✨│ai-help

[proper shale]

what colab/fork r u using

[full imp]

RVC v2 Disconnected

[proper shale]

have u run set training variables?

[full imp]

I'm the blindest of the blinds, I was doing it again and for some reason I jumped over several steps

[molten pecan]

https://rentry.org/RVC_making-models

Hope that helps

[fluid topaz]

thanks but this guide only explain what tensorboard does, it doesn't show on how to start it on the first place

[molten pecan]

https://docs.aihub.wtf/guide-to-create-a-model/tensorboard-rvc this might be outdated

[proper shale]

it is

[molten pecan]

Thought so

[proper shale]

#✨│ai-help message put this on ur rvc folder

and run

[molten pecan]

A tensorboard is a way to graphically display a graph of a model's parameters and how they change over the time of training. This way you can know what the optimal number of epochs is. Usually the only values that are monitored are G_loss and D_loss for RVC models.

Thanks Alex for that

[brittle wing]

An someone please help me with the bots? I’ve been struggling with it for weeks ☹️

I’ve been trying out the Wordband cover bot, and when I use it, I don’t know where my covers are. Like when it finishes, how do I know? It doesn’t message me with a link to the result or tell me when it’s done and I’ve been so confused trying to use it!

[fluid topaz]

ok thanks its working

[wanton knot]

"/content/drive/MyDrive/project-main

NameError Traceback (most recent call last)
<ipython-input-1-59a7acf51464> in <cell line: 79>()
77 )
78
---> 79 training_log = train_index(model_name, 'v2')
80
81 for line in training_log:

NameError: name 'model_name' is not defined" this keeps happening on Train Index

never had this issue

[rare goblet]

Ayo? @wanton knot level 7 !!!

[icy thicket]

help

you hear something robot

[molten pecan]

did you log in on wordband first?

you gotta do it or else it won't give you the covers

https://rentry.co/WordBand

here's a guide on it :3

[molten pecan]

qué 😦

[brittle wing]

Ohhhhhhh ok ty

[rare goblet]

Ayo? @brittle wing level 2 !!!

[molten pecan]

did you try reloading the colab page?

[frozen iron]

trying to extract features when training a model floods me with such errors:

    x = self.cnn(self.unet(mel)).transpose(1, 2).flatten(-2)
  File "C:\Users\jkier\AppData\Local\Programs\Python\Python39\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "D:\Applio-RVC-Fork\lib\modules\infer\rmvpe.py", line 426, in forward
    x, concat_tensors = self.encoder(x)
  File "C:\Users\jkier\AppData\Local\Programs\Python\Python39\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "D:\Applio-RVC-Fork\lib\modules\infer\rmvpe.py", line 303, in forward
    x = self.bn(x)
  File "C:\Users\jkier\AppData\Local\Programs\Python\Python39\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "C:\Users\jkier\AppData\Local\Programs\Python\Python39\lib\site-packages\torch\nn\modules\batchnorm.py", line 171, in forward
    return F.batch_norm(
  File "C:\Users\jkier\AppData\Local\Programs\Python\Python39\lib\site-packages\torch\nn\functional.py", line 2450, in batch_norm
    return torch.batch_norm(
RuntimeError: Expected tensor for argument #1 'input' to have the same type as tensor for argument #2 'weight'; but type torch.cuda.FloatTensor does not equal torch.cuda.HalfTensor (while checking arguments for cudnn_batch_norm)```

for some reason i can't train anything in applio

feature extraction does this

and then training the model does very quick epochs and leaves gpu usage on 0%

esentially not training the model at all

i've even tried a full clean windows installation

[brittle wing]

Please help..I feel like Sisyphus right now 💀

When I hit train model, it says it's done but doesn't even show up in the weights folder

And it finishes weirdly fast so I know I probably did something wrong

[frozen iron]

yeah i feel like ive got a similar issue rn

[rare goblet]

Ayo? @frozen iron level 1 !!!

[brittle wing]

God what are we going wrong.

[rare goblet]

Ayo? @brittle wing level 4 !!!

[frozen iron]

idk it randomly stopped working

[brittle wing]

if someone could help me work on this model/say how to make one using voice clips, thatd be awesome

[radiant whale]

guys

how to download our model from kaggle?

