Edresson Casanova

Hi, I don't understand, because you use ZeroPadding1D in this model. you are adding 2048 zeros in the second shape dimension. example: when input shape is: (1,280,161) after pass in...

Hello, I found the work done by Baidu, focused on Voice Cloning very interesting, do you think it would be interesting to implement and analyze its performance? Look: [Neural Voice...

It fixes #2280

Currently, we can't load the XTTS model without providing a `speaker_file_path` or provide a `checkpoint_dir`. This PR fixes it.

Run test_run() every time that saves a checkpoint and not in the epochs end.

4

Currently, we save checkpoints without audio samples for each one. In this way, we cannot evaluate the models correctly. My suggestion is to run test_run() every time that we save...

# What does this PR do ? It implements Nemotron-VoiceChat Speech Decoder.