Are you a social media content curator? Skip the tedious editing process and get polished video highlights in minutes. viral-clips-crew watches and listens to long-form content, extracting the most striking and potentially viral segments, ready for publication on social media.

viral-clips-crew helps you repackage your valuable content in new and engaging ways to capture attention on social media and drive traffic back to the original long-form piece. Whether you're looking to refresh your own content or recycle content from other creators, this tool streamlines the process, making content repurposing effortless and efficient.

This project requires:

• Python 3.7+
• CrewAI
• OpenAI API key and Google Gemini API key

All required Python libraries are listed in pyproject.toml.

1. Clone this repository to your local machine:

   git clone https://github.com/alexfazio/viral-clips-crew.git

2. Install Poetry to automatically manage project dependencies:

   pip install poetry

3. Install the required Python packages using Poetry:

   poetry install

4. Open .env and insert your OpenAI API key and Google Gemini API key.

After setting up, drag your desired clip into the input_files directory.

Gemini can process videos up to 1 hour in length. If you are using the OpenAI API, please ensure that the clip is less than 15 minutes in length. The current LLM context windows are approximately 15 minutes.

Run viral-clips-crew using Poetry with the following command:

```shell
poetry run python app.py
```

This will kickstart the process from beginning to completion.

Final output will be in the subtitler_output directory.

If you like this project and want to support it, please consider leaving a star. Every contribution helps keep the project running. Thank you!

The code for viral-clips-crew is intended for demonstrative purposes and is not meant for production use. The API keys are hardcoded and need to be replaced with your own. Always ensure your keys are kept secure.

Thanks to Rip&Tear for his support on the early prototype build.

This project is licensed under the terms of the MIT license.