[wanton knot]

i figured out it was the audio file, somehow

[rancid pine]

I've tried using a smaller dataset, still comes up with cracks. I've already ensured it was clean, I do not understand what could be disrupting it..

I suppose this isn't making my life easier either

It sounds worse, even

[rare goblet]

Ayo? @rancid pine level 4 !!!

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation by country

🇧🇷 Brasil (PT-BR)

https://docs.aihub.wtf/v/brazil

🇪🇸 Spain

https://docs.aihub.wtf/v/spain/

🇫🇷 France

https://docs.aihub.wtf/v/france

🇩🇪 Germany

https://docs.aihub.wtf/v/germany

🇯🇵 Japan

https://docs.aihub.wtf/v/japan

🇰🇷 Korea

https://docs.aihub.wtf/v/korea

🇵🇱 Poland

https://docs.aihub.wtf/v/poland

🇷🇺 Russia

https://docs.aihub.wtf/v/russia

🇮🇹 Italy

https://docs.aihub.wtf/v/italy

[rancid pine]

I'm fucked, kek

The audio flats at 16kHz

I'll try it iout

[brittle wing]

I'm currently playing around with training with RVC, specifically with RavenCutie21's modification, and I have nvidia cuda 1.17 installed on my computer, but when I launch the webapp the console says no compatible nvidia gpu found. Is there a minimum gpu that can be used or did I install the wrong version of cuda

[radiant whale]

any one know how to download our model in keggle?

[brittle wing]

It's a 2060 super, but I might test it with a 3060 to see if it works

[sand iris]

How'd you install it ?

(RVC, I mean)

[brittle wing]

Off of https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/releases/tag/updated1006v2 downloading the amd package, as I have a 6600 as my primary gpu in my system.. Actually I think I might have maybe found the solution

[rare goblet]

Ayo? @brittle wing level 1 !!!

[sand iris]

Oh yeah that's the thing

The AMD release doesn't contain (most of) the CUDA packages

You'd need the Nvidia release

[brittle wing]

Don't really know what went through my mind to download the amd version ¯_(ツ)_/¯

Thanks for the help!

[olive otter]

Is there anyway using rvc on mobile

Or do you have to use a computer

[bleak yacht]

Why can this happen? If you don't !chmod 775 <path>, it's just permission denied. it's Google Colab.

[simple igloo]

Hi there, is there a way to hear myself in a discord call while talking?

I wanna be able to hear the voice changer version of me talking

[sand iris]

use the third sound output

"monitor"

[sand iris]

which notebook is that ?

[bleak yacht]

It's a test one, I just loaded the applio in there to test it.

[sand iris]

I don't think you need to make anything executable

you might have screwed up something

setup goes something like :
clone code
install requirements via pip
python infer-web.py (maybe needs a one line change to get a public url, or use ngrok)

[bleak yacht]

Definitely did not spoil anything, because in the same Kaggle everything works without any problems

[rare goblet]

Ayo? @bleak yacht level 3 !!!

[simple igloo]

Thank you, fixed the problem

[sand iris]

can you send the code over so I can check ?

[bleak yacht]

Yeah, just a minute.

[bleak yacht]

first cell:
import codecs
appl = codecs.decode('uggcf://tvguho.pbz/VNUvfcnab/Nccyvb-EIP-Sbex', 'rot_13')
!git clone $appl
cdd = codecs.decode('Nccyvb-EIP-Sbex', 'rot_13')
%cd $cdd
!pip install -q -r assets/requirements/requirements.txt

second cell:
import codecs
runapp = codecs.decode('vasre-jro.cl', 'rot_13')
!python $runapp --colab

This uses the Caesar cipher and the usual fork of the applio.

[sand iris]

ya

[bleak yacht]

I'm more interested in why it runs via bash/sh without using the python interpreter

[sand iris]

that's what it appears to be doing with run.py too

[bleak yacht]

where run.py is my fork made from applio, it does the same thing as the original applio, then it doesn't see the files, then no permissions at all

[sand iris]

last time I checked it (locally), I didn't need to change any permissions

stock git clone just worked

[bleak yacht]

That's the thing, it works locally, but it won't work on google colab.

