• 

  LumiTextAI

LumiText - AI Accessibility Helper

Inspiration

I was inspired to create LumiText after noticing how many people struggle with understanding complex texts or accessing information quickly. Accessibility tools are improving, but there’s still a gap for AI-driven assistants that simplify text, provide translations, and support people with reading or language difficulties. LumiText was born to make digital content more understandable and inclusive for everyone.

What it does

LumiText is an AI-powered tool that helps users read, understand, and interact with text more easily. Key features include:

• Simplifying complex paragraphs into easy-to-read summaries.
  
• Translating text into multiple languages.
  
• Providing explanations for difficult terms or phrases.
  
• Helping users with accessibility needs to navigate content faster and more effectively.

How I built it

I built LumiText using:

• Python for backend logic and AI interactions.
  
• Streamlit for creating an interactive and user-friendly web interface.
  
• Gemini API / OpenAI GPT API for natural language processing, text simplification, and translations.
  
• Modular code design with app.py for the main interface and utils.py for helper functions.
  

I also used .env files to securely store API keys and ensured smooth integration between the frontend and AI backend.

Challenges I ran into

• API rate limits: I had to optimize requests and caching to prevent hitting the Gemini API limits.
  
• Text simplification quality: Ensuring that summaries remained accurate while being easier to read was tricky.
  
• Deployment issues: Deploying on Streamlit Community Cloud required careful configuration of file paths and environment variables.
  

Accomplishments that I'm proud of

• Built a fully functional AI accessibility assistant that works in a browser.
  
• Implemented multi-language support and real-time text simplification.
  
• Created a clean and interactive interface that is easy for users of all ages to use.
  

What I learned

• How to integrate AI APIs into a live web app effectively.
  
• Best practices for separating app logic and utilities for cleaner code.
  
• Deployment strategies for Streamlit apps and handling environment variables securely.
  

What's next for LumiText

• Add voice input and text-to-speech features for enhanced accessibility.
  
• Improve AI summarization accuracy and context awareness.
  
• Expand language support and possibly integrate with browser extensions.
  
• Collect user feedback to refine UX and add personalization features.
  

Built With

• blip
• gemini
• ollama
• python