Inspiration
Visually impaired users face difficulty when typing Braille using a standard QWERTY keyboard. It's easy to miss or misplace dot combinations, leading to incorrect words and frustrating communication. I built this project to solve that challenge by offering a simple yet intelligent Braille AutoCorrect system tailored to their needs, improving both speed and accuracy in real-time.
What I Built
This project is a real-time Braille input autocorrect and suggestion system. It listens for QWERTY keystrokes used to emulate Braille dot patterns and provides:
Auto-correction of common Braille input errors
Top 3 word suggestions based on the input pattern
A simple text interface for real-time typing feedback
Tech Stack
Python.
.txt file for dictionary and word frequency data
NLP logic: Levenshtein Distance, edit distance for error tolerance
What I Learned
Designing user-first assistive tools: how important simplicity and responsiveness are for accessibility
Implementing real-time feedback loops with minimal delay
Using string similarity algorithms like edit distance and prefix matching for correction logic
Challenges Faced
Mapping multiple dot combinations to accurate words without overcorrecting
Handling real-time keystroke detection across all six Braille dots
Designing an interface that could realistically assist a visually impaired user without full screen reader integration (future step)
Future Work
Adding GUI(using Tkinter).
Integrate LLM-based correction for contextual predictions
Add audio feedback using text-to-speech for suggestions
Make the system compatible with screen readers (e.g., NVDA, JAWS)
Turn this into a browser extension or mobile app for wider usability
Log in or sign up for Devpost to join the conversation.