karkirowle

It's written in the README.md, but it should be tensorflow-gpu 1.8. I can confirm that it also works reasonably well, with tensorlfo-gpu 1.9.

I can confirm on my system that by creating an account and generating the API key this works (Ubuntu 18.04). I have relatively limited experience with Docker, so I cannot...

Thanks for this pull request @taras-sereda! It basically makes everything work that's wrong with the current version. I can confirm it's suitable for merge.

Thanks @rafaelvalle for your advice. Averaging helps a lot, and the latent sampling improves the naturalness a bit. I added some more samples for those who are interested [here](https://drive.google.com/open?id=1YUDE-eCN7_Hlhqx4OSFcVVXBeS9eBhYC), using...

@rafaelvalle thanks for your help. I tried and uploaded some time averaging samples now [here.](https://drive.google.com/open?id=1TwDn-df6LAosXNupFGhnC3Lw66zwLgFG) I find that these are less distinctive than the batch averaged ones. But I might...

> I found the cause of my problem - looks like I am using a newer version of PyTorch in which masked_fill_ does not work with uint8 tensors anymore. The...

If you are using the LJS model that might be expected as it is a single speaker model. You could try using the LibrITTS.

@ninaburns there is a bit of discussion why that's needed [in this issue](https://github.com/NVIDIA/flowtron/issues/9#issuecomment-629628184) Basically, if you want to use the latest PyTorch version you have to do this. This is...

Hi @DamienToomey ! If I understand right you are using the LibriTTS model and the (one) surprised wav file as style that is on the NVIDIA demo page that you...

@rafaelvalle No, I haven't. Which e-mail address? Did you use the e-mail address in the COLAB or on my blog? I can have a second look or try resending it...