hdmjdp
hdmjdp
> what is the "warm up iters ", I cannot find it in the config. By the way I have the same problem, the loss cannot go down as below:...
can you share your code > I suggest 1 @rishikksh20
> detached hidden state @rishikksh20 Does this refer to text encoder output?
@rishikksh20 After 100k,, does the prams of prodsody extractor update or just frozen?
> @MorganCZY Training the model on a multiple speaker corpus and it will generalize automatically. You can just listen the audio samples released by the authors, the results in the...
> > > @MorganCZY Training the model on a multiple speaker corpus and it will generalize automatically. You can just listen the audio samples released by the authors, the results...
> Dear author, thank you for your contribution for TTS, this is a big step in E2E TTS. But when I use ground truth duration aiming to train faster and...
@zepingyu0512 looking for a pytorch version too!
> yes better than hifi-gan with less training but in my experiments, the stftnet has larger machine noise in the audio, did you have it? also in the training, there...
I think this framework is similar to tacotron1, because you don't use wavenet-decoder instead of grifflim-decoder. So the tts-wav from your implements is whether better than tacotron1?