Transcript, not a blank page
Speech is recognised and every word is aligned to the sound, so cue boundaries land where the voice actually breaks.
AI video editor
AI video editor built around subtitles
Not a general-purpose timeline that happens to do captions. One job, done all the way through: the words, the timing, the look and the export.
Open the editor →The draft is already made
You do not start from an empty timeline. You start from a finished version and argue with it.
A timeline you can read
The waveform and the frame strip sit under the video, so you see what you are dragging a cue against.
Styling that is already applied
A design is chosen and laid out before you arrive; changing it is picking a preset, not building one.
What the AI actually does
Three jobs, each with a page of its own — and each one you can overrule.
Placement clear of faces
Frames are analysed for faces and objects, and captions are laid out where they do not cover them.
AI headlines
Two to four words carrying the point, timed to the voice and dropped on the frame as a headline.
B-roll and zooms
Your own clip or still over the talking head, plus zooms placed on the timeline.
Word-level highlighting
Words light up as they are spoken, because the timing is per word rather than per line.
Everything in one workspace
Transcript
Fix a word once and every place it appears follows.
Timeline
Drag cue boundaries against the waveform and the frame strip.
Styles
Designer presets, or build one: font, colour, outline, shadow, animation.
Fonts
A full font library, with the preview rendering what the export will.
Headlines
Key phrases large on the frame, on their own track.
B-roll
Overlays and zooms over the talking head.
Placement
Automatic layout clear of faces, moved by hand when the shot needs it.
Export
SRT, VTT and ASS files, or MP4, MOV, H.265 and WebM with the captions burned in.
On a tablet the side panels open over the workspace; on a phone the interface splits into Subtitles, Style and Timeline tabs.
What it is not
It is not a general video editor. There is no multi-track cutting, no colour grading, no transitions — the video you upload is the video you get back, with captions on it.
It is not an SRT file editor either. It works on subtitles it generated from your video; if you already have a finished .srt and only need to fix the text and timings, that is a separate tool and it needs no account.
It does not run on your machine. Recognition, frame analysis and rendering happen on the service, so a long video does not depend on your laptop.
From upload to export
Drop in a video and the pipeline starts on its own: audio is extracted, speech recognised, words aligned to the sound, frames analysed, and a first design laid out. The current stage is shown while it runs.
Edit what only you can decide — the wording, the emphasis, the look, and where the captions sit in the frame.
Export as a subtitle file for a player that draws captions itself, or as video with them burned into the frames.
FAQ
There is a free plan: 2 videos a month, up to 3 minutes each, with no card required. Longer videos and a higher monthly limit come with Pro.
Beautiful captions.
Better videos.
Captions Space generates AI subtitles, styles them with designer presets and exports the finished video.
Create captions →No credit card required
- ✓ Free plan available
- ✓ One-click instant export
- ✓ Built for creators
Runs in the browser — no install, no credit card, and a free plan with 2 videos a month.