AI decodes 'oinks' and 'grunts' to keep pigs happy

European scientists have developed an artificial intelligence (AI) algorithm capable of interpreting pig squeals, aiming to create a tool that could help farmers improve animal welfare. The algorithm has the potential to improve the welfare of pigs by alerting farmers to their negative emotions.

Scientists from universities in Denmark, Germany, Switzerland, France, Norway and the Czech Republic used recorded sounds made by thousands of pigs in different scenarios, such as play, isolation and competition for food, and found that grunts, winks and squeals revealed positive or negative emotions in pigs. While many farmers already have a good understanding of the health of their animals by observing them in pens, most of the tools available measure their physical condition.

The algorithm showed that pigs raised outdoors in free-range or organic farms were able to roam and dig in the dirt, producing fewer stressful reactions than conventionally raised pigs. The researchers believe that this method, once fully developed, could also be used to label farms and help consumers make informed choices.

The scientists used these findings to create an algorithm that employs artificial intelligence. Artificial intelligence not only helps us process a large number of sounds, but also automatically categorize them,...

Short grunts usually indicate positive emotions, while long grunts usually indicate discomfort, such as pigs pushing against each other at the trough. High frequency sounds such as squealing or squeaking usually mean that the pigs are under stress, for example when they are distressed, fighting or separated from each other.