lokkelvin2 / tacotron2-tts-gui Goto Github PK

Hi I think the work you've done on this is amazing! :)
I am extremely interested in recent TTS developments but I have no coding experience whatsoever.

I've found it to be extremely difficult to understand anything I've read involving TTS and python.

Do you have any plans to implement a Training GUI ?
All I really want to do is train my own model, but the path to my goal is obscured by the sheer density of my ignorance.

Much love, I hope you are well in these difficult times

Hi,

I tried running the tool by installing all requirements but I can't seem to get it to work.. on multiple devices. I am getting the following error:

ModuleNotFoundError: No module named 'glow'

What am I missing? I installed everything, even torch.

I know it says "Allows CPU-only synthesis via a toggleable switch. Portable exe file is available (that runs on CPU only)."
But in the screenshot provided, it shows the "GPU Mode" toggled on. Where is there an installation file, versus the portable EXE File provided on the github page? I want to be able to utilize the GPU Mode for faster text conversion.

model.pt is a zip archive (did you mean to use torch.jit.load()?)

is there a way to fix this

Hi, I have done that but whenever I try to generate a TTS the output sounds really bad, e.g. it's like a blib, see video:
https://www.youtube.com/watch?v=vcGLX1OtjyM& at 0:03 you should hear it.

Originally posted by @maandagdev in #4 (comment)

I am also running into this.

model.pt is a zip archive (did you mean to use torch.jit.load()?)

how can i fix this?

python gui.py
pygame 2.0.1 (SDL 2.0.14, Python 3.8.0)
Hello from the pygame community. https://www.pygame.org/contribute.html
Traceback (most recent call last):
File "gui.py", line 31, in
from secrets import TOKEN # for debugging
ImportError: cannot import name 'TOKEN' from 'secrets' (C:\Program Files\Python38\lib\secrets.py)

Sorry to be a bother, I've been trying to get it to work but no luck, I've been able to solve (most) errors myself, but don't know what to do about this one. Thanks in any case.

I want to use the GUI form my pretrained model which is Tacotron 2 + WaveNet. Can you give me directions I'm not sure where to modify the code to implement it for my case?

Hello,

I've tried to download your application. I've been able to correctly plug the pretrained models, but apparently the donations from StreamElements are not pulled.

Moreover, I would like to understand how to properly train my own model(s): could you point me to some instructions?

Thanks.

Hello, sorry to bother.

I tried my best to resolve most errors than have arisen but can't seem to solve this issue at all.
Many thanks,

python gui.py
pygame 2.0.2 (SDL 2.0.16, Python 3.7.0)
Hello from the pygame community. https://www.pygame.org/contribute.html

C:\Users\Tester\AppData\Local\Programs\Python\Python37\lib\site-packages\torch\serialization.py:453: SourceChangeWarning: source code of class 'torch.nn.modules.container.ModuleList' has changed. you can retrieve the original source code by accessing the object's source attribute or set torch.nn.Module.dump_patches = True and use the patch tool to revert the changes.
warnings.warn(msg, SourceChangeWarning)

Traceback (most recent call last):
File "gui.py", line 768, in run
get_interruptflag = self.get_interruptflag)
TypeError: infer() got an unexpected keyword argument 'progress_callback'

Using:
Torch 1.2.0+cu92
Tensorflow 2.6.0
matplotlib 3.4.3
numpy 1.19.5
scipy 1.7.1

not sure if any of that info is useful