Match the timing before generation
A large duration difference makes loops, reverse motion or speed changes more noticeable. Use source footage close to your voice-track length when you want natural pacing.
AI AVATAR / SHOTTIDE
AI lip sync for the video you already have.
Upload a video with a clear face and prepare a new voice track. Choose what happens when the durations differ, then review the cost before synchronizing.
Choose the right starting point
Video lip sync changes mouth motion in existing footage to follow a new audio track. This workflow uses Sync Lipsync 2 with your source video and normalized audio. It replaces the soundtrack; it does not translate the script or preserve background music automatically.
How to lip-sync an existing video
- 01
Choose an existing video
Upload MP4, MOV or WebM, or select a saved video. Preview it and choose footage with a clearly visible speaking face.
- 02
Prepare the replacement audio
Import a recording or generate speech from a script. Listen for pronunciation and pacing, then mark the audio reviewed.
- 03
Set duration handling and generate
Cut to the shorter input, loop or bounce the video, remap its speed, or keep the longer duration with silence padding. Review the measured output duration and credit ceiling.
What to check before you publish
Review the full frame
Inspect the mouth, face outline and background across the finished clip. Active-speaker detection is optional; it does not guarantee correct selection in every multi-person shot.
Inputs, limits and credit costs
- Video upload
- Up to 1,000 MB, 3 minutes and 4K. Large files are compressed to MP4; review fine details after import.
- Audio and script
- Imported audio: up to 30 MB and 30 minutes. Generated speech: up to 6,000 characters.
- Cost
- 6 credits per started output second. Speech generation is charged separately.
- Sound and captions
- New audio replaces the original soundtrack. Transcribe the result, review and style captions, then export a copy. Transcription is charged separately.
Frequently asked questions
What if my video and audio have different lengths?
Choose one of five duration modes. Cut-off uses the shorter duration; silence uses the longer duration; loop, bounce and remap follow the audio duration. The studio shows the resulting quote before generation.
Can I animate a photo instead?
Use AI Avatar Generator for a still portrait. AI Lip Sync takes an existing video as its visual source.
Does it translate my video automatically?
No. Supply the replacement audio or write a script in your target language. Check meaning and pronunciation before generating the lip-sync video.
Are the maximum input lengths a quality guarantee?
No. Input limits describe what the workflow accepts. Face visibility, source quality, timing and model behavior affect the result, and long outputs require careful review.