Money-Printer

An AI-powered video generation tool that creates videos by combining generated stories, images, and text-to-speech narration.

Features

• Story generation using Ollama LLM
• Image generation with DALL-E
• Text-to-speech narration using Azure TTS
• Automated video synthesis
• Configurable output resolution and video length
• Robust error handling and retry mechanisms

Requirements

• Python 3.x
• Ollama server running locally or remotely
• DALL-E API key
• Azure TTS API key (for voice narration)
• Required Python packages (see requirements.txt)

Installation

1. Clone the repository
2. Install dependencies:

   pip install -r requirements.txt

Configuration

Set the following environment variables:

• DALLE_API_KEY: DALL-E API key (required)
• AZURE_TTS_KEY: Azure Text-to-Speech API key
• AZURE_TTS_REGION: Azure region (default: ukwest)
• AZURE_TTS_VOICE: Voice ID (default: en-US-JasonNeural)
• OLLAMA_ENDPOINT: Ollama API endpoint (default: http://localhost:11434)
• OLLAMA_MODEL: Ollama model name (default: mistral)

Usage

Run the script with:

python main.py [options]

Options:

• --ollama-endpoint: Override Ollama API endpoint
• --ollama-model: Override Ollama model
• --dalle-key: Override DALL-E API key
• --output-dir: Specify custom output directory

Output

Generated videos are saved in the output directory. The process creates:

1. A generated story using Ollama
2. A matching image using DALL-E
3. Audio narration using Azure TTS
4. Final video combining the image and narration