Academic Background

In 2022, Paul earned his PhD in computer science from the University of Toulon in France, where he developed computational methods to analyze passive acoustic data from marine mammals. His research lies at the intersection of methodological development and applied bioacoustics.

To address common challenges in bioacoustic analysis, including vocalization detection, repertoire description, and fundamental frequency estimation, he proposes deep-learning frameworks applicable across species and in low-data regimes. These tools have enabled him to study the structure and function of vocalizations in various species, including cultural trends in fin whale songs, complexity in orca call sequences, individuality in gibbon calls, and contact calling in pilot whales.

Current Research

Paul joined the biology cluster at ARI to conduct a comparative study of musicality in animal songs. This project proposes analyzing spontaneous songs produced in ecological contexts by humpback whales, humans, and gibbons. First, extracting melodic features automatically is crucial to standardize findings across species, populations, and individuals. This, in turn, will allow testing for fundamental components of pitch-based musicality and help identify the potential innate and cultural drivers of musical behavior.