Sentioscope

Speech and voice controls, as each vendor documents them

Changing one line of dialogue after the clip exists

Four entries document a way to change what is said in video that already exists. Three move the mouths to new audio and keep the picture. One also offers to act a short clip again with the new line, which fits the delivery to the words at the cost of a slightly shifted picture. As of 2026-10-13.

Two ways to change a line in a finished clipA lip-sync pass keeps every frame and redraws the mouth to new audio. A re-performance generates the clip again so the delivery fits the new words, and the picture may shift slightly. Either way the new line has to fit the length of the shot.A line in a finished clip has to changeLip syncSame picture, newmouthEvery frame kept; theoriginal ambience isdropped on one entry.Re-performSame shot, actedagainDelivery fits the words;documented for clips up to15 seconds.Voice onlySame words, new voiceA different library voicefor that clip alone.The new words still have to fit the seconds on screen
Fig. 1 Lip sync keeps the approved picture; re-performing keeps nothing but the storyboard.
What each entry documents about changing speech in a finished video, and what stays fixed. Recorded 2026-10-13.
ModelWhat can changeWhat stays as it was
PixVerseThe speech, from built-in or custom voices, with the mouth re-syncedThe supplied video, up to sixty seconds
SceneMixerThe words or the voice, by lip sync or by re-performing clips up to 15 secondsThe picture frame for frame on lip sync; the ambience is dropped
sync-3The mouth, matched to new audio or to textThe supplied video
SynthesiaThe language, in a dubbing project with locales addedThe uploaded video asset

Inclusion rule. Entries whose documentation describes changing the speech of a video that already exists, whether generated by that entry or supplied to it. Order. Alphabetical by model name.

1Moving the mouth or performing the line again

A lip-sync pass keeps every frame and redraws the mouth to match new audio. The picture that was approved stays approved; what changes is the speech and the shape of the lips. A re-performance generates the clip again with the new line, so the timing, breath and gesture fit the words, and the picture may move a little in the process.

Only one entry documents both, priced differently: lip sync at a credit a second on any length, re-performance at twice the tier's generation rate and only up to fifteen seconds.

2The new line has to fit the time already on screen

A lip-sync pass cannot lengthen the shot. A replacement line that runs longer than the original has to be delivered faster or cut, which is why the redub documentation advises keeping the new words short enough to fit. At roughly three spoken words a second the arithmetic is quick, and doing it before pressing the button saves a pass.

3The voice can change without the words

Swapping who a line sounds like is a separate operation from changing what it says. One entry lists every line of a clip with its speaker and a voice that can be exchanged for another from its library, applied to that clip only, while the character's own voice setting stays as it was.

That separation is useful in review, where a performance can be right and the casting wrong, and the fix should not touch the performance.

4Where the changed words are kept

Tools that work on an uploaded cut have no script to update, so the production's own copy drifts from what is on screen unless someone edits it by hand. One entry writes the changed lines back into the storyboard, so the text and the clip agree after the redub.

  • Voice source
    Text to speech supports both built-in voices and custom voices created from user-provided sample audiocatalogue or supplied samplePixVerse, speech and lip sync guide / recorded 2026-09-22
  • Per-character binding
    The chosen speaker id is passed as the lip-sync speaker on the generation requesta speaker id per callPixVerse, speech and lip sync guide / recorded 2026-09-22
  • Length per call
    Audio and video are each capped at sixty seconds and one hundred megabytesPixVerse, speech and lip sync guide / recorded 2026-09-22
  • Changing a line
    Lip 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 keptSceneMixer, redub guide / recorded 2026-10-13
  • Changing a line
    Re-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 redoneSceneMixer, redub guide / recorded 2026-10-13
  • Where the new words are recorded
    Lines changed in a redub are written back into the storyboardSceneMixer, changing a line guide / recorded 2026-10-13
  • Changing the voice in a finished clip
    In a redub each line carries a voice that can be swapped for another from the library without changing the words, for that clip onlySceneMixer, voice change guide / recorded 2026-10-13
  • Audio source
    The accepted pairs are video with audio, video with text, image with audio and image with textsupplied or read from textSync, sync-3 model documentation / recorded 2026-09-22
  • A route into another language
    The model page describes sync-3 as designed for global dubbing at native qualitysync, sync-3 model page / recorded 2026-10-13
  • A route into another language
    The API reference lists creating a dubbing project from an uploaded video asset and adding locales to itSynthesia, API reference / recorded 2026-10-13

5Sources

Every line is read from the vendor documentation linked on the speech controls page, checked 2026-10-13. What counts as documented is on how read. Nearby: Second language, Keeping a voice, Lip-sync.