christin, PhD candidate with Nicolas Lecomte, was also central to the development of this AI tool.

Dr Lecomte, one of the lead researchers, explains: “Due to significant changes in animal populations worldwide, there is an urgent need for automated tools, such as acoustic monitoring, to track shifts in biodiversity. However, the AI models used to identify species in acoustic monitoring lack comprehensive reference libraries.”

“With ECOGEN, you can address this gap by creating new instances of bird sounds to support AI models. Essentially, for species for which limited recordings in the wild exist, such as those that are rare, elusive, or sensitive, you can expand your sound library without further disrupting the animals or conducting additional fieldwork.”

The researchers say that creating synthetic bird songs in this way can contribute to the conservation of endangered bird species and also provide valuable insight into their vocalisations, behaviours and habitat preferences.

The ECOGEN tool has other potential applications. For instance, it could be used to help conserve extremely rare species, like 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.

The tool could benefit other types of animals as well. “While ECOGEN was developed for birds, we’re confident that it could be applied to mammals, fish (yes they can produce sounds!), insects and amphibians,” adds Dr Lecomte.

As well as its versatility, a key advantage of the ECOGEN tool is its accessibility, since it is open source and can be used on even basic computers.

ECOGEN works by converting real recordings of bird songs into spectrograms (visual representations of sounds) and then generating new AI images from these to increase the dataset for rare species for which there are few recordings. These spectrograms are then converted back into audio to train bird sound identification tools. In this study, the researchers used a dataset of 23,784 wild bird recordings from around the world, covering 264 species.

Article: Guei, A.-C., Christin, S., Lecomte, N., & Hervet, É. (2023). ECOGEN: Bird sounds generation using deep learning. Methods in Ecology and Evolution, 00, 1–11.

This article was adapted and republished with permission from the Université de Moncton and the British Ecological Society.

Up next

From student to astronaut: developing the skills to understand the world

Postsecondary students and postdoctoral researchers play a crucial role in the Canadian research ecosystem by pushing back the boundaries of knowledge.

Contact Newsletter

Get highlights of things happening at NSERC delivered to your email inbox. View all Newsletters

  • Twitter
  • Facebook
  • LinkedIn
  • Youtube
  • Instagram