Open ZhaZhaFon opened 2 years ago
I have the same question.
I found the answer in the previous issue #3 #8 #32. However, #3 and #8 said that the pre-trained models are trained on Vox1&2&Libri-other while #32 said Vox2. Hope that this can be clarified. Thanks a lot.
Hi, thanks for sharing
I wonder how the pre-trained model was trained. What data did you use to train the encoder ? LibriSpeech ? VoxCeleb ?