sayanoai / rvc-studio Goto Github PK
View Code? Open in Web Editor NEWThe best looking and most functional webui for RVC related tasks. See website for UI demo:
Home Page: https://rvc-studio.streamlit.app/
License: MIT License
The best looking and most functional webui for RVC related tasks. See website for UI demo:
Home Page: https://rvc-studio.streamlit.app/
License: MIT License
I need to help to fix this error that pops of when i want to train the data:
Traceback (most recent call last):
File "D:\AI\VoiceUI\RVC-Studio\preprocessing_utils.py", line 72, in pipeline
audio = load_audio(path, self.sr)
File "D:\AI\VoiceUI\RVC-Studio\lib\audio.py", line 19, in load_audio
raise RuntimeError(f"Failed to load audio: {e}")
RuntimeError: Failed to load audio: ffmpeg error (see stderr output for detail)
os.getlogin()
used in the Chat page is not supported inside docker containers.
It can be fixed either by giving a default name or using getpass.getuser()
.
file webui/components.py, line 168:
with urllib.request.urlopen(url) as response:
the problem is that no import for urllib was declared in the file. please provide a hotfix
how can I output a stereo audio file instead of mono?
Hello, I have a problem when start a rvc server.
I press start button and after server api loading nothing happens:
At inference page after pressing refresh data button nothing happens and i cant choose voice model:
Please, help me ๐
I get FileNotFoundError: [Errno 2] No such file or directory: '/usr/local/bin/python /app/api.py --port=5555 --host=0.0.0.0'
when trying to start the RVC Server. RVC Studio is running in a docker container.
Traceback:
File "/usr/local/lib/python3.8/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 552, in _run_script
exec(code, module.__dict__)
File "/app/pages/0_RVC_Server.py", line 65, in <module>
start_server(host=state.host,port=state.port)
File "/app/pages/0_RVC_Server.py", line 31, in start_server
p = subprocess.Popen(cmd, cwd=BASE_DIR)
File "/usr/local/lib/python3.8/subprocess.py", line 858, in __init__
self._execute_child(args, executable, preexec_fn, close_fds,
File "/usr/local/lib/python3.8/subprocess.py", line 1720, in _execute_child
raise child_exception_type(errno_num, err_msg, err_filename)
There are both api.py in the app dir, and python binary in /usr/local/bin
inside the container.
2023-09-15 19:07:49.805 Uncaught app exception
Traceback (most recent call last):
File "C:\Users\myhom\AppData\Local\Programs\Python\Python39\lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 552, in _run_script
exec(code, module.dict)
File "C:\RVCStudio\RVC-Studio\pages\3_Text_to_Speech.py", line 140, in
state.tts_audio = generate_speech(state.tts_text,speaker=os.path.basename(state.model_name).split(".")[0],method=state.tts_method, device=state.device)
File "C:\Users\myhom\AppData\Local\Programs\Python\Python39\lib\ntpath.py", line 216, in basename
return split(p)[1]
File "C:\Users\myhom\AppData\Local\Programs\Python\Python39\lib\ntpath.py", line 185, in split
p = os.fspath(p)
TypeError: expected str, bytes or os.PathLike object, not NoneType
i get this error when i try to train my model further and i have no idea how to fix this issue.
