Turn a rambling voice-over into a clean narration
You recorded it in one take, with the ums and the restarts. AI turns that into a clean script, shows it to you, and re-voices the video with a studio voice timed to the original.
Three steps, a couple of minutes
Drop in the recording
AI listens and writes down what was said.
Review the clean script
Fillers, false starts and repeats are gone; names, numbers and button labels are kept. Edit any line before a voice is recorded.
Pick a voice
Each line is spoken where you said it and fitted to its slot. The original sound is muted or kept quietly underneath.
Why use CutLuma for this
You approve the script first
Nothing is voiced until you have read the rewrite, so the narration says what you meant.
In sync with the screen
Lines start where the original sentence started and are only hurried slightly to fit.
One undo
The whole narration is one step: remove it and the original sound is back.
Questions? Answers.
Which languages work?
English and Spanish with natural voices; French, Chinese, Japanese and Korean with one basic voice each.
Is it free?
It uses transcription, the text model and AI voice, which are part of the paid plans and use the monthly AI time.
Can I keep my own voice?
Yes: keep the original quietly underneath, or skip the re-voicing and use “Edit video as text” to cut the fillers out of your own recording.
Turn a rambling voice-over into a clean narration
Included in every paid plan · The editor itself is free
Polish my narration






