Scientists have created a pig translator decoding their emotions.
By Andrew Joseph
Researchers have developed an algorithm able to translate a pig’s grunts into an emotional equivalent, which could be used to monitor animal well-being in real-time.
We know that domestic pigs are smart animals, and show a wide range of vocal expressions.
bazilfoto/iStock/Getty Images Plus photo
Per previous research, high frequency squeals and screams have been associated with negative pig emotions, while the lower grunts indicating positive or neutral emotions.
But until now, it’s the emotions between the high and low pig noises that have been unidentified.
While the focus of animal wellness has always been on physical health with numerous systems available to monitor for, the research team noted that animal mental health wellness is also important for a pig’s overall health.
The impact of emotions on pig vocalization is not a new subject, but has only been restricted to specific call types produced by animals of a specific age in a specific environment.
According to the study performed by an international team of researchers led by the University of Copenhagen, the ETH Zurich, and France’s National Research Institute for Agriculture, Food and Environment (INREA) and published in the Scientific Reports journal this year, it has used thousands of acoustic recordings of pigs from birth to death to translate the sounds into emotions across various conditions and life stages.
Using over 7,000 audio recordings - both positive and negative situations in both commercial and experimental venues - the team of researchers from the Czech Republic, Denmark, France, Germany, Norway, and Sweden designed an algorithm.
“There are clear differences in pig calls when we look at positive and negative situations.”
The study noted that positive emotions occur, for example, when a piglet suckles from its mother, or being reunited with family members after a separation.
Examples of negative emotions were observed via separation, infighting, castration, and slaughter.
The researchers also created experimental environments it felt would evoke more nuanced emotions in the middle of the emotional spectrum, such as areas with toys or food, or areas lacking any stimuli.
They also placed, in separate tests, new or unfamiliar objects for pig interaction.
To examine for any changes, where possible the researchers monitored and recorded the pig’s calls, behavior and heartrates.
Analyzing the data from the 7,414 audio recordings from 411 pigs, the team looked for patterns in pig sounds as a function of emotions.
While the highs and lows of pig sounds have been previously studied, the research indicated new patterns of what pig’s experienced in greater detail.
“There are clear differences in pig calls when we look at positive and negative situations,” stated researcher Elodie 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,” she continued. “By training an algorithm to recognize these sounds, we can classify 92% of the calls to the correct emotion.”
The researchers used a neural network to develop the algorithm.
Briefer said that with the devised algorithm able to decode pig grunts, it needs someone to develop it into a usable sound recognition app for farmers to utilize to monitor pig welfare. She suggested that with additional data research, it could be used to understand the emotions of other animals.