Towards Interpretable Depression Detection: Linking Acoustic Features to DSM-5 Indicators

Depression diagnosis still leans on subjective self-reports that can miss how someone actually behaves. This paper links speech acoustics to DSM-5 depressive-behavior indicators through a transparent Linkage Framework that, unlike black-box models, maps concrete features such as pitch variability, pauses, and tempo to clinical criteria, producing interpretable, indicator-level outputs and running locally on commodity hardware to protect privacy. Early results on DAIC-WOZ show directionally consistent associations for psychomotor change and concentration difficulty.