This project provides an automated solution for generating short, high-definition videos from simple text prompts or keywords. It targets content creators and marketers looking to streamline video production, offering a full pipeline from scriptwriting and asset sourcing to final rendering with customizable audio and subtitles.

How It Works

The system employs a modular architecture, processing user input through distinct stages: AI-generated scriptwriting, sourcing royalty-free video clips, text-to-speech for narration, subtitle generation, and background music integration. It supports various LLM providers for script generation and offers flexible configuration for video dimensions, clip duration, and audio/subtitle styling.

Quick Start & Requirements

• Windows: Download and extract the one-click launch package. Run update.bat then start.bat.
• Other OS: Docker deployment is recommended (docker-compose up). Manual deployment requires cloning the repo, installing dependencies (pdm sync), and installing ImageMagick.
• Prerequisites: Python 3.x, Docker (for Docker deployment), ImageMagick (for manual deployment). Network access is crucial, and VPNs should be in global mode.
• Resources: Minimum 4-core CPU, 8GB RAM. GPU is not required.
• Docs: Web UI, API Docs

Highlighted Details

• Supports multiple LLM providers including OpenAI, Moonshot, Azure, Gemini, and Ollama.
• Offers both Web UI and API interfaces for flexibility.
• Customizable video output: 9:16 (1080x1920) and 16:9 (1920x1080) resolutions.
• Batch generation capability for creating multiple videos simultaneously.

Maintenance & Community

The project is actively maintained, with recent updates including new Azure TTS voices. It acknowledges support from "佐糖" (PicWish). Community interaction is encouraged via GitHub issues and pull requests.

Licensing & Compatibility

The project is released under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

The README advises against using Chinese characters or special characters in installation paths due to potential issues. Some LLM providers or subtitle generation methods (like Whisper) may require specific configurations or model downloads, which can be complex for less experienced users. Stability of free GPT-3.5 wrappers is noted as poor.