This Python tool automates the creation of YouTube Shorts from long-form videos, targeting content creators and video editors. It leverages AI to identify engaging segments, transcribe audio, detect speakers, and perform vertical cropping, significantly reducing manual effort in repurposing content.

How It Works

The tool downloads YouTube videos, transcribes them using Whisper, and then employs GPT-4 to pinpoint the most compelling highlights. Speaker detection is integrated, and the identified segments are automatically cropped vertically to suit the Shorts format, optimizing them for viewer engagement.

Quick Start & Requirements

• Install: pip install -r requirements.txt
• Prerequisites: Python 3.7+, FFmpeg, OpenCV, OpenAI API key.
• Setup: Clone repo, create/activate virtual environment, install dependencies, set OPENAI_API in .env.
• Usage: python main.py (prompts for YouTube URL).
• Resources: Requires an OpenAI API key.
• Docs: vadoo.tv API, YouTube Tutorial, Medium Tutorial

Highlighted Details

• Automates video download from YouTube URLs.
• Utilizes OpenAI's GPT-4 for highlight extraction.
• Integrates Whisper for video transcription.
• Performs speaker detection and vertical cropping for Shorts.

Maintenance & Community

• Project is at v0.1, indicating early development.
• Contributions are welcome via pull requests.
• Issues can be reported on the GitHub repository.

Licensing & Compatibility

• Licensed under the MIT License.
• Permissive for commercial use and integration into closed-source applications.

Limitations & Caveats

This is a v0.1 release and may contain bugs. Specific performance or accuracy benchmarks for highlight extraction are not provided.