GitGuide

Four references in one place: the open-source repos behind AI video and media production, plus the AI, automation and self-hosting tools around them; the apps and platforms that sit alongside them; the terminology of the business, regulatory and commercial; and the technical vocabulary plus how we actually work with Claude Code.
Study materials

New here? Start with these three.

Everything here is a card β€” click one to open the full detail. There are three tabs: Open-source repos (what you can run yourself), Apps & websites (the paid platforms), and Terminology (the words that appear in every brief and regulation β€” start there if the vocabulary is new to you).

On the repos side there are 37, and you do not need all of them. Read in this order:

  1. FFmpeg β€” the foundation. Resizing, blur-padding, burning in subtitles, joining clips, encoding. Nearly every other tool here shells out to it, so learning its filter syntax pays back everywhere.
  2. faster-whisper β€” turns speech into text with per-word timings. That word timing is what makes karaoke-style captions possible.
  3. Remotion β€” videos written as React components. This is how templated video gets made: one composition, different data, many outputs.

After that, pick by the job in front of you. Use the Purpose filter to jump to a category, and the Hardware filter to see only what your machine can actually run.

Minimum PC by hardware class

Audio / video in faster-whisper word timings drive captions FFmpeg captions + 3 sizes Finished clip
The core repos chain like this: Whisper's word timings drive the captions FFmpeg burns in. Remotion and Playwright are alternate front-ends that also render out through FFmpeg.

Repositories

Purpose
Hardware