[sand iris]

Wait how is it running infer-web.py yet also trying to run it afterwards

[bleak yacht]

XD, I'm just now realizing

• he's trying to do it through sh instead of the python interpreter

[true vortex]

the name speaks for itself
while training something accessed the memory without permission, which resulted in an error
check your antivirus

[radiant whale]

guys

should i change folder name for your datasets?

[sand iris]

I took a big look at it but I don't understand why it's doing that

[rare goblet]

Ayo? @sand iris level 53 !!!

[sand iris]

Only thing that could be happening is that the python command magically got symlinked to sh

Somehow

can you !bash -c "ls -all $(which python)"

[sand iris]

Okay so python is indeed python

why the fuck is it using shell

[bleak yacht]

google's a funny guy.)

[sand iris]

Shitty idea :

put

#!/usr/bin/env python

on the very first line of every python file

This way even if it's sh that somehow runs it

it'll use the python interpreter

[bleak yacht]

should I fork applio and write it in every file?))))))

[sand iris]

Probably not

It's a very bad idea but it could just work

[bleak yacht]

it's a very crazy idea.

[sand iris]

But I'm still confused: why does it properly launches infer-web.py and then tries to run it via shell

[bleak yacht]

maybe try installing python 3.9 and making it the main interpreter?

[sand iris]

3.10 works

[candid lichen]

Heyo, need help getting a index file for my voice model on colab

[bleak yacht]

Well, then I don't know what to do. :/

[candid lichen]

I had made one of my own voice and have the pth, but I also wanna redue it and also have the index

[sand iris]

I'm as confused as you are about this lol

[bleak yacht]

while in the same kaggle the same version of python and set of libraries and everything works.

that's exactly what google colab does.

Can the --colab flag do that?

[sand iris]

wait can you link me your fork ?

[bleak yacht]

it doesn't differ anything from the latest version of applio except the name of the startup file, how does it help you?

[sand iris]

name of the startup file ?

Did you remove infer-web.py

[bleak yacht]

I renamed it to run.py to prevent Kaggle from blocking startup, in previous versions of the fork it didn't have any impact

[sand iris]

Oh wait no the thing you sent earlier is the original Applio code, nevermind

[bleak yacht]

yea, original applio

[rare goblet]

Ayo? @bleak yacht level 4 !!!

[bleak yacht]

lol, if you run the file without the --colab flag and listen to port 7865 through local tunnel everything works, what's wrong with the flag).

[sand iris]

The flag only creates a public gradio URL

Apparently

[candid lichen]

Is index files suppose to be over 800mb?

Cause the managed to find it and the whole file was over 800MB

I*

[bleak yacht]

I realized how to solve the problem, just instead of --colab write --paperspace and --pycmd python, the script will work without listening to the port

[bleak yacht]

Are you talking about .index? If yes, the index size depends on the dataset size

[candid lichen]

Yeah

I'm just tryna add index to my pth so I can rezip it and use it again

But trying to just add the folder itself, then the voice.pth itself and see if that works instead of just adding the files but if that doesn't I'll just take the files out of the folder and rezip the whole model

[blissful pike]

what should I do if my voice repeats itself as an echo after I have spoken a phrase?

[proud elbow]

it's supposed to be for 1-1.5 hour dataset

[fluid topaz]

can someone give a hint about whats the menu for the thing i pointed with red mark, i understand everything else

oops should be in #1159290066074607626

but it won't let me send pic

[dull gazelle]

why the fuk im getting tenserboard not found when i try to install applio

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation by country

🇧🇷 Brasil (PT-BR)

https://docs.aihub.wtf/v/brazil

🇪🇸 Spain

https://docs.aihub.wtf/v/spain/

🇫🇷 France

https://docs.aihub.wtf/v/france

🇩🇪 Germany

https://docs.aihub.wtf/v/germany

🇯🇵 Japan

https://docs.aihub.wtf/v/japan

🇰🇷 Korea

https://docs.aihub.wtf/v/korea

🇵🇱 Poland

https://docs.aihub.wtf/v/poland

🇷🇺 Russia

https://docs.aihub.wtf/v/russia

🇮🇹 Italy

https://docs.aihub.wtf/v/italy

[brittle wing]

@hidden drum

[hidden drum]

so what do i do on the folder

[brittle wing]

Open up a command prompt in the folder.

