Researchers describe a multilingual eye-sign communication system for augmentative and alternative communication
Original source
Thrivaad: A Multilingual, Predictive Eye-Sign-Based AAC System Powered by Optimized Deep Learning. (opens in a new tab)Compass summarised this from the study's abstract.
Study details
- Studied in
- Human
A study describes Thrivaad, a webcam-based communication system that uses eye movements to select text in English, Malayalam and Hindi. It combines an optimized YOLOv5 deep-learning model with text prediction to reduce the eye gestures needed to form a message. The study used 35,263 eye-sign images from 60 participants across three age groups.
Why this matters
The system could be relevant to people who rely on eye-based augmentative and alternative communication, including some people with ALS, if it works reliably in real-world use. However, this study does not show that Thrivaad improves communication for people with ALS or change clinical treatment.
Limitations and context
The source is a primary journal article, but the supplied information does not report performance metrics, participant diagnoses, comparisons with existing systems, or testing in people with ALS. It describes a system trained on collected eye-sign images, so further testing would be needed to establish usability and reliability in everyday communication, including under fatigue and changing conditions.