Reproduce “Using voice analysis as an early indicator of risk for depression in young adults” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Klaus R. Scherer, Felix Burkhardt, Uwe D. Reichel, Florian Eyben, Björn W. Schuller
Researchers investigated acoustic voice analysis to identify early indicators of depression risk in young adults, finding that lower pitch, reduced vocal effort, and specific spectral balance changes correlated with high-risk individuals. The developed model classified high-risk participants with a cross-validated accuracy of 63.9%.
Key insights
- High-risk young adults for depression consistently exhibited acoustic patterns characterized by less vocal effort, stronger glottal constriction (creaky voice quality), and a lower mean fundamental frequency (pitch).
- Specific spectral balance parameters (e.g., alphaRatio UV mean, mfcc2 mean, slope V0-500 mean) and Mel-frequency Cepstral Coefficients (e.g., mfcc4 mean) are robust indicators of depression risk.
- Voice analysis offers a non-obtrusive and complementary method for early risk identification, potentially augmenting existing self-report and behavioral assessments in young adults.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.