MENTAL HEALTH AND DEPRESSION DETECTION

MENTAL HEALTH AND DEPRESSION DETECTION

NOOR FATHIMA ARFA

Undergraduate Student

B.M.S. College of Engineering

Bengaluru, India

noorfathima.cs19@bmsce.ac.in

PREMA

Undergraduate Student

B.M.S. College of Engineering

Bengaluru, India

prema.cs19@bmsce.ac.in

Dr. Prasad G R

Professor

B.M.S. College of Engineering

Bengaluru, India

prasad.cse@bmsce.ac.in

POOJA K

Undergraduate Student

B.M.S. College of Engineering

Bengaluru, India

poojak.cs19@bmsce.ac.in

SHIVANI GAHLOT

Undergraduate Student

B.M.S. College of Engineering

Bengaluru, India

shivani.cs19@bmsce.ac.in

Abstract- In many humanities-related professions, understanding a person's interior emotional condition is crucial. Emotions are a crucial part of our existence because of the profound influence they have on people's physical and mental wellbeing. Depression is a prevalent mental illness. The sluggish, hesitant, monotonous voice that is a notable trait of sad people may change as affective sensing technology advances with a focus on acoustic qualities.The system's goal is to use voice and facial expressions to classify emotions and depressive states. The system's goal is to use voice cues to classify emotions and depressive states. The system will employ

Haar based classifier and CNN model for facial emotion detection. The model for predicting depression based on the attributes is 3DCNN. The content of the speech itself is another aspect of speech that can be utilized to gauge a person's emotional state. It contains the crucial data that sentiment analysis can utilize to anticipate depression. You can convert speech to text using the Google Speech API. After feature extraction, the acquired text can be trained and tested using an SVM model.

Keywords : Depression,3D CNN, SVM, DAIC-WOZ, MFCC feature extraction