Kalam2

Kalam2 is a full-stack handwriting generation application. It consists of a FastAPI backend for generating SVG handwriting using machine learning, and a Next.js frontend for user interaction and visualization.

Features

• Generate realistic handwriting in multiple styles
• Download handwriting as SVG (including A4 page layout)
• Customizable style, color, and layout options
• API endpoints for integration

---

Backend (FastAPI)

Setup

1. Install uv
2. Install dependencies and activate virtual environment:

   uv sync
   source .venv/bin/activate

3. Start the backend server:

   python main.py

   The API will be available at http://localhost:8000/api/v1.

Main Endpoints

• POST /api/v1/handwriting/generate — Generate handwriting from text lines
• POST /api/v1/handwriting/a4page — Generate handwriting formatted for A4 pages
• GET /api/v1/handwriting/styles — List available handwriting styles
• GET /api/v1/handwriting/styles/{style_id} — Get details and preview for a style

---

Frontend (Next.js)

Setup

1. Install Node.js.
2. Install dependencies:

   pnpm install
   # or
   npm install

3. Start the development server:

   pnpm dev
   # or
   npm run dev

   The app will be available at http://localhost:3000.

Usage

• Enter text and select a handwriting style to generate handwriting.
• Use the A4 page generator for multi-page output.
• Download generated SVGs for use in documents or printing.

---

Deployment

• The frontend can be deployed on Vercel or any static hosting supporting Next.js.
• The backend can be deployed on any server supporting Python and FastAPI.

---

License

MIT License

---

Acknowledgements

• Built with FastAPI, Next.js, and TensorFlow.
• Handwriting model and styles are from https://github.com/sjvasquez/handwriting-synthesis.