Voice to caption and text to voice on one StarCut timeline
Captions and voiceovers are not side exports. In StarCut they land on the timeline next to the cut, so you can tighten dialogue and narration in the same project.

Two directions, one project
StarCut’s speech tools go both ways. Voice to Caption takes audio or video you already have and writes captions with precise timestamps. Text to Voice takes a script and generates narration you can place on the timeline. Both results stay in the project, not in a separate download folder you have to import again.
That matters when the cut is still moving. A caption you regenerate after trimming a pause should still sit on the same track. A voice line you rewrite should drop into the same edit without starting over.
Voice to Caption: from speech to timed text
Upload the audio or video that carries the dialogue. The agent transcribes with frame-accurate timing, then can flag pauses, breaths, and filler words so the captions stay tight with the picture. Kinetic or styled text effects are part of the studio — not a second app that only returns an SRT.
After captions land, change them by hand. Fix a name, split a line, or move a cue. The point of running speech-to-caption inside StarCut is that the text remains an editable layer on the timeline.

Text to Voice: from script to narration
Type or paste the script, choose a voice, and generate narration in seconds. The clip joins the project as audio you can trim, duck under music, or retime next to picture. Adjusting pace and placement happens in the same studio that holds the cut.
If a line is wrong, change the script and regenerate that piece rather than rebuilding the whole timeline. The surrounding edit should still be there.
Walkthrough: caption a take, then voice a CTA
A short talking-head clip plus one closing line is enough to exercise both directions without a long brief.
- 01Open a project that already has the talking-head footage on the timeline.
- 02Ask for Voice to Caption on that clip. Confirm captions appear with timing you can scrub against the picture.
- 03Edit one caption by hand — a proper noun or a split line — so you know the layer is yours.
- 04Write a short closing CTA script and run Text to Voice. Place the narration on an audio track under the end card.
- 05Trim the voice clip and nudge a caption so they meet cleanly. Export only when both picture and speech are right.
Hear it, see it, speak it — then keep scrubbing the same timeline.
Questions
Can StarCut caption existing video?
Yes. Voice to Caption transcribes audio or video into timed captions you can still edit on the timeline.
Does text-to-voice only give me a download?
No. Generated narration lands in the project as audio you can place, trim, and revise next to the cut.
Can I change captions after the agent runs?
Yes. Captions are editable layers. Fix wording, timing, and style in the same studio without leaving the project.
Open the studio
Cut in the browser, keep the timeline editable, and install the plugin when another agent host should drive the same project.