Generalized Dilated CNN Models for Depression Detection Using Inverted Vocal Tract Variables

TL;DR

This paper develops a generalized dilated CNN classifier for depression detection using vocal tract variable features, demonstrating improved cross-database accuracy and robustness over previous models.

Contribution

It introduces a CNN model trained on vocal tract variable features from multiple databases, enhancing generalizability in depression detection.

Findings

ACFs from Vocal Tract Variables are effective features.

Model achieves ~10% accuracy improvement in cross-corpus evaluations.

Fusion of TVs and Mel-Frequency Cepstral Coefficients further boosts performance.

Abstract

Depression detection using vocal biomarkers is a highly researched area. Articulatory coordination features (ACFs) are developed based on the changes in neuromotor coordination due to psychomotor slowing, a key feature of Major Depressive Disorder. However findings of existing studies are mostly validated on a single database which limits the generalizability of results. Variability across different depression databases adversely affects the results in cross corpus evaluations (CCEs). We propose to develop a generalized classifier for depression detection using a dilated Convolutional Neural Network which is trained on ACFs extracted from two depression databases. We show that ACFs derived from Vocal Tract Variables (TVs) show promise as a robust set of features for depression detection. Our model achieves relative accuracy improvements of ~10% compared to CCEs performed on models trained on a single database. We extend the study to show that fusing TVs and Mel-Frequency Cepstral Coefficients can further improve the performance of this classifier.