Like humans, animals have their own emotions, but their "language" and way of expression vary. In the similar sounds of human beings, in fact, there are different emotional information of animals.
An international team of researchers, involving the University of Copenhagen, ETH Zurich and the French National Institute for Agriculture, Food and the Environment (INRAE), recently developed an algorithm to decode pig emotions in an article published in Nature's journal Scientific Reports.
Image credit: Farmers Weekly
Studies of emotional sound indicators so far have generally been limited to specific types of calls generated by animals of a particular age, the study said. Domestic pigs are a highly developed species of vocal communication, and the acoustic characteristics of vocalization vary depending on the environment. To conduct the study, the researchers analyzed 7,414 pieces of sound data from 411 pigs, covering a wide range of scenarios from birth to slaughter (death).
Previous studies have shown that the vocalizations of domestic pigs can be divided into high frequency (HF) and low frequency (LF), with high frequency calls (such as screams) associated with negative emotions and low frequency calls associated with positive or neutral emotions.
Image courtesy of Unsplash
The researchers' study of pigs' voices was also accompanied by behavioral observations and, where possible, heart rate monitoring. After the study, an association between behavior and emotion was established. When the pig is reunited with its family, crowded together, and receives positive emotional scenes such as breastfeeding, it will trigger the pig's approaching or searching behavior.
In situations where there is stress, isolation, fighting with other pigs, slaughtering and castration, etc., the pigs will be emotionally negative.
In addition to verifying the link between high-frequency calls and negative emotions, low-frequency sounds and positive emotions. After a more thorough analysis of the sound data, the researchers also found that two specific acoustic characteristics and frequencies were equally important in understanding the mood of pigs: duration and amplitude of the call.
Elodie Briefer, an author from the University of Copenhagen, said: "When we look at positive and negative situations, there is a clear difference in the call of pigs. In positive cases, the call is much shorter and the amplitude fluctuates very little."
Image courtesy of Scientific Reports
The researchers then used neural networks to develop an algorithm that could recognize the pig's call and thus translate the emotional features in the pig's voice. The researchers say the initial iteration of the algorithm, with 92 percent accuracy, correctly translates pigs' emotions from their calls.
The researchers' long-term goal is to develop an app that can monitor the emotional state of pigs in real time, and farmers can also improve the welfare of animals.
It is worth mentioning that in addition to pigs, Briefer thinks that this analysis method may be used to develop a universal translator that can be applied to other types of mammals to track emotions through the sounds emitted by animals.
Click "Watching"
It is the greatest motivation for us