Researchers at the University of Michigan, in collaboration with the National Institute of Astrophysics, Optics and Electronics in Puebla, Mexico, are exploring how artificial intelligence (AI) can help decode dog vocalisations.

This innovative study aims to distinguish between different types of vocalisations, such as playful barks or aggressive growls, and could even identify characteristics like age, breed, and sex of the dog.

AI models, originally developed for human speech, are being adapted to interpret animal communication, offering a new approach to understanding our canine companions.

The findings were presented at the Joint International Conference on Computational Linguistics, Language Resources, and Evaluation.

The study was conducted with a dataset of barks from 74 dogs in Tepic and Puebla, Mexico. The dogs, aged between 5 to 84 months, included breeds like Chihuahuas, French Poodles, and Schnauzers.

Researchers recorded the dogs’ vocalisations in their natural environments, capturing authentic responses to various stimuli, including encounters with strangers, playful interactions, affectionate owners, and simulated attacks.

By using advanced AI technology to analyse these vocalisations, the research aims to improve human-dog communication and enhance animal welfare by better understanding their needs and emotions, ultimately strengthening the bond between humans and their pets.

#MaltaDaily