Backend
SIGMA: speaker-isolated, symptom-structured multimodal estimation of depression severity from clinical interviews
arXiv cs.SD0 views
arXiv:2611.23348v1 Announce Type: new Abstract: Clinical interviews are central to depression assessment but time-consuming and rater-dependent. Automated severity estimation offers scalable screening, yet existing methods often retain interviewer speech (confounding participant cues), fuse modalities late (losing fine-grained cross-modal dynamics), or score symptoms independently (ignoring inter-symptom correlations). We present SIGMA, a speaker-aware, symptom-structured framework for depression-severity estimation from semi-clinical interviews. SIGMA uses speaker diarization to restrict input to participant speech, cross-attention so that text, audio and video condition one another during encoding, and a graded-response head that predicts the eight questionnaire items jointly through a shared latent severity trait with item-specific ordered thresholds. On the Extended Distress Analysis Interview Corpus, SIGMA achieves state-of-the-art performance (CCC 0.762, RMSE 4.13, MAE 3.30 over five seeds). Ablations confirm that isolating participant speech and joint, structured symptom prediction each substantially improve concordance and estimation accuracy.
Source:https://arxiv.org/abs/2611.23348PreviousShared Randomness and Parallel Rounds in Permutation SamplingNextGreat Listeners, Poor Decision-Makers: Rethinking Agentic Audio AI
Related posts
What Jobs Can a Fresher Get After Learning Python? A Practical Guide
0dev.to
Defensive Coding for Real-Time Scoreboard Components
0dev.to
Dev News Digest: 19 Nov 2026, 06:00
0dev.to News
Download Website as HTML: Get a Runnable Project From a Figma Capture
0dev.to Design
A bilingual article’s canonical URL changes from a draft decision to an API result
0dev.to Discuss
Signs of Wear: Evaluating Intelligent Speed Assistance Under Synthetic Adverse Conditions
0arXiv eess.IV