Joshua Larouche

Could you submit a PR?

https://drive.google.com/drive/folders/1YuOoV3lO2-Hhn1F2HJ2aQ4S0LC1JdKLd

The schedulers are created like this: ``` scheduler_g = torch.optim.lr_scheduler.ExponentialLR(optim_g, gamma=h.lr_decay, last_epoch=last_epoch) scheduler_d = torch.optim.lr_scheduler.ExponentialLR(optim_d, gamma=h.lr_decay, last_epoch=last_epoch) ``` and the epoch info is loaded from the state dict: ` last_epoch...

Yes the link I provided contains both the do_ file (discriminator weights and optimizer state) and g_ file, (generator weights)

"Am I using mel outputs from tacotron paired with matching original source audio?" Yes, except the mels from tacotron need to be Ground Truth Aligned meaning the durations are identical...

It can learn any language or for that matter, any sound; as long as the training data is good. It's nothing more than a sum of sine waves. There is...

Thanks for the info, I will check that out! I have around 7,000 samples of my voice and I trained at 48Khz. 11 hours 45 mins. It is my own...

Hi I retrained using the latest repo and it was a bit better but still wound up getting the end pitches wrong by the end of training. It starts out...

Hi One thing you mentioned was adding silence to the mel spectrogram. I thought I could add silence by playing with duration of spaces, but it turns out, most words...

@joseluismoreira Hi My dataset was collected by me speaking into a Rode NT1 microphone. I used a tool that I wrote to make it easier to record the samples. You...