🐷 AI Will Help to Understand animals and Take Better Care of Them
Scientists from the University of Copenhagen have trained AI to distinguish between positive and negative emotions in seven different ungulate species—cows, sheep, horses and Przewalski's horses, wild boars, pigs, and goats. This is the first study to use algorithms to study the language and feelings of multiple species at once.
"It has the potential to revolutionise animal welfare, livestock management, and conservation, allowing us to monitor animals' emotions in real time," says Élodie F. Briefer, Associate Professor at the University of Copenhagen and the study author.
Animals can't tell us how they feel with words. On farms, in zoos, or in natural parks, determining that an animal is unwell can be done through hormone-level indicators in its blood or by observing indirect signs, such as a lack of appetite, reduced movement, and other behavioral changes.
It's often too late to help them, as the disease might have spread too far. Better understanding the animals' "language" will help save them from unnecessary suffering and improve their well-being.
The model was trained on several thousand recorded vocalizations produced by these species of ungulates in different emotional states. Algorithms analyzed them by pitch, frequency, volume, and duration. AI can distinguish between positive and negative states with nearly 90% accuracy.
It turns out that, for example, horses' positive and negative emotions can be quite easily distinguished through sound, thanks to their rich and prolonged calls. In contrast, deciphering cows' emotions through mooing is much more challenging. But all ungulates have common patterns of joy expression: quieter, shorter, and with less variation in pitch.
🐳 This study is not the only example of AI use in bioacoustics. Last year, researchers from MIT, in collaboration with Project CETI, developed an AI model capable of predicting various "codes"—the sounds produced by sperm whales. Researchers from Tel Aviv University used an AI-powered voice-recognition program to analyze thousands of bat calls. They found that these animals have different communication styles. For example, mothers use a "baby language" with cubs.
How AI is helping to understand pets:
🐕 AI Collar Helps... Talk to Your Dog
