Built for our own launches, first.

TwelveAI ships several products. Every launch deserved a video, and most went out without one. Twelve Studio is the tool we built to change that.

Your words lead

The script is the source of truth. Scenes, captions and timing follow what you wrote, not the other way round.

Videos are composed from your text, screenshots and clips. The assistant helps with words; it never fabricates footage, numbers or customers.

Voice and captions run on TwelveAI's own pipeline. Projects stay editable, and the same project renders the same way every time.

A small set of components, each doing one job well.

Narration: Open-source speech models running on our servers, trimmed and levelled for clean audio.
Captions: Word timestamps measured from the generated voice, then matched back to your exact text.
Composition: Every frame laid out in code, so one project can be composed correctly for any screen.
Writing assistant: OpenAI, used only when you ask it to draft, polish or fit your script.

Twelve Studio sits alongside TwelveAI's banking infrastructure, business tools and developer platform.