The problem
Traditional study interfaces often depend on visual controls. Blind and visually impaired students need learning tools that support speech and screen readers throughout the study flow.
What we built
I built Braillience with my team at Cal Hacks 12.0 in 2025. Teachers can upload PDF learning materials and generate AI flashcards. Students navigate their study sets through voice commands and use an in-browser AI tutor.
How it works
A React frontend connects to a Node.js and Express backend for authentication, document uploads, flashcards, and voice-learning sessions. Letta provides stateful learning assistance, with Gemini in the AI stack. Browser speech recognition and synthesis support spoken interaction.
The learning flow
The interface supports commands to begin learning, reveal an answer, move between cards, and repeat content. Quizzes and progress views extend the same study workflow beyond the initial flashcard set.
The focus
Accessibility is part of the main interaction design: a hands-free study experience, clear navigation, and screen reader support. The repository and demo show the project flow and its implementation.