MP4 in · MP4 out · any LMS

The product changed.
The training video didn't.

Upload the old recording, edit the narration where it's wrong, and download the same video with updated audio. No studio booking, no chasing down whoever recorded it.

10 free minutes · No credit card required

How it works

  1. 1

    Upload the existing video

    The module you already published, in MP4, MOV, MKV, or AVI.

  2. 2

    AI transcribes the narration

    Every sentence comes back time-stamped and editable.

  3. 3

    Update what went stale

    Change the three lines about the old menu. Leave the rest exactly as it was.

  4. 4

    Publish the new version

    Download an MP4 and drop it back into your LMS.

What sends a module back for rework

The interface moved

A release renamed a menu and the narration now describes something that isn't there. The screen capture is still fine; the words aren't.

The policy changed

A compliance figure, a threshold, or a process step is now wrong. Fix the sentence that states it and leave the module intact.

The narrator has gone

The subject-matter expert who recorded it left the company. You can still change what the module says.

It needs to ship in more languages

Translate the narration into any of 57 languages and generate a version timed to the original cuts.

Not a re-record, and not a voice clone

Re-recording means finding the original narrator, matching their mic and their room, and hoping the new pacing still fits the screen capture. Voice cloning solves the wrong half of the problem — it changes how the module sounds while keeping the words that went out of date. Recastr changes the words and holds every sentence to the timing it already had, so a module you spent days producing can be corrected in an afternoon.

Change three lines, not the whole module

Only edited segments are regenerated. A wording fix in one section leaves the other twenty minutes of narration exactly as it was, which keeps your review cycle short.

One module, 57 languages

Translate the transcript and generate narration for each target language, all timed to the same cuts. Regional teams get the same module rather than a differently-paced remake.

Watch it work

Frequently asked questions

Will the result work in our LMS?
Recastr returns a standard MP4, which every major LMS accepts as a video asset. Recastr does not build SCORM or xAPI packages — if your course is wrapped in one, swap the MP4 inside it the same way you would any other updated asset.
Can I update just one section?
Yes. Edit only the sentences that changed and Recastr regenerates only those. The rest of the narration is not re-synthesized and does not shift.
What if the updated line is longer than the original?
Each sentence is generated into the timeslot it originally occupied. If your rewrite runs long, Recastr speeds that sentence up slightly to fit rather than pushing everything after it out of sync. Rewrites that stay close to the original length sound the most natural.
What if the original narrator has left?
It makes no difference. Recastr works from the recording, not from the person. Pick any available voice for the updated module, or keep one consistent voice across your whole library.
Can we localize the updated version?
Yes — 57 languages. Translate the transcript after editing it and generate a voiceover per language, each one timed to the original cuts.
How long can a module be?
Up to 30 minutes per module. Longer courses are usually already split into modules, so run each one separately.
Does it handle more than one presenter?
When more than one speaker is detected in the recording, the transcript is split by speaker and you can assign a distinct voice to each.
What if we still have the original subtitle files?
Transcription is optional if your course materials already include the captions — hand the SRT or VTT straight to the SRT-to-audio workflow and it will generate narration timed to your existing cues.

Bring your library back up to date

10 free minutes included · No credit card required