If you're on Windows 11, just right-click and click "Open in Terminal"

[hidden drum]

oh okay

[brittle wing]

Just to make sure, you have Python installed and in your PATH right?

[hidden drum]

it take me to powershell

[hidden drum]

yeah

[brittle wing]

Okay sweet. Powershell is just the default set, it's okay.

Run python -m venv [insertnamehere] (I just call it venv lol)

[hidden drum]

okay so i do that cmd

[brittle wing]

Just do it in the terminal it opened

:P

[hidden drum]

wait so i copy the brackets or no

[brittle wing]

no

[hidden drum]

okay

[brittle wing]

python -m venv venv

for example ^

[hidden drum]

it gave me blank now

[brittle wing]

That's fine.

Run this command now.

It'll go into the virtual environment.

[hidden drum]

can you copy the run i have to put in the cmd

[brittle wing]

.\venv\Scripts\activate

(replace venv with whatever you called your environment)

[hidden drum]

wait i dont get it i get a red error code

[brittle wing]

Take a screenshot.

[hidden drum]

.\venv\Scripts\activate : File C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voi
ce-Conversion-WebUI-main\venv\Scripts\Activate.ps1 cannot be loaded because running scripts is disabled on this
system. For more information, see about_Execution_Policies at https:/go.microsoft.com/fwlink/?LinkID=135170.
At line:1 char:1

• .\venv\Scripts\activate

•   + CategoryInfo          : SecurityError: (:) [], PSSecurityException
    + FullyQualifiedErrorId : UnauthorizedAccess

this one

[brittle wing]

Ohhh you have to edit your policies, one sec.

Set-ExecutionPolicy Unrestricted -Scope Process

[hidden drum]

okay i didnt get any error code

what next

[brittle wing]

You're in the venv?

(it should say venv in brackets)

[hidden drum]

no but i did the policie cmd

[brittle wing]

oh

try running the venv thing again

[brittle wing]

this

[hidden drum]

okay now it shows venv with green bracket

[brittle wing]

Sweeto!

Now run pip install -r .\requirements.txt

That will install the requirements inside of your venv environment.

[hidden drum]

what do i run for 3060 graphics card

[brittle wing]

Just that will be fine

[hidden drum]

like how much do i run when im making ai

[brittle wing]

Wdym by that?

[sand iris]

any reason y'all are setting up using a venv rn ? Do the provided releases not work ?

[hidden drum]

like what do i set the settings to on local while making ai on a 3060 graphics card

[brittle wing]

Yes, I'm showing them how to run mainline so that their requirements installed don't mess with any other requirements in case they install anything else.

[hidden drum]

bc i remember when i was making ai my pc was overheating asf

[brittle wing]

It's good practice to run different projects in venvs.

[sand iris]

There's runtime releases with mainline

which is why I'm wondering

[brittle wing]

Yes I know

But as I say it's good practice to do this inside of a venv due to that reason.

[hidden drum]

why when i make ai on local my pc overheats asf

it feels like a oven

[brittle wing]

We'll get there when you got it open xD

[hidden drum]

when making like ai on a 3060 graphics card lol

[brittle wing]

are you running the requirements installation command?

[hidden drum]

