How to reach the same result as the example

The example is very good, but I have not achieved this effect with my own voice (Mandarin). The lip movement is not that accurate. how can I set the model parameters or adjust the voice(like volume)?

Hello @user119975! Thanks for checking out Audio2Face. We do support several multiple Latin based languages such as Russian, French, Spanish, and Italian. Non-Latin based languages, such as Asian languages, have more variation which the team is currently working on solving.

@user119975 We will have improved Mandarin support not far from now. It is in the plan. Stay tuned for that