Abstract
Automatic depression detection typically models the semantic content and acoustic characteristics of participant speech during clinical interviews. However, the interactional timing between the clinician and participant remains comparatively under-modeled. We investigate conversational temporal dynamics, specifically dyadic turn-pair timing, as a primary modality fused with self-supervised encoders.
Evaluated on the DAIC-WOZ dataset, we compare a compact 24-dimensional timing module against frozen WavLM-large and RoBERTa-large baseline detectors. This temporal module achieves the highest single-modality performance on the development set. Furthermore, a convex-weighted late fusion strategy improves overall performance to 0.804 and 0.669 macro-F1 on the development and test sets, respectively. The learned fusion effectively assigns zero weight to acoustics, demonstrating that conversational timing serves as a lightweight, interpretable complement for dyadic depression screening.
Blogger's Review: This study explores a new avenue for depression detection by incorporating conversational temporal dynamics, showcasing the potential of timing information in emotional analysis. It highlights the value of multimodal fusion, enhancing detection performance while providing interpretability, paving the way for innovative clinical applications.