short-video-maker

Short Video Maker exposes a Model Context Protocol (MCP) server along with REST endpoints to automate the creation of short-form videos. The server combines text-to-speech, automatic captions, background video search via Pexels, and background music to produce a finished video from a simple text prompt. You can interact with it either through MCP calls for programmatic integration or via the REST API for quick experiments and workflows. The Web UI is also available for interactive generation if you run the server locally. Typical use cases include generating YouTube Shorts, Instagram Reels, or TikTok-like clips from a concise script, with configurable voice, captions styling, and background media.

Prerequisites:

• Docker installed on your machine
• Optional: a Pexels API key for background video search
• Basic familiarity with MCP (Model Context Protocol) and REST endpoints

Install and run with Docker:

1. Pull and run the Docker image (choose the variant you need):

# Tiny image (lower resource usage)
docker run -it --rm --name short-video-maker -p 3123:3123 -e LOG_LEVEL=debug -e PEXELS_API_KEY=YOUR_KEY gyoridavid/short-video-maker:latest-tiny

# Normal image
docker run -it --rm --name short-video-maker -p 3123:3123 -e LOG_LEVEL=debug -e PEXELS_API_KEY=YOUR_KEY gyoridavid/short-video-maker:latest

# CUDA image (GPU acceleration)
docker run -it --rm --name short-video-maker -p 3123:3123 -e LOG_LEVEL=debug -e PEXELS_API_KEY=YOUR_KEY --gpus=all gyoridavid/short-video-maker:latest-cuda

2. If you prefer docker-compose, you can define the service similarly and expose port 3123. Example docker-compose snippet:

version: "3"
services:
  short-video-maker:
    image: gyoridavid/short-video-maker:latest-tiny
    environment:
      - LOG_LEVEL=debug
      - PEXELS_API_KEY=YOUR_KEY
    ports:
      - "3123:3123"
    volumes:
      - ./videos:/app/data/videos

3. To use via MCP/REST, send requests to http://localhost:3123 (or your host IP if running remotely).

Prerequisites for NPM/npx usage (alternative):

• Node.js and npm installed
• This repository may provide npm/npx usage as an alternative to Docker; if so, follow the project-specific npm instructions in the README.

Note: Replace YOUR_KEY with your actual Pexels API key. The Tiny image focuses on minimal resource usage, Normal is a balanced option, and CUDA leverages GPU acceleration for larger models."

Tips and common issues:

• Ensure your host has sufficient RAM and CPU resources (the project recommends at least 3–4 GB RAM and multiple CPU cores).
• If you encounter OOM errors with Remotion, consider using the Tiny image or adjusting CONCURRENCY and VIDEO_CACHE_SIZE_IN_BYTES as documented in the Docker usage notes.
• The REST API and MCP endpoints are designed to be used by automation tools (n8n, Zapier, custom scripts). You can pass the text prompt, language, and selected media options to customize the generated video.
• If you do not need the Web UI, you can run headlessly and rely on MCP/REST for integration.
• When using Pexels, set up a valid API key; without it, background video search may fail.
• For local testing, ensure ports are not blocked by firewalls and that the container has network access to reach external services (TTS, Whisper, Pexels, etc.).