All case studies
Accessibility
Assistive AI that turns gestures into speech
A multimodal system that analyzes video and produces text, speech, and images for audiences with special needs.
- Gesture to text and speech
- Multilingual output
- Real time interpretation
The challenge
The client wanted to help people who communicate through hand gestures interact more easily with others, in several languages and in real time.
Our solution
We built a computer vision and NLP pipeline that detects and interprets hand gestures from video, translates them into letters and sentences, and produces audible output. The system can also generate images or text from the same input, with multilingual support.
The outcome
The product delivers real time interpretation and creative content generation, giving its users three accessible ways to communicate from a single input.
