VeedLipsyncNode - ComfyUI Built-in Node Documentation - ComfyUI

Re-sync a speaker’s mouth in a video to a new speech track with VEED Lipsync. Use it to dub a clip or replace its dialogue while keeping the original footage. The output keeps the video’s resolution at 25 fps and is as long as the audio, with one speaker per video.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	Model to use. Selecting a model reveals the parameters specific to it: `video`, `audio` and `seed`.	DYNAMIC_COMBO	Yes	`"lipsync-2.0"`
`video`	Footage of one front-facing speaker. If it is shorter than the audio, it plays forward and backward until the audio ends.	VIDEO	Yes	-
`audio`	New speech, up to 600 seconds. Sets the video length.	AUDIO	Yes	Up to 600 seconds
`seed`	Seed controls whether the node should re-run; results are non-deterministic regardless of seed. (default: 42)	INT	Yes	0 to 2147483647

The node stops with an error when the speech track is longer than 600 seconds, or when the VEED job fails.

Outputs

Output Name	Description	Data Type
`VIDEO`	The re-synced video: the original footage with the mouth matched to the new speech.	VIDEO

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): 1a1e854b6923b5bad0fd6f3a79e76c70231dee31a12e9cf0b22be6bdcc741168