Lipsync syncs a voiceover to a face in an existing video or photo. You pick a source file, attach a voiceover, and Workroom generates a new video with audio-driven lip movements.
Before you begin
You need two files:
- A source file — a video with a visible face (Lipsync Labs, Lipsync 2 Pro) or a portrait photo (InfiniteTalk). See Lipsync models to choose the right model for your input.
- A voiceover — an audio recording. You can create one in Voice Studio.
Select your source
Click the Video button in the control bar (or Image if you're using InfiniteTalk). A modal opens with three tabs: All, Favorites, and Uploaded. Click any thumbnail to select it, or use Upload new file to add a file from your computer.
Add a voiceover
Click the Voiceover button. The modal lists your saved voiceovers with a preview play button and duration display. Select one, or use Upload new file to add a new audio file.
Generate
When both inputs are selected, the Generate button activates. Click it to start processing.
View your result
The generated video appears directly on the Lipsync page once processing is complete. Use the playback controls to preview it. To generate again with different inputs, swap the video or voiceover and click Generate once more.