verssae / flask-tacotron2-tts-web-app

flask+tornado based NVIDIA tacotron2+waveglow tts web app

⚡️ My GitHub Activity Graph (last 31 days)

📌 Pinned Gist

flask-tacotron2-tts-web-app's Issues

Decoder Max

I was able to stop the "Warning! Decoder Max" and weird sounds if I always use a punctuation like period.

model?

which tacotron2 pretrained models can we use? can you show an example of using it

error when running app.py

When running app.py, an error occurs as follows: AttributeError: 'WN' object has no attribute 'cond_layer'.

How to use this???

How to use this???

export PYTHONPATH="$PYTHONPATH:/path/to/flask-tacotron2-tts-web-app"

What should I do?

Quick answer please

Training

Do you have instructions on how to train, which git commit of tacotron2 to use
I ask because your repo is one of the only few that the models actually work, all the other repos have issues. I want to train a new model and be able to use it with your code

mellotron intergration?

Could you integrate this
https://github.com/NVIDIA/mellotron

improved inference speed

Have you seen this?
https://github.com/Saurabh-29/Waveglow_Inference_in_CUDA

ERROR when attempting to run the app.py

Hi,

i checkout the code and installed all prerequisites,and placed trained models in models folder.

Facing below issue when trying to run the application.
D:\anaconda3\envs\final\lib\site-packages\pydub\utils.py:170: RuntimeWarning: Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work
warn("Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work", RuntimeWarning)
D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py:671: SourceChangeWarning: source code of class 'glow.WaveGlow' 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)
D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py:671: SourceChangeWarning: source code of class 'torch.nn.modules.conv.ConvTranspose1d' 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)
D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py:671: 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)
D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py:671: SourceChangeWarning: source code of class 'torch.nn.modules.conv.Conv1d' 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 "D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py", line 308, in _check_seekable
f.seek(f.tell())
AttributeError: 'NoneType' object has no attribute 'seek'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "app.py", line 12, in
t2s = T2S(model)
File "D:\flask-tacotron2-tts-web-app\text2speech.py", line 40, in init
self.update_model(model_choice, self.max_duration_s)
File "D:\flask-tacotron2-tts-web-app\text2speech.py", line 90, in update_model
self.model.load_state_dict(torch.load(self.checkpoint_path, map_location=torch.device('cpu'))['state_dict'])
File "D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py", line 594, in load
with _open_file_like(f, 'rb') as opened_file:
File "D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py", line 235, in _open_file_like
return _open_buffer_reader(name_or_buffer)
File "D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py", line 220, in init
_check_seekable(buffer)
File "D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py", line 311, in _check_seekable
raise_err_msg(["seek", "tell"], e)
File "D:\anaconda3\envs\final\lib\site-packages\torch\serialization.py", line 304, in raise_err_msg
raise type(e)(msg)
AttributeError: 'NoneType' object has no attribute 'seek'. You can only torch.load from a file that is seekable. Please pre-load the data into a buffer like io.BytesIO and try to load from it instead.
(final) PS D:\flask-tacotron2-tts-web-app>

Environment details:

python 3.6

CUDA 10.1

pytorch 1.9.1

tensorflow 1.13.1

please throw some light to proceed further