AI-Driven Multimodal Diagnosis of Developmental Speech and Language Disorders (DSLD)
Voice AI
Multimodal
Diagnosis
Fairness
Overview
This project develops an AI-driven, multimodal diagnostic framework for developmental speech and language disorders (DSLD), enabling earlier, fairer, and more objective identification of communication difficulties in children.
What we do
We integrate voice-based acoustic and linguistic analyses with clinical and behavioral measures to build interpretable machine-learning models that can support clinical decision-making.
Key components
- Voice-based acoustic and linguistic feature extraction
- Multimodal data integration (speech, language, clinical variables)
- Interpretable and fairness-aware modeling
- Foundations for clinical decision-support tools