Depression Markers in Speech: An Approach based on Tract Variables Dynamics

2026-07-28

AI Summary

Current speech-based biomarkers for depression lack characterization of the dynamic properties of the vocal apparatus. This study addresses this gap by introducing dynamic features of vocal tract variables and employing nonlinear dynamical methods—specifically, the maximum Lyapunov exponent, correlation dimension, and sample entropy—to quantify predictability, complexity, and randomness in speech, thereby extending conventional acoustic analysis. Experimental results on the Androids corpus demonstrate that the proposed biomarkers effectively differentiate individuals with depression from healthy controls in both read and spontaneous speech tasks, with statistically significant effect sizes (Cliff’s delta). These findings offer a novel, interpretable, and objective voice-based biomarker for depression diagnosis.

Abstract

This study identifies new depression biomarkers based on the dynamical properties of tract variables, which represent geometric features describing the configuration of the speech articulators. A key advantage of this approach lies in its ability to quantify aspects of the articulatory process that have not been previously explored in the context of depression, namely predictability, complexity, and randomness. These properties are respectively characterised using the Largest Lyapunov Exponent, the Correlation Dimension, and the Sample Entropy. Thorough experiments were conducted on the Androids Corpus, a publicly available dataset comprising 64 speakers diagnosed with depression by clinicians and 54 control speakers with no reported history of mental health conditions. The results indicate that the proposed biomarkers effectively discriminate between the depressed and control speakers, as evidenced by the high Cliffs delta values across both read and spontaneous speech.

Problem

• depression biomarkers
• speech articulators
• tract variables
• dynamical properties
• mental health

Innovation

• tract variables
• dynamical biomarkers
• Largest Lyapunov Exponent
• Correlation Dimension
• Sample Entropy