A groundbreaking study from the University of Copenhagen reveals that artificial intelligence can accurately decode emotions in seven ungulate species, such as cows and pigs. By training a machine-learning algorithm to analyze vocal patterns, researchers achieved 89.49% accuracy in distinguishing between positive and negative emotions. This pioneering work not only marks the first cross-species emotional recognition using AI but also has profound implications for animal welfare, as it offers a way to monitor emotions in real-time, enhancing livestock management and conservation efforts.
"This breakthrough provides solid evidence that AI can decode emotions across multiple species based on vocal patterns. It has the potential to revolutionise animal welfare, livestock management, and conservation, allowing us to monitor animals' emotions in real time."
Researchers have successfully trained a machine-learning model to distinguish between positive and negative emotions in seven different ungulate species, achieving an impressive accuracy of 89.49%.
The model analyzes the acoustic patterns of vocalizations, marking the first cross-species study to detect emotional valence using AI, leading to new insights in animal communication.
By identifying and analyzing thousands of vocalizations from ungulates in different emotional states, the researchers pinpointed key acoustic indicators that predict emotional valence.
Collection
[
|
...
]