This project aims to classify goat emotions based on vocalizations using machine learning techniques. We utilized the VOCAPRA dataset, extracting a comprehensive range of temporal and spectral features for analysis.
To explore the feature space, K-means clustering was applied for visualization. Subsequently, we trained and evaluated three classification models:
- Support Vector Machine (SVM)
- Random Forest
- Neural Network.
Our findings demonstrate that all models effectively classify vocalizations, with the Neural Network achieving the highest accuracy.