ReadEaseAI

AI-Powered Accessibility Platform for Inclusive Reading Experiences

ReadEaseAI transforms PDF documents into personalized, accessible reading experiences tailored for individuals with diverse learning needs and disabilities. Powered by Claude AI and OpenAI, our platform adapts content in real-time to support dyslexia, visual impairments, autism, and ADHD.

✨ Features

Core Capabilities

🎯 Multi-Modal Accessibility: Four specialized reading modes optimized for different disabilities
🤖 AI-Powered Transformation: Claude Sonnet 4.5 intelligently adapts content structure and complexity
🔊 Text-to-Speech: Natural voice narration powered by OpenAI TTS
📚 PDF Intelligence: Advanced document parsing with context-aware extraction
🎨 Customizable Display: Adaptive fonts, spacing, and visual layouts
♿ WCAG Compliant: Designed with accessibility-first principles

AI Features

Semantic Simplification: Multi-level text complexity reduction
Conversational Q&A: Interactive document exploration with memory
Structured Learning: Auto-generated lessons with quizzes and visual aids
Content Chunking: ADHD-optimized micro-lessons with engagement tracking
Voice Navigation: Hands-free document interaction for blind users

🧩 Disability-Specific Modes

1. Dyslexia Support (`/dyslexia`)

Optimized for reading comprehension challenges

✅ OpenDyslexic Font: Specialized typeface with weighted bottoms
✅ Adjustable Letter Spacing: 0-5px customization via sidebar
✅ Three Reading Levels:
- Mild: Simplified sentence structure (15-20 words)
- Moderate: Middle-school vocabulary (≤15 words/sentence)
- Severe: Elementary-level language (≤10 words/sentence)
✅ Color Overlays: Reduce visual stress
✅ Text-to-Speech: Synchronized audio reading

AI Processing: Claude analyzes text complexity and rewrites content while preserving meaning (api/levels)

2. Visual Impairment / Blindness (`/blindness`)

Complete audio-first experience

✅ Conversational AI Assistant: Ask questions about document content
✅ Voice-Controlled Navigation: Spacebar-activated speech recognition
✅ PDF-to-Narration Conversion: Describes text AND images
✅ Keyboard-Only Controls: Full accessibility without mouse
✅ Conversation Memory: Context-aware dialogue across questions

Interaction Flow:

Press & hold SPACEBAR → Ask question
Release SPACEBAR → Submit query
Claude analyzes document → Generates spoken answer
Press SPACEBAR during playback → Interrupt/new question

AI Processing:

/api/narrate - Converts PDF to conversational narration
/api/conversation - Handles Q&A with document context

3. Autism Spectrum Support (`/autism`)

Structured, predictable learning environment

✅ Chunked Content: Manageable sections with clear progression
✅ Concrete Examples: Abstract concepts explained literally
✅ Visual Vocabulary Aids: Term → Definition → Example tables
✅ Interactive Assessments:
- True/False with explanations
- Multiple choice (4 options)
- Short answer with rubrics
✅ "Draw-It" Visual Learning: Diagram instructions with labels
✅ Spaced Repetition: Auto-generated review schedule
✅ Age Personalization: Content adapts to grade level

AI Processing: /api/generate-lesson returns strict JSON schema:

{
  "summary": ["5-7 bullet points"],
  "vocabulary": [{"term": "...", "definition": "...", "example": "..."}],
  "questions": {
    "trueFalse": {...},
    "multipleChoice": {...},
    "shortAnswer": {...}
  },
  "drawIt": {"title": "...", "labels": [...], "caption": "..."},
  "reviewPlan": ["Tomorrow", "In 3 days"]
}

4. ADHD Support (`/adhd`)

Engagement-optimized micro-learning

✅ Micro-Lessons: 100-150 word chunks with emoji anchors
✅ Focus Timer: 25-minute Pomodoro sessions
✅ Gamified Progress: 🌱→🌿→🌳→🏆 milestone tracking
✅ Active Highlighting: Visual focus on current section
✅ Break Reminders: Prevents burnout
✅ Interactive Checkpoints: Quick questions after each chunk
✅ 6 Voice Options: Personalize narration (alloy, echo, fable, onyx, nova, shimmer)

Content Structure:

📌 Quick Summary (3-5 bullets)
  ↓
🎯 Micro-Lesson 1 (100-150 words)
  ↓
✅ Checkpoint Question
  ↓
🎯 Micro-Lesson 2...

