Greetings. Thanks for making your adjustment open-sourced. Many changes are useful.
I am assuming that you would not provide the trained model because of the training data license.
But is it possible that you help to use your trained model to test the attached data and show the effect?
The video attached is from AVSpeech dataset. And the audio is from wild.
Hi @primepake
Greetings. Thanks for making your adjustment open-sourced. Many changes are useful.
I am assuming that you would not provide the trained model because of the training data license. But is it possible that you help to use your trained model to test the attached data and show the effect?
The video attached is from AVSpeech dataset. And the audio is from wild.
Archive 2.zip
Looking forward to your reply! :)