Entropy-Dominated Temporal Vocal Dynamics as Digital Biomarkers for Depression Detection

TL;DR

This study demonstrates that entropy-based temporal vocal biomarkers significantly enhance automated depression detection by capturing dynamic conversational patterns beyond static features.

Contribution

The paper introduces entropy-driven temporal biomarkers as a novel approach that outperforms traditional pooled features in depression detection from speech.

Findings

Entropy biomarkers achieved an AUC of 0.646, outperforming pooled features.

Entropy biomarkers showed statistical significance with p = 0.017.

Dynamic entropy features were more stable and informative than other complexity measures.

Abstract

Automated depression detection often relies on static aggregation of conversational signals, potentially obscuring clinically meaningful behavioral dynamics. We investigated whether entropy-driven temporal biomarkers improve depression detection beyond standard pooled features using the DAIC-WOZ corpus. Using 142 labeled participants, we reconstructed utterance-level acoustic trajectories and compared pooled temporal baselines, trajectory dynamics, Shannon entropy biomarkers, recurrence quantification, sample entropy, fractal complexity, and coupling biomarkers under leakage-aware validation. Static pooling achieved an AUC of 0.593, trajectory dynamics improved performance to 0.637, and entropy biomarkers produced the strongest statistically significant improvement over pooled baselines (AUC 0.646; nested cross-validated AUC 0.615; permutation p = 0.017). Entropy biomarkers outperformed recurrence, coupling, sample entropy, and fractalbased features, with several biomarkers stable across folds. These findings suggest depression-related signal may lie less in average acoustic levels than in entropy of conversational dynamics, supporting temporally informed digital phenotypes for mental-health assessment.