From recording to finished video
Five steps, every one of them authored as text and rendered by a deterministic pipeline.
Capture your product
Record a terminal session or a browser walkthrough, or point decodx at clips you already have. Captures are cached, so you record once.
Author the script
Narration lives in the spec as plain text you can review in a pull request. No dragging waveforms, no re-syncing audio when the wording changes.
Give it a voice
Premium AI narration in dozens of voices, with an optional talking-head avatar presenter. Every voice is available on every plan.
Brand and polish
Title cards, gradient backgrounds, background music, animated captions, and chapters — all declared in the spec, applied the same way every render.
Render and share
Out comes final.mp4 plus captions, chapters, and a thumbnail. Change one scene and re-render in seconds — cached assets are reused, free.
Everything in the box
The building blocks that turn a screen recording into something worth watching.
Deterministic renders
The same spec and assets always produce the same MP4. No model in the render loop.
Content-hash cache
Only what changed re-renders. Iterating on a video costs seconds, not another export.
Avatar presenter
Add a talking-head avatar that can stay on screen for the whole scene, or just a beat.
Captions & chapters
Accurate burned-in captions and chapter markers generated from your narration.
Background music
A built-in music library with automatic ducking under narration.
Brand kit
Logos, colors, and fonts applied consistently across every scene and video.
Thumbnails
A poster frame produced with the video, ready for your player or docs.
One spec, many outputs
The same source renders your video, its captions, and a written guide together.