Researchers at Virginia Tech in the United States are utilizing sound and machine learning to comprehend cow behavior and emotions, which may help farmers reduce their suffering. They want to capture audio from cows, calves, and beef cattle on the pasture by attaching tiny recording devices to their halters or collars. Then they’ll employ machine learning to detect indicators of stress or disease.
The researchers want to learn how cows transmit discomfort by matching the data with saliva cortisol samples collected from the cows. They can determine if the cows are stressed by matching the data with saliva cortisol tests. This technique is non-invasive, like utilizing a microphone during a speech.
Chen intends to create a computational pipeline that combines audio data with pre-trained machine learning models and interactive visualization. This will be open-source and freely accessible to the public, providing a tool that will allow a broad range of individuals to have a better understanding of the levels of animal welfare in any given situation.
The data might also assist determine which cows burp less, since cow burps contain a lot of methane, a greenhouse gas. They will compare the DNA of cows who burp more and those that burp less in order to determine if such burping levels are inherited. They will also examine the impact of rumen modifiers (food additives that reduce methane production).
