AI YouTube Shorts Generator is a Python tool that automatically converts long-form YouTube videos into viral-ready vertical shorts by detecting highlight moments and auto-cropping them for TikTok, Reels, and YouTube Shorts.
The tool solves the time-intensive problem of manually identifying and editing engaging moments from lengthy videos. It combines GPT-class language model highlight detection with Whisper speech transcription to identify the most viral-worthy segments, then automatically reframes them to 9:16 vertical aspect ratio. The approach eliminates the need for manual clip selection and manual vertical cropping, delivering ranked clips ready for social distribution without watermarks or per-clip processing fees.
Creators, agencies, and developers should choose this tool if they want to avoid subscription costs and per-minute caps imposed by commercial alternatives. It suits projects requiring batch processing of multiple videos, full control over the virality detection algorithm, and self-hosted deployment. The README explicitly compares it to Opus Clip, Vidyo.ai, Klap, and SubMagic, highlighting that this tool offers unlimited processing at no subscription cost, no watermarks on free usage, an editable highlight algorithm rather than a black box, support for any output aspect ratio and resolution, batch processing capabilities, JSON API output, self-hosting, and white-label licensing under MIT terms—features the commercial alternatives either restrict to paid tiers or do not offer.
The project maintains active development with regular commits addressing feature requests and bug fixes. The codebase shows consistent refinement of the core highlight detection and cropping pipeline. Documentation is kept current with usage examples and API integration guidance. The tool integrates established external services for transcription and LLM inference rather than reimplementing those components, focusing engineering effort on the clip selection and vertical reframing logic that differentiates it from competitors.