A deep learning model has been developed to identify and count chicken distress calls, which could potentially improve the welfare of chickens raised on crowded commercial farms. The tool aims to address the serious welfare concerns faced by billions of chickens worldwide, who often live in poor conditions with limited mobility. Chicken distress calls, characterized by sharp and short “cheep” sounds, can provide insight into the animal’s health and growth rate. However, identifying these calls becomes challenging in environments with thousands of chickens cheeping together. The AI model, trained on recordings from large broiler chicken farms, demonstrated an accuracy of around 85% in detecting distress calls. The next steps involve implementing measures to reduce distress calls, such as providing more space and enrichment opportunities for the chickens. This study contributes to the growing evidence that animal emotions can be assessed using machine learning techniques, highlighting the importance of evaluating animal welfare.
| Signal | Change | 10y horizon | Driving force |
|---|---|---|---|
| AI detects chicken distress calls | Improve chicken welfare | Chickens raised in better conditions | Concern for animal welfare |
| AI can identify and count distress calls | Improve conditions on commercial farms | Farmers use AI to enhance chicken welfare | Desire to improve farm conditions |
| Chicken distress calls predict health and growth | Understanding chicken well-being | Chickens raised with better welfare measures | Concern for chicken welfare |
| AI can measure and monitor animal emotions | Assessment of animal welfare | Improved assessment of animal emotions | Importance of animal welfare assessment |