Skip to content

Add captions and subtitles to video

Captions from the audio, styled as text layers

Captions are text layers with a schedule. Premation transcribes the composition's audio into cues, imports SRT or VTT you already have, lets you restyle and animate each cue like any text, and writes SRT or VTT back out for platforms that want a sidecar.

Download Premation

Free · AGPL-3.0-only · macOS & Windows

The panel that does it

Composition (Type)Text tool · Ctrl T

Set in motion.
Cut to the beat.

Inter Display · 148 px · -30 · normal text · 2 lines · 27 glyphs

The editor’s own panel, rebuilt on its chrome. It works: drag, toggle, queue.

How to do it

  1. 01

    Transcribe or import

    The Transcript panel on the left rail: Transcribe reads the audio layer (the model downloads once, with progress in the job tray). Or File ▸ Import an .srt / .vtt.

  2. 02

    Fix the words

    Edit the text in the Transcript panel; the cue timing follows the words. Text-based editing can also cut the video by deleting sentences.

  3. 03

    Style the cues

    Each cue is a text layer: Character panel for the face, a text animator for word-by-word entrances, a shape layer behind for a caption box.

  4. 04

    Export both ways

    Burn them in with any video export, and write the sidecar with Export Captions (.srt) or (.vtt). Chapters from markers export alongside.

Questions

Does transcription need an account or a cloud?
No. The model downloads once and runs on your machine; the job tray shows the download and the transcription progress.
Can captions be animated word by word?
Yes. A cue is a text layer, so a text animator with a range selector in words does exactly that.