Depression Detection from Speech Emotions using MFCC based Recurrent Neural Network

Depression is a mental health disorder that affects millions of people worldwide. While there are many effective treatments for depression, the first step is often detecting the condition. In recent years, researchers have explored the use of machine learning algorithms to detect depression in speech patterns. Recurrent neural networks (RNNs) are a popular type of deep learning algorithm that can be used for this task. In this study, depression detection in speech using RNNs is used. The proposed system uses Mel-frequency cepstral coefficients (MFCCs) as input features to an RNN model. The RNN model is trained on an emotion dataset of speech recordings from individuals. The model is then used to classify new speech recordings based on stress/depression in each emotion. The experimental results show that RNNs are prominent for depression detection in speech. Future work could focus on improving the accuracy of the system by incorporating additional features in RNN architecture.