Inspiration
Too often, Deaf and hard-of-hearing people are left out of conversations in classrooms, meetings, and public talks. We wanted to create something that makes those spaces instantly inclusive—no extra setup, no barriers—just communication that works for everyone.
What it does
Project Hands Up converts live spoken words into American Sign Language (ASL) through an animated avatar. A user speaks into the app, and in real time, their words are transcribed and signed. Deaf users can follow the conversation without needing captions or interpreters.
How we built it
Frontend: Built in React for a clean, responsive user experience.
Backend: Developed with Python using FastAPI to handle requests quickly and efficiently.
AI Layer: Used the Gemini API for speech-to-text and language processing. ASL Output: Integrated an animated signing avatar that translates text into ASL. Deployment: Hosted under a .tech domain to make it easily accessible.
Challenges we ran into
Making ASL output accurate and natural, since direct word-for-word translation doesn’t always work. Coordinating speech recognition and avatar signing in real time without noticeable lag. Learning new frameworks (FastAPI, Gemini API) and integrating them under time pressure.
Accomplishments that we're proud of
Built a functioning prototype that translates live voice input into sign language. Seamlessly combined React, FastAPI, and Gemini into a real-time system. Focused on accessibility and impact, not just technical novelty.
What we learned
Real accessibility design requires thinking about user experience first—not just technical feasibility. How to build a full-stack app quickly using modern frameworks like FastAPI and React. The challenges of bridging language, culture, and technology when working with ASL.
Log in or sign up for Devpost to join the conversation.