shorts-video-maker

To make Shorts from a YouTube video, this Skill automates highlight extraction by downloading the source with yt-dlp, transcribing audio with Whisper, and cropping three 9:16 vertical clips. You can use automatic transcript-based selection or manually specify segments.

Yes, you need Whisper and ffmpeg installed to generate vertical Shorts. The Skill requires Python 3.8+ and uses Whisper for audio transcription and ffmpeg for video downloading and 9:16 cropping operations.

Yes, you can manually select video segments for clip generation. The Skill supports script-driven manual selection, allowing you to specify exact timestamps or use a transcript file to define the three highlight clips instead of relying on automatic extraction.

The best way to automate transcription and 9:16 cropping is running the full pipeline, which uses yt-dlp for video downloading, Whisper for transcription, and ffmpeg for vertical cropping to output three ready-to-upload Shorts.

Yes, yt-dlp supports downloading long YouTube videos for Shorts creation. The Skill uses yt-dlp to fetch the source video, then processes the file with Whisper and ffmpeg to extract and crop three highlight segments into vertical clips.

Transcription may fail if Whisper or ffmpeg dependencies are improperly configured or if the Python environment is below 3.8. Ensure all required dependencies are correctly installed and the source video downloaded successfully before running the clip generation pipeline.