Vocalization Patterns in Laying Hens - An Analysis of Stress-Induced Audio Responses

This study leverages Convolutional Neural Networks (CNN) and Mel Frequency Cepstral Coefficients (MFCC) to analyze the vocalization patterns of laying hens, focusing on their responses to both visual (umbrella opening) and auditory (dog barking) stressors at different ages. The aim is to understand how these diverse stressors, along with the hens’ age and the timing of stress application, affect their vocal behavior. Utilizing a comprehensive dataset of chicken vocal recordings, both from stress-exposed and control groups, the research enables a detailed comparative analysis of vocal responses to varied environmental stimuli. A significant outcome of this study is the distinct vocal patterns exhibited by younger chickens compared to older ones, suggesting developmental variations in stress response. This finding contributes to a deeper understanding of poultry welfare, demon-strating the potential of non-invasive vocalization analysis for early stress detection and aligning with ethical live-stock management practices. The CNN model’s ability to distinguish between pre- and post-stress vocalizations highlights the substantial impact of stressor application on chicken vocal behavior. This study not only sheds light on the nuanced interactions between stress stimuli and animal behavior but also marks a significant advancement in smart farming. It paves the way for real-time welfare assessments and more informed decision-making in poultry management. Looking forward, the study suggests avenues for longitudinal research on chronic stress and the application of these methodologies across different species and farming contexts. Ultimately, this research represents a pivotal step in integrating technology with animal welfare, offering a promising approach to transforming welfare assessments in animal husbandry.