LipDub 2.0 speeds up short-form localization
LipDub 2.0 is a new lip sync model with no training step, making it 8x faster than LipDub 1.0. It handles single-speaker videos up to 2 minutes at 1080p in 80+ languages.
What it does
LipDub 2.0 is a new lip sync model for short-form video. It skips the preprocessing and per-speaker training that LipDub 1.0 needs, so jobs start rendering as soon as you upload.
It handles side profiles, turns, and hands or objects in front of the face, for videos up to 2 minutes with a single speaker, at up to 1080p, in 80+ languages.
Results and limits
LipDub says it's 8x faster than LipDub 1.0. Limits: one speaker and 2 minutes max; for longer or multi-speaker videos, use LipDub 1.0.
Worth knowing
There are no training fees, just the same per-minute rate. It's in the API and in the app's Translation and Dialogue Replacement projects, preselected for eligible videos.
Try it yourself
Create a Translation project, upload [a 30-second single-speaker clip], pick [target language], and compare turnaround with your current pipeline.