Audio Frequency-Time Dual Domain Evaluation on Depression Diagnosis

TL;DR

This paper introduces a novel deep learning-based method utilizing frequency-time dual domain analysis of voice signals for improved depression diagnosis, addressing current challenges in mental health assessment.

Contribution

It proposes a new multimodal voice analysis approach combining frequency and time domains with deep learning for depression detection.

Findings

High classification accuracy in depression diagnosis

Effective utilization of voice's dual domain features

Provides new tools for depression screening and diagnosis

Abstract

Depression, as a typical mental disorder, has become a prevalent issue significantly impacting public health. However, the prevention and treatment of depression still face multiple challenges, including complex diagnostic procedures, ambiguous criteria, and low consultation rates, which severely hinder timely assessment and intervention. To address these issues, this study adopts voice as a physiological signal and leverages its frequency-time dual domain multimodal characteristics along with deep learning models to develop an intelligent assessment and diagnostic algorithm for depression. Experimental results demonstrate that the proposed method achieves excellent performance in the classification task for depression diagnosis, offering new insights and approaches for the assessment, screening, and diagnosis of depression.