The cat family Felidae is one of the most successful carnivore lineages today. However, the study of the evolution of acoustic communication between felids remains a challenge due to the lack of fossils, the limited availability of audio recordings because of their largely solitary and secretive behavior, and the underdevelopment of computational models and methods needed to address acoustic evolutionary questions. This study is a first attempt at developing a machine learning-based approach to the classification of felid calls as well as the identification of acoustic features that distinguish felid call types and species from one another. A felid call dataset was developed by extracting audio clips from diverse sources. The audio clips were manually annotated for call type and species. Due to the limited availability of samples, this study focused on the Pantherinae subfamily. Time-frequency features were then extracted from the Pantherinae dataset. Finally, several classification algorithms were applied to the resulting data. We achieved 91% accuracy for this Pantherinae call type classification. For the species classification, we obtained 86% accuracy. We also obtained the most predictive features for each of the classifications performed. These features can inform future research into the evolutionary acoustic analysis of the felid group.

No abstract