Please add the possibility to select a reader in EdgeTTS (list of all readers in all languages)
That would be a very good feature
Mine just seams like it gets stuck here and never changes no matter how long i wait. it says "INFO:jc_v2_40k:{'train': {'log_interval': 200, 'seed': 1234, 'epochs': 20000, 'learning_rate': 0.0001, 'betas': [0.8, 0.99], 'eps': 1e-09, 'batch_size': 5, 'fp16_run': True, 'lr_decay': 0.999875, 'segment_size': 12800, 'init_lr_ratio': 1, 'warmup_epochs': 0, 'c_mel': 45, 'c_kl': 1.0}, 'data': {'max_wav_value': 32768.0, 'sampling_rate': 40000, 'filter_length': 2048, 'hop_length': 400, 'win_length': 2048, 'n_mel_channels': 125, 'mel_fmin': 0.0, 'mel_fmax': None, 'training_files': './logs\jc_v2_40k/filelist.txt'}, 'model': {'inter_channels': 192, 'hidden_channels': 192, 'filter_channels': 768, 'n_heads': 2, 'n_layers': 6, 'kernel_size': 3, 'p_dropout': 0, 'resblock': '1', 'resblock_kernel_sizes': [3, 7, 11], 'resblock_dilation_sizes': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'upsample_rates': [10, 10, 2, 2], 'upsample_initial_channel': 512, 'upsample_kernel_sizes': [16, 16, 4, 4], 'use_spectral_norm': False, 'gin_channels': 256, 'spk_embed_dim': 109}, 'model_dir': './logs\jc_v2_40k', 'experiment_dir': './logs\jc_v2_40k', 'save_every_epoch': 5, 'name': 'jc', 'total_epoch': 1000, 'pretrainG': 'models\pretrained_v2\f0G40k.pth', 'pretrainD': 'models\pretrained_v2\f0D40k.pth', 'version': 'v2', 'gpus': '0', 'sample_rate': '40k', 'if_f0': 1, 'if_latest': 1, 'save_every_weights': '1', 'if_cache_data_in_gpu': 0}"
And on the home page I already downloaded everything. I did have it train a feature index and it did work.
Extract Vocals
TypeError: object of type 'NoneType' has no len()
Traceback:
File "C:\Users\xxx\AppData\Local\Programs\Python\Python310\lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 552, in _run_script
exec(code, module.dict)
File "C:\Users\xxx\Desktop\RVC-Studio-main\pages\1_Inference.py", line 193, in
with st.expander(i18n("inference.split_vocals.expander"),expanded=not (state.input_audio_name and len(state.uvr5_params.model_paths))):
Hi I Tried to put model in RAV folder and index in RAV /.index / directory
but its not showing any model. I even downloaded using the home interface to download the models but still it's not showing/loading any voice model .
will be grateful if someone guide me .
Thanks
So, I spent all night training, only to discover it needs more epocs, how to continue training?
Without a continue option, that's a lot of wasted power.
Description
The button for models scan(Refresh data
in Inference tab) doesn't do anything.
The scan doesn't work with or without the RVC server. The RVC server, however, can see the model right after the start:
{"status":"OK","rvc":["repetur"],"uvr":[],"denoise":[]} # output at http://localhost:5555/, where repetur is model name
The model and index are inside models/RVC
folder(index is in models/RVC/.index
), the model was trained with RVC-Studio.
No errors are being thrown in the console after pushing the button. The only errors i've encountered during my user journey from Home tab to training and up to infer are:
Invalid URL 'None': No scheme supplied. Perhaps you meant https://None?
Invalid URL 'None': No scheme supplied. Perhaps you meant https://None?
Invalid URL 'None/preprocess': No scheme supplied. Perhaps you meant https://None/preprocess?
Invalid URL 'None/postprocess': No scheme supplied. Perhaps you meant https://None/postprocess?
'<' not supported between instances of 'NoneType' and 'int'
'<' not supported between instances of 'NoneType' and 'int'
'<' not supported between instances of 'NoneType' and 'int'
but as far as i can tell they do not interfere with anything and the model training(which is the only thing i've done so far besides downloading hubert, ffmpeg, etc.) went fine.
building the docker image goes succesfully
when launching with 'docker container run rvc-studio' the web page comes up with,, 'refuse to connect to 0.0.0.0'
may be missing something. not familiar with docker.
Resolved
when running docker had to add '--net=host' to get internet connectivity. eg. 'docker run --net=host rvc-studo'
TIA
I spent over 30 hours trying to troubleshoot this. I followed every bits of details there is on this. I absolutely need help now.
After I initiate the server, made sure everything is green, went to the inferring page, reloaded (refresh) button, selected my pth and mp3, 1-click does NOT want to come on
2023-09-23 18:23:43.056 Uncaught app exception
Traceback (most recent call last):
File "E:\conda-envs\RVC-Studio\lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 552, in _run_script
exec(code, module.dict)
File "E:\RVC-Studio\pages\1_Inference.py", line 166, in
disabled=not (state.uvr5_params.model_paths and state.input_audio_name and state.model_name)):
AttributeError: 'NoneType' object has no attribute 'model_paths'
I already reinstalled and it still doesn't work.