AI Processing: /api/convert transforms PDFs with ADHD-specific constraints (short sentences, emojis, whitespace optimization)

🛠 Tech Stack

Frontend

Technology	Version	Purpose
Next.js	15.2.4	React framework with App Router
React	19.0.0	UI library
TypeScript	5.x	Type safety
Tailwind CSS	4.x	Utility-first styling
Radix UI	Latest	Accessible component primitives
Lucide React	0.487.0	Icon library
react-markdown	10.1.0	Markdown rendering

Backend / AI

Technology	Version	Purpose
Claude AI	Sonnet 4.5	Document processing, Q&A, content transformation
OpenAI	4.91.1	Text-to-Speech (TTS-1, TTS-1-HD)
pdf-parse	1.1.1	PDF text extraction
pdfjs-dist	5.1.91	PDF rendering
Web Speech API	Native	Browser speech recognition (blindness mode)

Infrastructure

Vercel: Deployment platform
Vercel Analytics: Usage tracking
Environment Variables: Secure API key management

🚀 Getting Started

Prerequisites

Node.js 18.x or higher
pnpm (recommended) or npm/yarn
Anthropic API Key (Get one here)
OpenAI API Key (Get one here)

📦 Installation

Clone the repository

git clone https://github.com/yourusername/ReadEaseAI.git
cd ReadEaseAI

Install dependencies
```
pnpm install
# or
npm install
```

Set up environment variables

cp .env.example .env

Edit .env and add your API keys:

ANTHROPIC_API_KEY=your_anthropic_key_here
OPENAI_API_KEY=your_openai_key_here

Run the development server
```
pnpm dev
# or
npm run dev
```
Open your browser Navigate to http://localhost:3000

🔑 Environment Variables

Create a .env file in the root directory:

# Required
ANTHROPIC_API_KEY=sk-ant-xxxxx
OPENAI_API_KEY=sk-xxxxx

# Optional
NODE_ENV=development
NEXT_PUBLIC_ANALYTICS_ID=your_analytics_id

💻 Usage

Basic Workflow

Select Accessibility Mode: Choose from Dyslexia, Blindness, Autism, or ADHD on the home page
Upload PDF: Drag & drop or select a PDF document
AI Processing: Document is analyzed and transformed (10-30 seconds)
Customized Reading: Interact with content optimized for your needs

Mode-Specific Tips

Dyslexia Mode

Use the sidebar to adjust letter spacing (find your sweet spot)
Try different reading levels to match your comfort
Enable TTS for audio reinforcement

Blindness Mode

Press and HOLD spacebar to ask questions
Navigate with Tab/Enter for keyboard-only control
Ask: "Summarize page 3" or "What is the main argument?"

Autism Mode

Progress through sections linearly for predictable structure
Use the Draw-It section for visual reinforcement
Review vocabulary before diving into content

ADHD Mode

Set a realistic timer goal (start with 10 minutes)
Complete checkpoints to maintain engagement
Celebrate milestones (track your 🌱→🌿→🌳→🏆 journey)

🔌 API Documentation

Core Endpoints

`POST /api/narrate`

Convert PDF to audio narration

Request:
{
  file: Buffer,          // PDF file as buffer
  fileName: string       // Original filename
}

Response:
{
  narration: string,     // Conversational narration text
  success: boolean
}

Model: Claude Sonnet 4.5 (4000 tokens)

`POST /api/conversation`

Q&A with document context

Request:
{
  question: string,
  documentContext: string,
  conversationHistory: Message[]
}

Response:
{
  answer: string,        // Audio-optimized response
  success: boolean
}

Model: Claude Sonnet 4.5 (1000 tokens)

`POST /api/generate-lesson`

Generate autism-friendly structured lesson

Request:
{
  file: Buffer,
  age: number,
  sectionNumber: number
}

Response:
{
  summary: string[],
  vocabulary: Vocabulary[],
  questions: Questions,
  drawIt: DrawItSection,
  reviewPlan: string[]
}

Model: Claude Sonnet 4.5 (4000 tokens)

`POST /api/convert`

Transform PDF for ADHD mode

Request:
{
  file: Buffer,
  mode: "adhd" | "dyslexic" | "deaf" | "autism"
}

Response:
{
  converted: string,     // Markdown content with "---" separators
  success: boolean
}

