chickens
(Photo : Pexels / Engin Akyurt )

A team of Japanese researchers claims that they were able to use AI to translate the clucks and noises of chickens.

AI Chicken Language Translator?

This novel feat was documented in a preprint that is yet to undergo peer review. The research team was reportedly able to develop a system that can interpret chicken's different emotional states. These covered the fowls' feelings of anger, fear, hunger, excitement, contentment, and distress.

The technology relied on an AI technique that the researchers refer to as Deep Emotional Analysis Learning, according to professor Adrian David Cheok from the University of Tokyo.

According to the team, the method depends on complex math algorithms. They also claim that the technology can be used to adapt to the changing vocal patterns of chickens. In theory, this means that the tool will get better in deciphering as time passes.

ALSO READ: AI Might Help Humans Talk to Animals, Opening New Doors for Communicating With Nature

How the Novel AI Chicken Translator Works

The team tested their system by recording and analyzing samples taken from 80 chickens. These samples were then fed to the algorithm that is in charge of connecting the vocal patterns with the birds' different emotional states.

The researchers collaborated with eight veterinary surgeons and animal psychologists. Doing so enabled them to distinguish the mental state of the chicken with an accuracy rate that was shockingly high.

The experiments show how AI and machine learning could potentially be used to recognize chicken emotional states based on the sounds that they make. The remarkable detection probabilities for every emotion shows that the model was able to learn how to capture and decipher relevant features and patterns of the chicken vocalizations.

Findings, however, should still be taken with a grain of salt. The researchers acknowledge that their model's accuracy could be influenced by the specific chicken breed as well as other environmental conditions. The training and evaluation dataset may also be unable to fully capture the entire scope of the emotional variations and states of the birds. Not to mention, chickens also have other ways to communicate, including body language.

Nevertheless, Cheok explains that this is quite a leap for the scientific world and that it is just the start. Cheok adds that they hope to use such ML and AI methods for other animals and lay down a framework for intelligence across various industries that involve animals. He adds that knowing animal feelings can enable humanity to design a better world for these creatures.

RELATED ARTICLE: Clothing 'Wearable Algorithm' Makes CCTVs Identify Humans as Animals, Confuses AI Recognition

Check out more news and information on Artificial Intelligence in Science Times.