Signify

• 

  light mode

• 

  light mode

• 

  light mode

• 

  light mode

• 

  dark mode

• 

  dark mode

• 

  dark mode

Inspiration

We were inspired by the need for accessibility in everyday communication. Many people in the Deaf and Hard-of-Hearing community face barriers when interacting with others who don’t know sign language. We wanted to build a tool that makes conversations more inclusive and accessible for everyone.

What it does

Our project is a real-time sign language interpreter. It recognizes hand gestures and translates them into text or speech, allowing smoother communication between sign language users and non-signers.

How we built it

We used computer vision to detect hand gestures and map them to sign language letters/words. We trained models with datasets of sign language gestures, then integrated them into a web application for easy accessibility. The app was built using Python, react.js, mediapipe, scikit-learn, and OpenCV, with a simple and intuitive UI.

Challenges we ran into

• Training models accurately with limited data
• Handling differences in lighting, angles, and hand positions
• Making real-time recognition fast enough to be practical
• Designing a clean and accessible interface within the hackathon timeline

Accomplishments that we're proud of

• Built a working prototype that successfully recognizes and interprets sign language gestures
• Created a user-friendly application focused on accessibility
• Learned how to integrate machine learning with computer vision in a short time
• Took a meaningful step toward inclusive communication

What we learned

We learned how to combine AI, computer vision, and accessibility design principles to solve real-world problems. We also gained experience collaborating under time pressure and adapting quickly to challenges.

What's next for Signify

We want to expand Signify to support full sentences, more gestures, and multiple sign languages. Future improvements could include mobile integration, support for speech-to-sign translation, and accessibility partnerships to bring the tool to schools, workplaces, and public services.

Built With

• mediapipe
• opencv
• python
• python-opencv
• react.js
• scikit
• scikit-learn