Israeli researchers said Monday they have built an artificial intelligence system that determines when farmed fish are hungry by analyzing the sounds they make while eating—an approach designed to help aquaculture operators feed more precisely, reduce pollution, and improve animal welfare.
The team, from the National Center for Mariculture in Eilat and the Israel Institute of Technology, trained machine-learning models to pick out the distinct biting sounds made by sea bream as they snap at food pellets, while filtering out the background noise typical of fish farms. By tracking those bite signatures in real time, the system can gauge whether fish are still actively feeding, have had enough, or may be experiencing stress.
Give the gift of hope
We practice what we preach:
accurate, fearless journalism. But we can't do it alone.
- On the ground in Gaza, Syria, Israel, Egypt, Pakistan, and more
- Our program trained more than 100 journalists
- Calling out fake news and reporting real facts
- On the ground in Gaza, Syria, Israel, Egypt, Pakistan, and more
- Our program trained more than 100 journalists
- Calling out fake news and reporting real facts
Join us.
Support The Media Line. Save democracy.
Researchers said feeding control is one of the biggest pressure points in aquaculture, a fast-growing source of global seafood. Many farms still lean on broad estimates based on fish weight and standard feeding charts. That can push farms into overfeeding—wasting money and degrading water quality—or underfeeding, which can stress fish, slow growth, and increase disease risk.
The sound-based method, slated for publication in the March issue of Computers and Electronics in Agriculture, is positioned as a scalable and automated way to optimize feeding while also tracking welfare indicators.
The team also reported early promise beyond finfish: Similar sound-monitoring approaches have helped detect stress and aggression patterns in shrimp and oysters, pointing to wider potential use across shellfish farming.