Brief

Researchers have developed a framework for analyzing speech features to aid in clinical decision-making for mental health care. This system uses perceptually grounded acoustic and linguistic characteristics, such as prosody, vocal quality, and semantic coherence, to identify objective cues related to depression, anxiety, and ADHD. By employing interpretable machine learning techniques like XGBoost with SHAP and LIME, the framework establishes stable associations between symptom severity and vocal irregularities, lexical-syntactic patterns, and affective tone, validated across both benchmark and clinical datasets. AI

Researchers have developed EmoTrack, a new framework designed to more accurately track depression severity from counseling transcripts. This system combines signals extracted by large language models with semantic embeddings from individual transcript turns. EmoTrack also incorporates a novel dataset, LongCounsel, which includes longitudinal data and supervision for repeated sessions, enabling better performance even with partial symptom disclosure across sessions. AI