• Thu. Aug 11th, 2022

mccoy.ventures

All content has been processed with publicly available content spinners. Not for human consumption.

AI that detects chicken distress calls could improve farm conditions

A deep learning model can pick out chicken distress calls from recordings taken at commercial farms, and could be used to improve chicken welfare

Life 29 June 2022

Cinematic close up shot of young male farmer is feeding from his hands ecologically grown white hen with proper genuine bio nutrient cereals for eggs laying in a barn of countryside agricultural farm.; Shutterstock ID 1899989971; purchase_order:

A farmer hand feeding a white chicken

HQuality/Shutterstock

An AI has been trained to identify and count chickens’ distress calls. Farmers could use the tool to improve conditions for chickens raised on crowded commercial farms.

As of 2020, there were more than 33 billion chickens around the world, according to the Food and Agriculture Organization of the United Nations. Many of these animals live in poor conditions, packed together with little ability to move around or do things chickens like to do. “Despite the basic concerns about not being hungry or not being thirsty, there are still serious welfare concerns about how they’re produced,” says Alan McElligott at City University of Hong Kong.

The frequency and volume of a chicken’s distress call – a sharp, short “cheep” – can predict the animal’s health and growth rate, according to McElligott’s previous research. But he says these calls can be hard to identify when there are thousands of chickens cheeping together – in some barns, it can be 25,000 or more. “They’re called barns, but they’re more like airline hangers,” says McElligott.

Advertisement

Listening to recordings made at large broiler chicken farms in southwestern China, McElligott and collaborators labelled chicken distress calls, distinguishing them from farm sounds and other chicken sounds like chirps of pleasure and trills that signal fear. With this labelled data, they trained several algorithms to identify distress calls from the background noise and measure their frequency and volume. When tested on other labelled recordings from the same farm, the best algorithm accurately detected distress calls around 85 per cent of the time.

The tool has not yet been used on a working chicken farm, and McElligott says there’s more work to do to understand the link between distress calls and a chicken’s well-being, but he says the next steps are somewhat obvious: “We should give them conditions in which, well, maybe they would produce less distress calls.” That could mean giving chickens more space, or other enrichments, such as giving them bales of straw to scratch at and climb on.

Elodie Floriane Mandel-Briefer at the University of Copenhagen in Denmark has developed similar tools to assess the emotions of pigs based on sounds and facial expressions. She says the study in chickens adds to growing evidence that animal emotions can be measured and monitored using machine learning. “Since animal emotions are an important part of their welfare, their assessment is crucial,” she says.

Journal reference: Journal of the Royal Society Interface, DOI: 10.1098/rsif.2021.0921

More on these topics: