Automated Classification of Depression Severity Using Speech - A Comparison of Two Machine Learning Architectures

Depression recognition using voice-based pre-training model

“…Exploration of depression detection based on voice signals began with the widespread adoption of two key approaches: traditional machine learning and neural networks.Gao et al used Random Forest Algorithm30 and Shi et al used SVM for depression recognition31 . Aharonson et al also chose machine learning methods to automatically categorize depression severity32 . Depression detection studies based on machine learning methods usually struggle to learn meaningful knowledge representations, and classification accuracy relies heavily on feature selection.…”