AI-Powered Reel Generator

An innovative AI-powered tool that generates full-featured video reels from simple text prompts. This project combines multiple cutting-edge AI technologies to automate the time-consuming process of reel creation.

🌟 Features

• Text-to-Reel Generation: Convert simple text prompts into complete video reels
• AI-Powered Components:
  • Image generation using state-of-the-art models
  • Voice synthesis and audio generation
  • Automatic transcription and subtitle creation
• Customization Options: Control various aspects of the generated reels
• User Authentication: Secure user accounts with JWT-based authentication

🛠️ Tech Stack

Frontend

• Framework: React with Vite
• State Management: React Context API
• Styling: CSS/SCSS
• API Communication: Fetch API
• Authentication: JWT token-based auth

Backend

• Framework: Django with Django REST Framework
• Database: PostgreSQL
• Authentication: JWT with Simple JWT
• Media Processing: MoviePy

AI Services

• Image Generation: DreamShaper XL
• Audio Generation: Suno Bark
• Transcription: Whisper
• Orchestration: LangChain
• Text Generation: Cohere

Infrastructure

• Containerization: Docker and Docker Compose
• Development Environment: Fully containerized local development

🚀 Getting Started

Prerequisites

• Docker and Docker Compose
• API keys for:
  • Cohere
  • LangChain
  • LangSmith

Quick Start

1. Clone the repository

   git clone https://github.com/yourusername/ai-reel-generator.git
   cd ai-reel-generator

2. Set up environment variables

   Create a .env.docker file in the root directory with the following variables:

   # Django settings
   DEBUG=1
   SECRET_KEY=your_secret_key
   
   # Database
   DATABASE_URL=postgresql://postgres:Fakhar27@db:5432/genAI
   
   # AI Service API Keys
   COHERE_API_KEY=your_cohere_api_key
   LANGCHAIN_API_KEY=your_langchain_api_key
   LANGCHAIN_PROJECT=your_langchain_project
   LANGSMITH_API_KEY=your_langsmith_api_key
   

3. Launch the application

   docker-compose up --build

4. Access the application

   • Frontend: http://localhost:5173
   • Backend API: http://localhost:8000
   • Admin interface: http://localhost:8000/admin

Stopping the Application

docker-compose down

To remove all data including database volumes:

docker-compose down -v

📁 Project Structure

ai-reel-generator/
├── backend/                # Django backend
│   ├── genAI/              # Main Django app
│   ├── Dockerfile          # Backend container configuration
│   └── requirements.txt    # Python dependencies
├── frontend/               # React frontend
│   ├── src/                # React source code
│   ├── public/             # Static assets
│   └── Dockerfile          # Frontend container configuration
├── ai-services/            # AI service containers
│   ├── image-generation/   # Image generation service
│   ├── sound-generation/   # Sound generation service
│   └── whisper/            # Transcription service
├── docker-compose.yml      # Docker Compose configuration
└── .env.docker            # Environment variables for Docker

🔧 Advanced Configuration

Scaling AI Services

For production deployments, you may want to scale AI services independently. This can be achieved by modifying the docker-compose.yml file:

services:
  image-service:
    deploy:
      replicas: 2

GPU Support

For improved performance, enable GPU support by modifying the AI service sections in docker-compose.yml:

services:
  image-service:
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]

🙏 Acknowledgements

• Cohere for text generation capabilities
• LangChain for AI orchestration
• Suno Bark for audio generation
• OpenAI Whisper for transcription
• DreamShaper XL for image generation