Chickens are known as noisy creatures because they make noise when they are content or frightened. They even sometimes sing "buk, buk, ba-gawk" while laying an egg. Chicks also vocalize, a little bit different from adult chickens.
According to an article in Science, an international team of scientists developed a novel artificial intelligence (AI) that automatically determines which of the sounds chickens produce are distress calls to improve their welfare. The technology could be available commercially within five years so that farmers can save the lives of these birds before it is too late.
Novel Artificial Intelligence Can Determine Distress Calls From Chickens With 97% Accuracy
Distress Calls of Chickens
Distress calls from chickens are triggered by environmental stressors rather than loud noises, according to Poultry World. Even as chicks, they utter distress call characterized by high-pitched and repetitive chirps to attract their mother hen's attention for food or because they are uncomfortable.
On the other hand, distress calls from chickens are not triggered by startling stimuli like loud noises. Instead, discomfort or risky environments, such as heat and cold stress, maternal signaling of threat, food and water restriction, and social isolation, prompt them to make distress calls.
However, being in constant distress is not good for them, especially the young ones. Early life stress is energetically costly, affecting their growth and immune system and increasing mortality risks. Like humans, increased levels of stress hormones called cortisol could suppress weight gain, and increase oxidative damage and heart problems development.
AI Improves Welfare of Chickens
The Food and Agriculture Organization of the United Nations estimated that there are approximately 33 billion chickens around the world. Unfortunately, many of them are in poor conditions and packed together in stressful environments that endanger their overall welfare.
New Scientist reports that the new AI developed was trained to identify and count the distress calls of chickens that farmers can use to improve the conditions of their poultry. It can detect that sharp, short chirp that signals distress so farmers can address them.
McElligott and collaborators have labeled calls from the recorded audio of chickens in barns in southwestern China, as chicken chirps of pleasure and fear. Initial testing of the algorithm accurately detected distress calls 85% of the time, but further improvements have increased it to 97%.
The technology has not been used on a working chicken farm, but the team is planning to conduct another experiment to understand the distress calls of chickens and their well-being but with some conditions.
A similar tool was also developed in Denmark by Elodie Floriane Mandel-Briefer at the University of Copenhagen that determines the emotions of pigs based on the sounds and facial expressions made.
The team described their technology in full detail in the study titled "Automated Identification of Chicken Distress Vocalizations Using Deep Learning Models," published in the Journal of the Royal Society Interface.
RELATED ARTICLE: How to Know If Your Cat is Happy? AI-Powered App Uses Camera to Tell a Feline's Mood
Check out more news and information on Artificial Intelligence in Science Times.
