WILDLABS logoWildlabs.net : The conservation technology network

WILDLABS logoWildlabs.net : The conservation technology network

discussion / Acoustics  / 6 April 2026

Looking for opportunities in AI for Bioacoustics and Environmental Monitoring

Leonardo Mannini

Leonardo Mannini

AI researcher specializing in bioacoustic signal analysis, multi-species classification, and embedded audio monitoring systems.

Groups

Hi everyone,

My name is Leonardo Mannini, and I’m an AI engineer with a strong interest in bioacoustics and conservation technology.

I recently completed a research role at FBK (Fondazione Bruno Kessler), where I developed a multi-species bird sound classification system using machine learning. The project involved training custom models on labeled audio datasets, applying techniques such as data augmentation and knowledge distillation, and deploying lightweight models on low-power devices like AudioMoth.

I am now looking for opportunities to continue working in this field — either as a research engineer, AI engineer, or developer, in projects related to bioacoustics, environmental monitoring, biodiversity assessment, or acoustic data analysis.

I am particularly interested in roles where machine learning can support ecological research, and I would also be happy to contribute to collaborations or ongoing projects.

If anyone knows of open positions, collaborations, or teams working in this space, I would be very grateful to connect.

Best,  
Leonardo



Chris Nicholas
@chrisgnicholas  | he/him/it
Society for Conservation GIS
IT 'Forest Gump' veteran in remote sensing/location services/XR/IoT ML/Agentic AI
Conversation starter level 2
Poster level 1
Commenter level 2

7 April 2026 3:26am

nothing paid; , but I've got two problems: 

1) I collected several hours of Audiomoth recording from the Los Amigos Biological Station at night near a mosquito and frog-filled pond, at a very high sampling rate. Supposedly there are > 20 different kinds of bats there, and dozens of different frogs, that were just going nuts. I was going to just try processing with BattyBirdNet to see if anything from North America was even close for the bats; I have no idea how to process them for South American frogs. Any help, pointers, appreciated

2) as per this thread in the Edge Computing thread, working on a remote compute "appliance" device, that can be remotely administered/upgraded, and can routinely crunch vast amount of locally-collected data (bioacoustics, drone, camera trap, etc) stuff on-site, because an Audiomoth or two collecting 64G SD cards full of data quickly overwhelms a Starlink device.  Starting with Birdnet-analyze and BattyBirdnet, looking to build out that functionality