Artificial Intelligence Deciphers Language of Crows

The researchers used a machine learning model developed by the Earth Species Project to recognize crow vocalizations and distinguish them from ambient sounds. They first had to manually classify 7,000 recordings to train the system. Once trained, the AI was able to identify patterns in the data, including a possible correlation between a specific call and a defensive behavior. The researchers found that when a predator attacks a nest, a crow emits a brief call, and other crows soon join to defend the nest. They hypothesize that this call may serve as a distress signal, but further study is needed to confirm this.