Automated speech-based screening of depression using deep convolutional neural networks

TL;DR

This paper presents a novel CNN-based method for automated depression detection from speech, achieving promising accuracy and offering a potential objective tool for mental health screening.

Contribution

It introduces a new deep learning approach using residual CNNs and spectrograms for depression detection in speech, with a comprehensive experimental evaluation.

Findings

Achieved 77% accuracy in depression detection

Used spectrograms as input for CNN models

Demonstrated the effectiveness of residual CNN architectures

Abstract

Early detection and treatment of depression is essential in promoting remission, preventing relapse, and reducing the emotional burden of the disease. Current diagnoses are primarily subjective, inconsistent across professionals, and expensive for individuals who may be in urgent need of help. This paper proposes a novel approach to automated depression detection in speech using convolutional neural network (CNN) and multipart interactive training. The model was tested using 2568 voice samples obtained from 77 non-depressed and 30 depressed individuals. In experiment conducted, data were applied to residual CNNs in the form of spectrograms, images auto-generated from audio samples. The experimental results obtained using different ResNet architectures gave a promising baseline accuracy reaching 77%.