Short Video Maker

🚀 Short Video Generation Tool

The Short Video Generator is an open - source automated video production tool that can generate high - quality short videos based on simple text input. It integrates various advanced technologies, including text - to - speech (TTS), speech recognition, background video selection, and music matching.

🚀 Quick Start

The Short Video Generator simplifies the process of creating short videos. With just text input, it can generate professional - grade videos by integrating multiple advanced technologies.

✨ Features

1. Text - to - Speech: Convert text into natural and fluent speech using Kokoro TTS.
2. Speech Recognition: Generate accurate text subtitles through Whisper CPP.
3. Background Video Selection: Select high - quality video clips relevant to the content from the Pexels platform.
4. Video Composition: Use the Remotion framework to arrange and render video elements and output professional - level videos.

📦 Installation

Installation Requirements

• Hardware Configuration:

  • CPU: Dual - core or higher is recommended.
  • Memory: At least 4GB RAM.
  • Hard Disk Space: Reserve at least 10GB for storing generated videos and temporary files.

• Software Dependencies:

  • Node.js (Recommended version: LTS).
  • FFmpeg (Version requirement: >=2.1.3).
  • Pexels API key.

Running Methods

Method 1: Use NPX Command - Line Tool

Start the service by directly running the following command:

npx short - video - maker

Method 2: Deploy Based on Docker

Build and run a Docker container:

docker build -t short - video - maker .
docker run -p 3123:3123 short - video - maker

📚 Documentation

Environment Variable Configuration

Create a .env file in the project root directory and add the following configuration items:

• PORT = 3123 (Optional, default is 3123).
• PEXELS_API_KEY = your_api_key.
• WHISPER_MODEL = model_name.

💻 Usage Examples

Basic Usage

Request Example

{
  "scenes": [
    {
      "text": "Welcome to the Short Video Generator. Let's explore the infinite possibilities of artificial intelligence together.",
      "searchTerms": ["Technology", "Future"]
    }
  ],
  "config": {
    "paddingBack": 3000,
    "music": "Science Fiction"
  }
}

Request Method

Send a POST request using Postman or the curl tool:

curl -X POST http://localhost:3123/api/short - video \
-H "Content - Type: application/json" \
-d '{"scenes":[{"text":"Sample text","searchTerms":["Keywords"]},{"config":{"paddingBack":3000,"music":"Genre"}}'

🔧 Technical Details

• Front - end: A visual editing interface based on React (coming soon).
• Back - end: Develop a RESTful API using Node.js.
• Core Modules:
  • Video Generator: Responsible for video composition and rendering.
  • Subtitle Generator: Provides high - precision speech recognition.
  • Resource Manager: Efficiently manages and schedules Pexels video resources.

📄 License

This project follows the MIT License. For detailed information, please refer to the LICENSE file.

API Documentation

REST API

1. GET /api/short - video/:id: Get the video file with the specified ID.
2. POST /api/short - video: Create a new video (supports batch processing).
3. DELETE /api/short - video/:id: Delete the specified video file.
4. GET /api/music - tags: Get the list of available music tags.

MCP Protocol

• GET /mcp/sse: Subscribe to the real - time event stream.
• POST /mcp/messages: Send control instructions to the MCP server.

🙏 Acknowledgments

Thanks to the contributions of the following projects:

• Remotion: Provides a powerful video generation framework.
• [Whisper.cpp](https://github.com/ggml - org/whisper.cpp): An excellent speech recognition tool.
• Pexels: A rich video resource library.
• FFmpeg: A powerful audio - video processing tool.
• Kokoro.js: An advanced TTS engine.