Comments (10)
You have to merge the pretrained models with the root of the repo. I think you put them in the wrong place. In the same folder you have demo_toolbox.py
you should have the directory encoder
, then saved_models
under it and finally pretrained.pt
under that. Right now, you don't have pretrained.pt
there and that's causing the error.
As for retraining in Romanian, I think that's going to be difficult unless you find a good Romanian dataset. Even then that's a lot of work to interface the dataset with the project.
from real-time-voice-cloning.
You would need hundreds of hours, and you need the text that comes with that data.
from real-time-voice-cloning.
The package is Unidecode
, not unicode. It's only been added with a recent PR. Pull the repo and install the requirements again.
from real-time-voice-cloning.
I followed your indications, and I've got an error (see below). I understand something is related with dataset. How do I properly configure dataset, because before I tried to download VoxCeleb1 and VoxCeleb2 I couldn't, it asked me some authentication and just I skipped this step.
PS: How do I train for other languages, not English, where I can get the data set, or how can I do it?
C:\Users\admin\Downloads\Real-Time-Voice-Cloning>python demo_toolbox.py "C:\User
s\admin\Downloads\Real-Time-Voice-Cloning\dataset_root"
WARNING: Logging before flag parsing goes to stderr.
W0619 00:22:47.159650 5648 deprecation_wrapper.py:118] From C:\Users\admin\Down
loads\Real-Time-Voice-Cloning\synthesizer\models\modules.py:91: The name tf.nn.r
nn_cell.RNNCell is deprecated. Please use tf.compat.v1.nn.rnn_cell.RNNCell instead.
Arguments:
datasets_root: C:\Users\admin\Downloads\Real-Time-Voice-Cloning\dataset_root
enc_models_dir: encoder\saved_models
syn_models_dir: synthesizer\saved_models
voc_models_dir: vocoder\saved_models
Traceback (most recent call last):
File "demo_toolbox.py", line 28, in
Toolbox(**vars(args))
File "C:\Users\admin\Downloads\Real-Time-Voice-Cloning\toolbox_init_.py", line 39, in init
self.reset_ui(enc_models_dir, syn_models_dir, voc_models_dir)
File "C:\Users\admin\Downloads\Real-Time-Voice-Cloning\toolbox_init_.py", line 85, in reset_ui
self.ui.populate_models(encoder_models_dir, synthesizer_models_dir, vocoder_
models_dir)
File "C:\Users\admin\Downloads\Real-Time-Voice-Cloning\toolbox\ui.py", line 26
9, in populate_models
raise Exception("No encoder models found in %s" % encoder_models_dir)
Exception: No encoder models found in encoder\saved_models
from real-time-voice-cloning.
You have not downloaded the pretrained models. Look for them in the readme. Don't worry, you can run the toolbox without datasets first. What is your native language?
from real-time-voice-cloning.
Romanian, but about pretrained models I've got downloaded before, what I did wrong?
from real-time-voice-cloning.
Could you suggest me some pre-trained reference of Romanian dataset source? or in other words, from what I should start?
from real-time-voice-cloning.
No, I don't know about any substantial one. M-AILABS doesn't have any Romanian, and the only one I found is SWARA which is way too small to do anything. Sorry.
from real-time-voice-cloning.
Last question if I have some records (romaine) can I use them to train the speech?
from real-time-voice-cloning.
File "demo_toolbox.py", line 2, in
from toolbox import Toolbox
File "C:\Users\dacia\Desktop\Real-Time-Voice-Cloning-master\toolbox_init_.py", line 1, in
from toolbox.ui import UI
File "C:\Users\dacia\Desktop\Real-Time-Voice-Cloning-master\toolbox\ui.py", line 2, in
from matplotlib.backends.backend_qt5agg import FigureCanvasQTAgg as FigureCanvas
File "C:\Users\dacia\AppData\Roaming\Python\Python37\site-packages\matplotlib\backends\backend_qt5agg.py", line 11, in
from .backend_qt5 import (
File "C:\Users\dacia\AppData\Roaming\Python\Python37\site-packages\matplotlib\backends\backend_qt5.py", line 13, in
import matplotlib.backends.qt_editor.figureoptions as figureoptions
File "C:\Users\dacia\AppData\Roaming\Python\Python37\site-packages\matplotlib\backends\qt_editor\figureoptions.py", line 11, in
from matplotlib.backends.qt_compat import QtGui
File "C:\Users\dacia\AppData\Roaming\Python\Python37\site-packages\matplotlib\backends\qt_compat.py", line 179, in
raise ImportError("Failed to import any qt binding")
ImportError: Failed to import any qt binding
PS C:\Users\dacia\Desktop\Real-Time-Voice-Cloning-master>
from real-time-voice-cloning.
Related Issues (20)
- Runtime error HOT 3
- How to deal with this bug? HOT 3
- AttributeError: module 'umap' has no attribute 'UMAP' on macOS Big Sur
- How can I decrease speed for cloning?
- How were the pretrained models trained?
- GUI is not started after running demo_toolbox.py
- Is this open source or paid? HOT 1
- RuntimeError Corrupted. HOT 1
- Ai voice cloning
- TypeError: melspectrogram() HOT 1
- What to do , i dont get it . Help! HOT 1
- i need help my brain hurts from this
- I have no idea what to do anymore with this, I need help.
- Failed to build webrtcvad when installing a package HOT 3
- Is this project abandoned? HOT 2
- Installing on M2 Chip error: metadata-generation-failed HOT 2
- the output only one second why HOT 1
- dataset
- [REQ] Switch to a GH-compliant license file HOT 2
- encoder training isn't stopping
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from real-time-voice-cloning.