Model: Claude Sonnet 4.5 (4000 tokens)

`POST /api/parse`

Extract and clean PDF text for dyslexia

Request:
{
  file: Buffer,
  useAI?: boolean        // Default: false
}

Response:
{
  text: string,          // Cleaned, dyslexia-friendly text
  success: boolean
}

Model: Claude Sonnet 4.5 (4096 tokens) - if useAI=true

`POST /api/levels`

Adjust reading complexity level

Request:
{
  text: string,
  level: "mild" | "moderate" | "severe"
}

Response:
{
  summary: string,
  rephrased: string
}

Model: Claude Sonnet 4.5 (4096 tokens, temp: 0.3)

`POST /api/tts`

Generate Text-to-Speech audio

Request:
{
  text: string,
  voice?: "alloy" | "echo" | "fable" | "onyx" | "nova" | "shimmer"
}

Response:
{
  audio: string          // Base64-encoded MP3
}

Model: OpenAI TTS-1

🏗 Architecture

Project Structure

ReadEaseAI/
├── src/
│   ├── app/
│   │   ├── api/                    # API Routes
│   │   │   ├── ai-process/         # Reading level simplification
│   │   │   ├── conversation/       # Blindness Q&A
│   │   │   ├── convert/            # ADHD content transformation
│   │   │   ├── generate-lesson/    # Autism lesson generation
│   │   │   ├── narrate/            # PDF to narration
│   │   │   ├── parse/              # PDF parsing
│   │   │   ├── tts/                # Text-to-Speech
│   │   │   └── tts_adhd/           # ADHD-specific TTS
│   │   │
│   │   ├── adhd/                   # ADHD mode pages
│   │   ├── autism/                 # Autism mode pages
│   │   ├── blindness/              # Blindness mode pages
│   │   ├── dyslexia/               # Dyslexia mode pages
│   │   ├── processed/              # Dyslexia reader view
│   │   ├── reader/                 # ADHD chunked reader
│   │   ├── refined/                # Alternative dyslexia reader
│   │   │
│   │   ├── fonts/                  # Font files (OpenDyslexic)
│   │   ├── layout.tsx              # Root layout with global sidebar
│   │   ├── page.tsx                # Home page (mode selection)
│   │   └── globals.css             # Global styles
│   │
│   ├── components/
│   │   ├── ui/                     # Reusable UI components
│   │   │   ├── button.tsx
│   │   │   ├── card.tsx
│   │   │   ├── dialog.tsx
│   │   │   ├── FileUploader.tsx
│   │   │   ├── fontSelector.tsx
│   │   │   ├── input.tsx
│   │   │   ├── slider.tsx
│   │   │   └── ...
│   │   └── app-sidebar.tsx         # Global sidebar (font/spacing controls)
│   │
│   ├── hooks/
│   │   └── use-mobile.ts           # Mobile detection hook
│   │
│   └── lib/
│       └── utils.ts                # Utility functions
│
├── public/
│   └── fonts/                      # Public font assets
│
├── .env                            # Environment variables (gitignored)
├── .env.example                    # Environment template
├── package.json                    # Dependencies
├── tsconfig.json                   # TypeScript config
├── next.config.ts                  # Next.js config
├── tailwind.config.ts              # Tailwind config
├── components.json                 # shadcn/ui config
└── README.md                       # This file

Data Flow

User uploads PDF
       ↓
Next.js API Route
       ↓
PDF Parser (pdf-parse)
       ↓
Claude AI Processing
  ├─ Narration generation
  ├─ Text simplification
  ├─ Lesson structuring
  └─ Q&A answering
       ↓
OpenAI TTS (optional)
       ↓
Client-side rendering
  ├─ Custom fonts
  ├─ Accessibility controls
  └─ Interactive UI

🤝 Contributing

We welcome contributions from the community! Here's how to get started:

Development Workflow

Fork the repository
Create a feature branch
```
git checkout -b feature/amazing-feature
```
Make your changes
- Follow TypeScript best practices
- Maintain accessibility standards
- Test across all disability modes

Commit with descriptive messages

git commit -m "feat(dyslexia): add color overlay customization"

Push to your fork
```
git push origin feature/amazing-feature
```
Open a Pull Request

