Algorithm translate pig grunts

Using more than 7,000 recordings of pig sounds, a team of international researchers has developed an algorithm that can distinguish sounds based on emotion.

The idea from this extensive research: to develop a tool and application that could be suspended in pigs. This way, the farmer can receive a signal if something is wrong, or he can measure whether improvements in the barn are also having an impact on the animals’ welfare. Of course, you could also consider an app where the authorities can monitor whether the animals are being looked after well enough in some pig houses.

The tool and app still need development, but what we already know is that the algorithm is already 92 percent accurate at distinguishing between negative and positive emotions. This would of course also be of interest to other animal species, but an exceptionally large amount of data was used for this study. If you also want to do this for cows, for example, there is probably still a lot to include.

You hear in this voice researcher Elodie Briefer from the University of Copenhagen. Read more about the research here: Pig grunts reveal their feelings† The paper can be found here: Classification of the calls of pigs produced from birth to slaughter according to emotional equivalence and production context†