How LipDub AI works
EasyDub
EasyDub is designed to streamline the single-speaker workflow, combining the uploading and labeling steps from our Advanced Workflow.
To get started, simply create a new project, upload the single-speaker video you want to lip sync, and that’s it!
LipDub AI will then automatically process your video. For the best results, make sure the clip is at least 30 seconds long to give the AI enough footage to train on.
Unlike the Advanced Workflow, EasyDub doesn’t allow you to upload additional training footage. If your clip is shorter than 30 seconds and you need to add extra footage, we recommend using the Advanced Workflow instead.
Upload
Add audio and train
Once your video has finished processing, click the “Create” button to add your audio. You can either upload your own audio file or choose “Translate with AI,” which will generate the audio clip directly in the app.
Our auto-translation feature translates your speaker’s voice from one language to another, cloned to sound as authentic as the original. Just select from one of our 29 language options, and you’re all set.
Now, LipDub AI will begin its comprehensive training process, which can take roughly 1-4 hours depending on the length and complexity of the footage. Our team is working to make this faster, but this is where LipDub AI builds the detailed model that delivers the realistic results we’re known for.
The best part? LipDub AI only needs to train on your original video once for the life of the project.
Generate and iterate
Your newly generated video is now ready to view, and since you’ve completed the training step, you can create as many iterations as you’d like.
Just upload new audio files or select different languages for auto-translation. It doesn’t matter if you started with auto-translation or your own audio—both options are available for every new version.
You can also queue up multiple iterations for efficient, simultaneous processing. LipDub AI uses the initial training to make subsequent versions quick and easy.
