SceneMixer: performed in the language, redubbed when a line changes
SceneMixer states that each line is performed in the chosen language while the shot renders, so no translated track is laid over finished footage. When a line changes after the clip exists, a redub keeps the picture frame for frame, voices the lines again and moves the mouths to the new audio. From 3 source pages, read 2026-09-12 to 2026-10-13; no maker pays to be listed.
| What the documentation settles | What it leaves to a take |
|---|---|
| The model performs the line in the chosen language while it renders the shot | How close the mouths land on the first take |
| Lip sync keeps the picture frame for frame, voices every line again and moves the mouths to match | How the result compares with acting the clip again |
| Re-perform acts the clip again with the new lines, at twice the tier's rate, on clips of 15 seconds or less | Which of the two suits a given line, which is decided clip by clip |
Inclusion rule. Only statements found on the vendor page named in the caption, and only ones bearing on this one field. Nothing here is filled in from generated output. Order. In the order the statements arrive on the vendor's page.
1Two moments, each with its own answer
The first answer is about method. Perform the line in the delivery language while the picture renders and the disagreement a dubbing stage exists to fix never arises. The second is about change. When a line is rewritten after the clip exists, the redub moves the mouths to the new audio instead of regenerating the shot.
In the second case the driver is named, the new audio, which is the kind of answer this column records as the strongest grade.
2What a redub keeps and what it drops
Lip sync keeps every frame and voices all of the clip's lines again, at one credit per second of clip, on clips of any length. The clip's own ambience is not kept; the episode's score is mixed back under the new version. Re-perform acts the clip again so the delivery fits the words, on clips of fifteen seconds or less, at twice the tier's per-second price.
Either way the changed lines are written back into the storyboard, so the script stays the record of what is said on screen.
3The other entries that name a driver
Four other entries answer this column by naming what a mouth follows, often a supplied recording. This one names two moments: the performance at render, and the new audio when a line is redubbed.
- MiniMax — whoever is on screen.
- sync-3 — the audio it is given.
- Synthesia — the spoken content, framed close.
- Wan2.2-S2V — the audio input.
- Audio sourceThe video model is described as performing the line in the chosen language while it renders the shotgenerated with the picture
- Changing a lineLip sync keeps the picture frame for frame, voices every line again and moves the mouths to match, dropping the original ambience, at 1 credit per second on clips of any lengthpicture kept
- Changing a lineRe-perform acts the clip again with the new lines so delivery fits the words, at twice the tier's per-second price, on clips of 15 seconds or lessperformance redone
- Where the new words are recordedLines changed in a redub are written back into the storyboard
- LanguagesDialogue is set per project to any of 15 languages, with Cantonese a 16th for dialogue onlya named list of 15
4Sources
Read from the languages guide at scenemixer.com on 2026-09-12. The same column across every entry is on lip-sync; everything this vendor publishes about speech is on SceneMixer. What counts as documented is on how read.