All other tabs work fine.
Hello! It's Antasma from the AI Hub Discord server.
Really sorry to write you through here, but that's the only way we found to reach you.
Could you contact me on Discord (@antasma245
) ?
Many thanks in advance and really sorry for the inconvenience.
I was able to train one model successfully, but for reasons I can't understand, all subsequent attempts fail to train, even after a re-install from scratch. I'm attaching a screencap of the error message. This github thread Lightning-AI/pytorch-lightning#11865 suggests that the error could be solved by setting an environment variable --
SETexport
NCCL_P2P_DISABLE=1` but that variable doesn't seem to persist after running streamlit run Home.py.
Is there a solution that I can implement in the code?
Checking if conda is already installed...
.cache folder does not exist. Creating it now.
Downloading Miniconda installer...
BITSADMIN version 3.0
BITS administration utility.
(C) Copyright Microsoft Corp.
Transfer complete.
Installing Miniconda...
Checking if conda is in the user path...
Something went wrong. Please check the installation log or submit a ticket at: https://github.com/SayanoAI/RVC-Studio/issues
Why?
Installed RVC-Studio on two Debian 12 linux machines
1 x Python 3.8.9 , 1 x 3.10.12
NO GPU only CPU,,,,,for completeness
ffmpeg-python IS installed
EDIT:
Set up stderr output and this is what the stderr throws
RVC-Studio/input/inoutfiles.zip: Invalid data found when processing input
`VC-Studio/preprocessing_utils.py", line 73, in pipeline
audio = load_audio(path, self.sr)
File "/home/brcisna/MyPrograms/rvcstudio/RVC-Studio/lib/audio.py", line 43, in load_audio
raise RuntimeError(f"Failed to load audio: {e}")
RuntimeError: Failed to load audio: ffmpeg error (see stderr output for detail)
/home/brcisna/MyPrograms/rvcstudio/RVC-Studio/datasets/elvis.zip->Traceback (most recent call last):
File "/home/brcisna/MyPrograms/rvcstudio/RVC-Studio/lib/audio.py", line 38, in load_audio
ffmpeg.input(file, threads=0)
File "/home/brcisna/.pyenv/versions/3.10.12/envs/RVC-Studio/lib/python3.10/site-packages/ffmpeg/_run.py", line 325, in run
raise Error('ffmpeg', out, err)
ffmpeg._run.Error: ffmpeg error (see stderr output for detail)
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/brcisna/MyPrograms/rvcstudio/RVC-Studio/preprocessing_utils.py", line 73, in pipeline
audio = load_audio(path, self.sr)
File "/home/brcisna/MyPrograms/rvcstudio/RVC-Studio/lib/audio.py", line 43, in load_audio
raise RuntimeError(f"Failed to load audio: {e}")
RuntimeError: Failed to load audio: ffmpeg error (see stderr output for detail)
end preprocess`
EDIT
Looking closer this is Windows specific,to download ffmpeg from the landing page of RVC-Studio
My mistake,sorry.
Debian 12
Python=3.10.12
Am having troubles when i do a training session. I get tracebacks in regards to ffmpeg,
ffmpeg-python is installed
Question:
In the Home.py , the following lines have me puzzled. as Linux contains NO .exe files
Am showing my stupidity,, but is there something in python that can mimic 'exe /Windows executables
Home.py
lines 33 & 38 , ffmpeg.exe
line #33 - def render_download_ffmpeg(lib_name="ffmpeg.exe"):
col1, col2 = st.columns(2)
is_downloaded = os.path.exists(lib_name)
col1.checkbox(os.path.basename(lib_name),value=is_downloaded,disabled=True)
if col2.button("Download",disabled=is_downloaded,key=lib_name):
line #38 - link = f"{RVC_DOWNLOAD_LINK}ffmpeg.exe"
with st.spinner(f"Downloading from {link} to {lib_name}"):
file_downloader((lib_name,link))
st.experimental_rerun()
TIA
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.