Lipdub Playground is Live

We left silent movies in the 1920s with the advent of synchronized sound. Today, we’re leaving silent AI videos in the past with the launch of Lipdub Playground.

Powered by Captions' Lipdub 2.0 model, Lipdub Playground allows you to add a voice to your AI-generated videos. Just type and watch your character speak the provided script aloud, complete with synced lip movement and body language.

Cutting-Edge Lip-Syncing Technology‍

Captions' Lipdub 2.0 model is a multi-stage transformer-based model that allows anyone to generate lip movement (that can be synced to audio) in dynamic environments — for this use case, complicated AI-generated scenes. It’s also zero-shot, meaning it requires only a few seconds of input video and can handle complex situations like walking and 4K footage.

How it works: ‍

Cast your character: Start by selecting one of our pre-generated AI characters or upload an AI video of your own. Pro tip: If you’re uploading your own AI video, try to use one where your character has slight lip movement. This provides more data for our model to work with, resulting in a better output.

‍
Share a script: Next, it’s time to give your character some lines. Start by pasting or typing a script for your character to read aloud. Alternatively, upload a voice recording of your own and your character will speak your audio aloud.

‍

‍Generate your video: Our proprietary Lipdub model will bring your AI character to life and speak the words you’ve either written or recorded. Lipdub Playground makes bringing your ideas to life simple, whether you're creating marketing videos, personal projects, or storytelling content.

Try Lipdub Playground for yourself here.
‍