pretrained model and single speaker adaption

mkotha / WaveRNN

A WaveRNN implementation

MIT License

199 stars 48 forks source link

Open ghost opened 5 years ago

ghost commented 5 years ago

hi, can you share a pretrained model and how to transfer learn to adapt to a single speaker?

triwoods commented 5 years ago

@mkotha Thanks for your implementation. Any plan to share a pre-trained model with VQVAE-WaveRNN?