Listen to the news
A team of researchers in Japan claims to have found a way to translate the cackling of hens into human speech with the help of artificial intelligence.
As detailed in a preprint yet to be peer-reviewed, the team led by University of Tokyo professor Adrian David Cheok has created "a system capable of interpreting a variety of emotional states in chickens, including hunger, fear, anger, contentment, excitement and suffering" by using "cutting-edge AI technique we call deep emotional analysis learning."
The technique "takes root in complex mathematical algorithms" and can even be used to adapt to the chickens' ever-changing voice patterns, meaning it is getting better at deciphering "chicken vocalizations" over time.
To test their new system, the team recorded and analyzed samples from 80 chickens. They then fed these samples to an algorithm to relate these vocal patterns to different "emotional states" of the birds.
Microsoft's Artificial Intelligence Team Accidentally Releases Company Secrets
Teaming up with "a team of eight animal psychologists and veterinary surgeons," the researchers say they were able to determine a chicken's mental state with surprisingly high accuracy.
"The results of our experiments demonstrate the potential to use AI and machine learning techniques to recognize emotional states in chickens based on their sound signals," the document said. "The high mean detection probabilities of each emotion suggest that our model learned to capture meaningful patterns and characteristics from the sounds of the chicken."
In their paper, the researchers acknowledge that the accuracy of their model may change with different breeds and environmental conditions, and that "the dataset used for training and evaluation may not encompass the full range of emotional states and variations of chickens."
'Generative AI is just a phase': Google co-founder DeepMind announces what's next
Then there are many other ways chickens communicate, from "other non-acoustic cues, such as body language and social interactions."