No. LipDub AI is 100% proprietary.
No, as long as you upload a video with at least 30-45 seconds of speaking time for the model to train on, LipDub AI will generate based on the length of your audio track.
The 30-45 seconds of training footage can be stitched together or looped, the model just uses it to learn how your speakers articulate.
Yes, you can generate multiple videos simultaneously, but you'll need to start each one individually.
For example, if you want to translate one video into five different languages, start by kicking off the first language. Then, while the first video is processing, you can begin generating the second language and so on.
Yes. Please reach out to Sales for more details.
The platform currently supports professional resolution MOV or MP4 files, up to 4K resolution, with both ungraded and graded footage.
Supported colorspaces include sRGB and Rec709. For best results, avoid manipulated footage, such as text overlays on faces or fade-in transitions.