search

NVIDIA / mellotron

Mellotron: a multispeaker voice synthesis model based on Tacotron 2 GST that can make a voice emote and sing without emotive or singing training data
BSD 3-Clause "New" or "Revised" License
854 stars 187 forks source link

LJS trained model? #32

Closed scottbouma closed 4 years ago

scottbouma commented 4 years ago

Thank you so much for sharing this! I have it working on Google Colab (thanks to @yhgon ). But I cannot find the pretrained LJS model which you mentioned on your website (https://nv-adlr.github.io/Mellotron). Can you post a download link to your LJS model? Thanks!

rafaelvalle commented 4 years ago

We do not have a model trained only on LJS. You can use the model trained on LibriTTS

rafaelvalle commented 4 years ago

We uploaded a model trained on LJS. Please take a look at our README!