Which shots can be dubbed: DUB (close-up), DUB-MEDIUM (a smaller face, dubbed through a crop) or NO, each with its reason (face too small, profile, mouth…

Lip-sync for music videos in ComfyUI: dub the singers' mouths to the song, shot by shot, with QA. Every input is a slot (bring a clip and song, or generate them with LTX-2.5, Qwen-Image-Edit and ACE-Step); vocal split, lyric alignment, lip kits, sync QA.

LSDub

project

source

shots

samples

song

cast

Which shots can be dubbed: DUB (close-up), DUB-MEDIUM (a smaller face, dubbed through a crop) or NO, each with its reason (face too small, profile, mouth hidden, no singing).

Name	Type	Default	Description
project	LSDUB_PROJECT		The LSDub project (song_offset, preset singers).
source	LSDUB_SOURCE		The video.
shots	LSDUB_SHOTS		The shots (from LSDub Shot Split).
samples	INT	61–50	Frames sampled per shot.
songopt	LSDUB_SONG		The song: its vocal marks which shots are sung.
castopt	LSDUB_CAST		The cast: judge only the singers' faces. Without it, the largest face in each shot is judged.

Name	Type	Description
shots	LSDUB_SHOTS	The shots with verdict, reason, face size, singer.
sheet	IMAGE	Contact sheet coloured by verdict.
report	STRING	Markdown: every shot, worst first, with its reason.

LSDub Triage

Inputs (6)

Outputs (3)