Zero Barriers

Inspiration

We noticed a persistent communication gap between Deaf and hearing communities, especially in real-time conversations. Our goal was to create a solution that could bridge this divide, making communication more inclusive and accessible for everyone.

What it does

Zero Barriers provides real-time translation of spoken language into ASL-like gestures through a digital avatar. As someone speaks, the avatar performs gestures that approximate American Sign Language, allowing deaf users to understand audio (Ex: listening to a video with audio).

How we built it

We combined web technologies, React, Typescript, and real-time audio processing to capture speech input. Animations were implemented using Mixamo avatars, which were programmed to perform sequences corresponding to ASL-like gestures. The system communicates with a Supabase backend to store sessions and handle real-time updates.

Challenges we ran into

Limited ASL animation resources: Mixamo doesn’t provide full ASL, so we had to creatively map gestures to convey meaning. Real-time processing constraints: Translating audio into animations with minimal delay required careful optimization. User experience design: Ensuring the avatar’s gestures were intuitive and understandable was tricky without a fully accurate ASL model. Limited Capabilites: Limited amount of words that can be translated for now (planning to expand)

Accomplishments that we're proud of

Achieved real-time gesture translation from spoken input. Built a fully functional prototype within a limited timeframe. Created a foundation that can later integrate accurate ASL datasets or ML-based gesture synthesis.

What we learned

The importance of creative problem-solving when perfect resources aren’t available. How to handle real-time audio streaming and animation sequencing in a web application. Gained insight into inclusive design principles for accessibility-focused technology.