Speech as a biomarker for supported diagnosis of major depressive disorder using self-supervised representations

AI Summary

Largest multicentre speech cohort (N=1,816; 23,608 samples) used to build diagnostic models for major depressive disorder.
Self-supervised deep learning (WavLM, HuBERT) on acoustic-prosodic features outperformed traditional openSMILE methods, achieving internal AUC 0.932 and external AUC 0.879.
Findings indicate a rapid, cost-effective, non-invasive speech biomarker approach for assisted clinical depression assessment and potential translation.

Nat Commun. 2026 Jun 15. doi: 10.1038/s41467-026-74122-9. Online ahead of print.

ABSTRACT

The diagnosis of major depressive disorder (MDD) urgently requires objective biomarkers for clinical translation. In this study, we established a multicenter cohort to date (N = 1,816; comprising 910 MDD patients and 906 healthy control subjects), using 23,608 standardized speech samples. Based on 6,373 acoustic-prosodic features, we develop a deep learning framework that employs a self-supervised architecture to leverage speech biomarkers. We perform a systematic comparative analysis among pretrained foundation models, including WavLM and HuBERT, and traditional acoustic features extracted from openSMILE. Model performance is evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. Our framework achieves an AUC of 0.932 in internal validation (n = 333), significantly outperforming conventional methods, while maintaining efficacy in external validation (n = 160, AUC = 0.879). Self-supervised representations demonstrate robust diagnostic accuracy compared to other models. Leveraging the large speech biomarker dataset, our findings provide a rapid, cost-effective, and non-invasive approach for assisted depression assessment.

PMID:42297825 | DOI:10.1038/s41467-026-74122-9

Document this CPD