Code Standards

TypeScript: Strict mode enabled, no implicit any
Accessibility: WCAG 2.1 Level AA compliance
Formatting: Use Prettier (run pnpm format)
Linting: Pass ESLint checks (pnpm lint)
Testing: Include tests for new features

Areas We Need Help

🌍 Internationalization: Support for non-English languages
📱 Mobile Optimization: Enhanced touch interfaces
🎨 Design: Improved visual accessibility
🧪 Testing: Unit/integration test coverage
📖 Documentation: Tutorials and user guides
♿ Accessibility: Additional disability modes (e.g., hearing impairment)

🗺 Roadmap

See FUTURE_ROADMAP.md for detailed expansion plans including:

Multi-Agent Systems: AI orchestration with LangGraph
RAG Integration: Vector database for document intelligence
MCP Support: Model Context Protocol for extensibility
Advanced Features: Eye-tracking, spatial audio, haptic feedback
Personalization: User profiles and adaptive learning

📊 Performance

PDF Processing: 10-30 seconds (varies by document size)
Text-to-Speech: Real-time streaming
API Response Times:
- Narration: ~15-20s (4000 tokens)
- Q&A: ~3-5s (1000 tokens)
- Lesson Generation: ~20-25s (4000 tokens)
Browser Support: Chrome, Firefox, Safari (latest 2 versions)

🔒 Privacy & Security

No Data Storage: PDFs are processed in-memory and discarded
API Key Security: Keys stored in environment variables, never exposed to client
HTTPS Only: All communication encrypted in transit
Client-Side Processing: Where possible, processing happens in browser (Web Speech API)

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

Anthropic: For Claude AI and accessibility research
OpenAI: For TTS capabilities
Radix UI: For accessible component primitives
Vercel: For hosting and deployment infrastructure
OpenDyslexic: For the dyslexia-friendly font

📞 Support

Issues: GitHub Issues
Discussions: GitHub Discussions
Email: support@readease.ai

🌟 Star History

If this project helps you or someone you know, please consider starring it on GitHub! ⭐

Built with ❤️ for accessibility and inclusion

Website • Documentation • Demo

Name		Name	Last commit message	Last commit date
Latest commit History 10 Commits
public		public
src		src
.gitignore		.gitignore
.vercel-deploy		.vercel-deploy
FUTURE_ROADMAP.md		FUTURE_ROADMAP.md
README.md		README.md
components.json		components.json
next.config.ts		next.config.ts
package-lock.json		package-lock.json
package.json		package.json
pnpm-lock.yaml		pnpm-lock.yaml
postcss.config.mjs		postcss.config.mjs
tsconfig.json		tsconfig.json

baveet256/ReadEaseAI

Folders and files

Latest commit

History

Repository files navigation

ReadEaseAI

📋 Table of Contents

✨ Features

Core Capabilities

AI Features

🧩 Disability-Specific Modes

1. Dyslexia Support (/dyslexia)

2. Visual Impairment / Blindness (/blindness)

3. Autism Spectrum Support (/autism)

4. ADHD Support (/adhd)

🛠 Tech Stack

Frontend

Backend / AI

Infrastructure

🚀 Getting Started

Prerequisites

📦 Installation

🔑 Environment Variables

💻 Usage

Basic Workflow

Mode-Specific Tips

Dyslexia Mode

Blindness Mode

Autism Mode

ADHD Mode

🔌 API Documentation

Core Endpoints

POST /api/narrate

POST /api/conversation

POST /api/generate-lesson

POST /api/convert

POST /api/parse

POST /api/levels

POST /api/tts

🏗 Architecture

Project Structure

Data Flow

🤝 Contributing

Development Workflow

Code Standards

Areas We Need Help

🗺 Roadmap

📊 Performance

🔒 Privacy & Security

📄 License

🙏 Acknowledgments

📞 Support

🌟 Star History

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Contributors 3

Uh oh!

Languages

1. Dyslexia Support (`/dyslexia`)

2. Visual Impairment / Blindness (`/blindness`)

3. Autism Spectrum Support (`/autism`)

4. ADHD Support (`/adhd`)

`POST /api/narrate`

`POST /api/conversation`

`POST /api/generate-lesson`

`POST /api/convert`

`POST /api/parse`

`POST /api/levels`

`POST /api/tts`

Packages