AI Short Video Creator

Tool to automatically create short clips with a background video and AI-generated captions. Can be used for YouTube Shorts, TikTok, Instagram Reels, Snapchat Spotlight.

No API keys are required, the videos are processed locally on your computer.

Table of Contents

• Example
• Requirements
• Installation
• Usage
• Contributing
• License

Example Output

Requirements

• Python >=3.11 - Download Here
• FFmpeg is bundled through imageio-ffmpeg, so a separate installation is not required. A full system FFmpeg installation is optional; its ffprobe improves media probing when available.

Captions are transcribed locally with NVIDIA's Parakeet TDT 0.6B v3 model running on onnx-asr (works on Windows, macOS and Linux, CPU-only by default). The model (~600 MB) is downloaded automatically from Hugging Face on first run and cached locally.

Installation

1. Clone the repository:

 git clone https://github.com/sw-aka/Short-Video-Creator.git

1. Install the package:

 pip install -e .

Usage

1. Move main videos (.mp4, .mov, .mkv, .avi, or .webm) into INPUT_VIDEOS
2. Move background videos in any of those formats into BACKGROUND_VIDEOS
3. Run the CLI:

svc

1. The edited videos are saved in OUTPUT_VIDEOS
2. If any video fails to process, the tool prints a failure summary and exits with a nonzero status.

The compatibility command python main.py and python -m short_video_creator provide the same folder-batch workflow. Directory defaults are relative to the current working directory and can be changed with --input-dir, --output-dir, --backgrounds-dir, and --processes.

Library API

from short_video_creator import Settings, create_short

output = create_short(
    input_video="clip.mp4",
    output_path="out/short.mp4",
    backgrounds_dir="backgrounds",
    settings=Settings(font_size=120),
)

create_short returns the output Path. Catch ShortVideoError or one of its exported subclasses to handle per-clip failures. The default caption font is bundled; pass Settings(font_path=Path("font.ttf")) to use a custom font.

Contributing

1. Fork the repository.
2. Create a new branch: git checkout -b feature-name.
3. Make your changes.
4. Push your branch: git push origin feature-name.
5. Create a pull request.

License

This project is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Dependencies are installed from PyPI under their own licenses and are not redistributed here. The Parakeet TDT 0.6B v3 model is released by NVIDIA under the CC-BY-4.0 license.