Is it possible to retrain the audio2face network using training data that I have prepared myself?

The question is exactly as in the title.
Where can I find information on how to make that happen if possible?

Hello! I’ve shared your post with the dev team for further assistance.

You may be interested in our Replicator Extension. Here is more information on that Replicator — Omniverse Extensions documentation

1 Like

Thank you for the information.

Does this mean that even if the desired result is not obtained during inference, it is possible to get even closer to the desired result by relearning using this function?

Also, in the linked example, images are used as input target, but is it possible to use the same feature in Audio2Face?