Lip-Sync is optional and costs extra. It's powerful when the video actually benefits from it, and a waste of credits when it doesn't. This article is a quick decision framework.
Enable Lip-Sync when:
The speaker is on camera most of the time and facing the audience.
The face is prominent in the frame — talking-head videos, interviews, product demos with a presenter, ads.
The mouth is clearly visible throughout.
Sufficient high-quality footage is available: The video must contain at least 10 seconds of clear, unobstructed footage of the speaker's face for the AI to process the lip-sync accurately.
You're publishing to audiences who expect a localized look (social, YouTube, ads, branded content). For these, Lip-Sync adds the visual polish that separates a dubbed video from a voice-over.
Skip Lip-Sync when:
The speaker is rarely on camera — voice-over videos, tutorials with screen recording, product videos with minimal presenter footage.
The video is mostly B-roll, animation, or motion graphics with no human face.
Shots are very wide or the face is too small to benefit.
The face is frequently occluded (close microphones, masks, heavy beards).
You're shipping internal content where the extra polish isn't worth the credit cost. For these, ship just the dubbed audio with the original video — it looks more natural than a forced Lip-Sync and saves you credits.
How to turn it on
Open your finished dub.
Open the Lip Sync tab on the dub detail page.
Click Start lip sync. You'll see the credit cost before you commit.
Wait for processing to finish — Lip-Sync runs independently of the main dub and usually takes a few minutes per segment. Lip-Sync is decided per subdub (per target language), so you can turn it on for the languages where it matters most and skip it elsewhere.
Lip-Sync always syncs to the dubbed audio of that subdub — in the standard workflow you can't point it at an audio file of your own, and it can't run without a dub. If your project needs your own audio track, ask Dubby in the chat — our team looks at such requests individually. See Lip-sync limitations and workarounds.
Run Lip-Sync as the last step
Lip-Sync should always be the final step in your workflow. It re-processes the whole video against the current audio, so it only pays off once the translation is locked.
If you edit the translation (or any segment) after Lip-Sync and then regenerate the video, the lip-synced version is discarded — it no longer matches the new audio, so Lip-Sync resets to not started and you'll have to run it again, which is charged again. (While Lip-Sync is running, translation editing is locked.)
The safe order: translate → review and adjust in the editor → update the video → and only when everything is 100% correct, run Lip-Sync.
Cost
1 credit per minute per subdub, on top of the dubbing cost. A 10-minute video dubbed into 3 languages with Lip-Sync for all three: 30 credits dubbing + 30 credits Lip-Sync = 60 credits total. Free trial: Your first Lip-Sync per subdub is free — enough to test the feature on a short clip before you commit on longer content.
Changed your mind after running it?
You can always download the dubbed-audio version without Lip-Sync — both are available on the dub detail page once processing finishes. Lip-Sync credits aren't refunded automatically if you decide not to use the result, so use the free-trial to preview before running it on longer content.
