LexiFlow

LexiFlow uses the Google Gemini API to transform complex text into accessible learning through AI-powered simplification, dyslexia-friendly rewrites, summaries, quizzes, and voice-guided reading.

Comment

Inspiration

Reading should never be a barrier to learning. However, many students, including people with dyslexia, ADHD, English language learners, and struggling readers, often find complex text difficult to understand. While modern AI tools can answer questions, they usually require users to know what to ask and how to write effective prompts.

We wanted to build a platform where AI works in the background, allowing users to simply focus on reading. This led to the creation of LexiFlow, an AI-powered accessibility platform that transforms complex content into personalized reading experiences using the power of Google's Gemini API.

What it does

LexiFlow is an AI-powered accessibility platform that makes difficult content easier to understand.

Users can paste text or upload PDFs, automatically analyze readability using the Flesch-Kincaid Grade Level, simplify content at multiple reading levels, generate dyslexia-friendly rewrites, listen using built-in text-to-speech, create summaries, extract key points, generate comprehension quizzes, and personalize their reading experience through accessibility settings.

At the core of these features is the Google Gemini API, which intelligently rewrites, summarizes, explains, and generates educational content while preserving the original meaning of the text.

The platform also includes a progress dashboard to help users monitor their learning journey over time.

How we built it

The frontend was built using React, Vite, and Tailwind CSS, providing a responsive and accessibility-first user interface.

The backend was developed with FastAPI and Python, exposing stateless REST APIs for text processing and AI-powered features.

The Google Gemini API serves as the intelligence behind LexiFlow. It powers text simplification, dyslexia-friendly rewriting, summaries, key-point extraction, contextual explanations, and comprehension quiz generation. Carefully designed prompts ensure that Gemini produces accurate, structured, and easy-to-understand responses while maintaining the original context of the text.

For document processing, PyMuPDF extracts text from uploaded PDFs, while TextStat calculates readability scores locally without requiring AI, making readability analysis both fast and cost-effective.

The application is deployed using Vercel and follows a lightweight, stateless architecture without requiring a database or user authentication.

How Gemini API was used

LexiFlow is powered by the Google Gemini API, which serves as the core intelligence behind the platform. Gemini is used to simplify complex text into multiple reading levels, generate dyslexia-friendly rewrites, produce concise summaries, extract key points, create comprehension quizzes, and explain difficult words or phrases in context. By combining carefully engineered prompts with Gemini's language understanding capabilities, LexiFlow provides personalized and accessible reading experiences while preserving the original meaning of the content.

Challenges we ran into

One of our biggest challenges was designing prompts that consistently generated structured and reliable responses from the Gemini API. Since LexiFlow relies heavily on AI-generated educational content, maintaining factual accuracy while simplifying complex information required several rounds of prompt engineering and testing.

We also spent considerable time configuring frontend-backend communication, handling CORS, deploying both services on Vercel, and ensuring smooth API integration.

Another challenge was creating accessibility features that remained intuitive while supporting users with different reading abilities and learning preferences.

Accomplishments that we're proud of

We're proud of building a complete accessibility platform rather than just another AI chatbot.

By leveraging the Gemini API, LexiFlow combines intelligent text simplification, dyslexia-friendly rewriting, AI-generated summaries, key points, contextual explanations, quizzes, and read-aloud functionality into one seamless workflow.

We're also proud of delivering a lightweight, stateless application that requires no database or user accounts while solving a meaningful real-world educational problem.

What we learned

This project gave us valuable hands-on experience with Google Gemini API integration, prompt engineering, FastAPI backend development, React application architecture, REST API design, accessibility-focused UI development, and deploying full-stack AI applications on Vercel.

Most importantly, we learned that AI creates the greatest impact when it removes barriers to learning instead of adding complexity for users.

What's next for LexiFlow

We plan to continue expanding LexiFlow with OCR support for scanned documents, EPUB support, multilingual text simplification powered by Gemini, browser extensions for webpage accessibility, offline AI capabilities, teacher dashboards, classroom analytics, personalized learning recommendations, and adaptive reading profiles.

Our long-term vision is to evolve LexiFlow into a comprehensive AI-powered accessibility platform that enables every learner to access information in the format that works best for them, regardless of their reading ability or learning style.

Built With

Share this project:

Updates