Oink Oink! Scientists use AI Pig translator to decode emotions through pig calls

The research, published in the journal Scientific Reports, shows how the researchers trained the algorithm to analyse acoustic signatures of 7,414 recordings of pig calls collected from 411 individuals throughout their life stages, including slaughter. The algorithm helps classify 92% of the calls to the correct specific emotion.
The algorithm can decode whether a pig is experiencing positive or negative emotions | Representative image

The algorithm can decode whether a pig is experiencing positive or negative emotions | Representative image

Photo : iStock
Scientists have “trained” a machine learning algorithm to decode whether an individual pig’s calls can be put down as a function of a positive or a negative emotion.
In a potential breakthrough for the field of animal behaviour, the new artificial-intelligence based “pig translator” will give insights into the well-being of pigs, especially in farms or slaughterhouses.
“We have trained the algorithm to decode pig grunts,” said Dr Elodie Briefer, animal behaviour and communication expert who co-led the study at the University of Copenhagen.
“Now we need someone who wants to develop the algorithm into an app that farmers can use to improve the welfare of their animals.”
The research, published in the journal Scientific Reports, shows how the researchers trained the algorithm to analyse acoustic signatures of 7,414 recordings of pig calls collected from 411 individuals throughout their life stages, including slaughter.
These recordings were collected from both commercial and experimental set ups.
Pigs wait to be fed in their enclosures at a farm
Pigs wait to be fed in their enclosures at a farm
Photo : iStock
The algorithm can decode whether a pig is experiencing positive emotions, such as happiness or excitement, or negative emotions like fear and distress.
Vocalisations associated with happiness or excitement were recorded in emotionally positive situations, for instance, when piglets suckle from their mothers or uniting with the family are being separated.
Negative situations included short social isolation, piglet fights, castration, handling and waiting in the slaughterhouse.
Researchers were able to decipher a pattern that gave further insight into what the pigs experienced in specific situations.
“There are clear differences in pig calls when we look at positive and negative situations,” said Dr. Briefer.
“In the positive situations, the calls are far shorter, with minor fluctuations in amplitude. Grunts, more specifically, begin high and gradually go lower in frequency.”
The algorithm helps classify 92% of the calls to the correct emotion, according to Dr Briefer.
Even though mental health is considered by farmers to be a crucial component of livestock’s well-being, animal welfare efforts are skewed primarily towards physical health.
Briefer and her colleagues believe their algorithm might pave the way for automated systems aimed at identifying calls to monitor the well-being of animals at farms.
End of Article