Accepted for/Published in: JMIR Formative Research
Date Submitted: Feb 27, 2026
Date Accepted: Sep 4, 2026
Voice-Derived Speech Features of Anhedonia in Adults With Depression: Exploratory Secondary Analysis of a Semi-Structured Clinical Interview
ABSTRACT
Background:
Anhedonia is a core dimension of depression that may manifest in alterations in speech production and expressivity. Voice-based digital phenotyping offers a scalable approach to identifying behavioral markers associated with depressive symptom dimensions, but preliminary work is needed to characterize speech signatures of anhedonia and evaluate their prospective relevance.
Objective:
To examine cross-sectional and short-term longitudinal associations between multimodal voice features and anhedonia symptoms in adults with depression prior to initiating a behavioral intervention.
Methods:
Depressed adults (N = 56) completed a semi-structured symptom interview before treatment initiation. Acoustic (pitch, spectral/MFCC principal components, jitter/shimmer), prosodic, linguistic (lexical richness, sentiment), topic-model, and automatic speech recognition (ASR) quality features were extracted. Permutation tests evaluated cross-sectional associations with baseline anhedonia and prospective associations with baseline-adjusted week-4 anhedonia outcomes (controlling for baseline symptoms and treatment condition; n = 36).
Results:
Higher baseline anhedonia was characterized by briefer, more lexically diverse speech (higher hapax ratio/lexical diversity; rs = 0.47), more negative sentiment, fewer total and content words (rs ≈ −0.43), lower ASR token confidence, and flatter pitch variation. Work/school-related topic content appeared less frequently. Longitudinally, reduced speech output and lower expressive complexity at baseline were associated with increases in anhedonia at week 4.
Conclusions:
Anhedonia was associated with a voice signature marked by reduced speech output, diminished prosodic variability, more negative sentiment, and sparser yet more lexically diverse language. Baseline markers of reduced verbal output and expressivity prospectively predicted increases in anhedonia. These findings are exploratory and require replication in larger samples with multiple-testing control and more naturalistic speech contexts. Clinical Trial: N/A
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.