Reviewed by Lexie CornerFeb 24 2025
According to a study published in iScience, researchers from the University of Copenhagen's Department of Biology successfully trained a machine-learning model to distinguish between positive and negative emotions in seven different ungulate species, including cows, pigs, and wild boars.
Wild boar with piglets. Image Credit: Anne-Laure Maigrot
Can artificial intelligence help us understand animal emotions? A pioneering investigation reveals that the answer is yes. By analyzing the acoustic patterns of their vocalizations, the model achieved an astonishing 89.49 % accuracy, making it the first cross-species study to detect emotional valence using AI.
This breakthrough provides solid evidence that AI can decode emotions across multiple species based on vocal patterns. It has the potential to revolutionize animal welfare, livestock management, and conservation, allowing us to monitor animals’ emotions in real-time.
Élodie F. Briefer, Study Last Author and Associate Professor, Department of Biology, University of Copenhagen
AI as a Universal Animal Emotion Translator
The researchers identified key acoustic indicators of emotional valence by analyzing thousands of vocalizations made by ungulates in different emotional states. Changes in duration, energy distribution, fundamental frequency, and amplitude modulation were the most significant indicators of whether an emotion was positive or negative. These patterns were notably consistent across species, suggesting that basic emotional vocalizations are evolutionarily conserved.
A Game-Changer for Animal Welfare and Conservation
The results of the study have wide-ranging implications. The AI-powered classification model could lead to the development of automated tools for real-time animal mood monitoring, potentially transforming livestock management, veterinary care, and conservation efforts.
Understanding how animals express emotions can help us improve their well-being. If we can detect stress or discomfort early, we can intervene before it escalates. Equally important, we could also promote positive emotions. This would be a game-changer for animal welfare.
Élodie F. Briefer, Associate Professor and Study Last Author, Department of Biology, University of Copenhagen
Next Steps: Expanding Research and Sharing the Data
The researchers have made their database of labeled emotional calls from the seven ungulate species publicly available to facilitate further research.
Briefer concluded, “We want this to be a resource for other scientists. By making the data open access, we hope to accelerate research into how AI can help us better understand animals and improve their welfare.”
This research enhances our understanding of animal emotions and our ability to respond to them, opening up new possibilities for research, conservation, and animal welfare.
Journal Reference:
Lefèvre, A. R., et al. (2025) Machine learning algorithms can predict emotional valence across ungulate vocalizations. iScience. doi.org/10.1016/j.isci.2025.111834.