yup i am

shows this rn Using cached pywin32-306-cp39-cp39-win_amd64.whl (9.3 MB)
Using legacy 'setup.py install' for ffmpy, since package 'wheel' is not installed.
Using legacy 'setup.py install' for antlr4-python3-runtime, since package 'wheel' is not installed.
Using legacy 'setup.py install' for future, since package 'wheel' is not installed.
Installing collected packages: rpds-py, attrs, urllib3, typing-extensions, sniffio, six, referencing, pycparser, pyasn1, numpy, mpmath, markupsafe, llvmlite, idna, h11, exceptiongroup, colorama, charset-normalizer, certifi, zipp, uc-micro-py, tzdata, tqdm, threadpoolctl, sympy, scipy, rsa, requests, PyYAML, pywin32, pytz, python-dateutil, pyreadline3, pyasn1-modules, platformdirs, packaging, oauthlib, numba, networkx, multidict, mdurl, jsonschema-specifications, joblib, Jinja2, httpcore, fsspec, frozenlist, filelock, cffi, cachetools, anyio, yarl, websockets, torch, toolz, tabulate, starlette, soundfile, scikit-learn, resampy, requests-oauthlib, regex, pyparsing, pydantic, portalocker, pooch, Pillow, pandas, omegaconf, markdown-it-py, lxml, linkify-it-py, kiwisolver, jsonschema, importlib-resources, importlib-metadata, humanfriendly, huggingface-hub, httpx, google-auth, fonttools, decorator, cycler, contourpy, click, audioread, async-timeout, antlr4-python3-runtime, aiosignal, Werkzeug, uvicorn, traitlets, torchaudio, tensorboard-data-server, semantic-version, sacrebleu, python-multipart, pygments, pydub, protobuf, orjson, mdit-py-plugins, matplotlib, Markdown, librosa, hydra-core, grpcio, gradio-client, google-auth-oauthlib, future, flatbuffers, ffmpy, fastapi, Cython, coloredlogs, bitarray, altair, aiohttp, aiofiles, absl-py, tornado, torchcrepe, tensorboardX, tensorboard, pyworld, python-dotenv, praat-parselmouth, onnxruntime-gpu, matplotlib-inline, json5, gradio, ffmpeg-python, faiss-cpu, fairseq, av

[brittle wing]

That's fine, it'll install those things.

[hidden drum]

never use ai in a very long time

[rare goblet]

Ayo? @hidden drum level 4 !!!

[karmic canopy]

how do i fix this

[rare goblet]

Ayo? @karmic canopy level 2 !!!

[hidden drum]

okay done whats next?

[brittle wing]

Those have already installed??

jeezus

[hidden drum]

yup

[brittle wing]

okay well, do python .\infer-web.py and then that should start RVC

[hidden drum]

what does the warning pip version mean

[brittle wing]

It's nothing to be concerned about, just tells you there's a new pip version.

[hidden drum]

2023-12-24 07:50:51 | INFO | configs.config | No supported Nvidia GPU found
overwrite preprocess and configs.json
2023-12-24 07:50:51 | INFO | configs.config | Use cpu instead
is_half:False, device:cpu
2023-12-24 07:50:54 | INFO | main | Use Language: en_US
Running on local URL: is this bad or no

im on the site btw just deleted the link

but is the no supported Nvidia GPU found bad @brittle wing

[brittle wing]

Not a bad thing, but I need to find the command to enable your GPU processing.

pip install numpy --pre torch torchvision torchaudio --force-reinstall --index-url https://download.pytorch.org/whl/nightly/cu118

this

[gusty kestrel]

No results found for the search arjit singh...

Powered by Applio — Make sure you spelled it correctly!

[hidden drum]

do i close the site @brittle wing

[brittle wing]

Awesome it's running then, you just have to sort out the gpu acceleration thing with that cu118 thing and it should be fine

[gusty kestrel]

I have found 2 results for the search anuv jain...

Anuv Jain (Indian Singer)

Owner: virginice
Uploaded: <t:1699910940:R>

Epochs

200

Technology

RVC

Algorithm

Harvest

Requested by sanskaar. · Dec 24, 2023, 12:53 PM

[brittle wing]

Press CTRL + C in your command prompt and it'll turn off the script

you don't need to exit and re-enter :)

the gradio interface you can close for now tho if you'd like.

[hidden drum]

bet K

[brittle wing]

@ancient arch #🔍│find-models.

[burnt root]

is there any way to find older ai voice models? because i remember there being a cloaker voice model that i used the crap out of but its gone now..

[brittle wing]

There's a model tracker somewhere, I'd have to ffind it though LOL.

[burnt root]

found one, but it doesnt have 350 epochs like the folder i extracted it to..

[hidden drum]

when your running an ai model does your pc feel like a fucking campfire inside your pc

[brittle wing]

Not necessarily.

[burnt root]

i deleted the contents for some reason and now i am regretting everything

[hidden drum]

why for me it does

[brittle wing]

Maybe you need to sort out your fan curves

[hidden drum]

Is this good

idk if its curve

[brittle wing]

nono I mean the fan curves in the BIOS Jace.

like the speed your fans go at with certain temps

xD

[hidden drum]

oh lol mb

[brittle wing]

How's the cu118 thing coming along?

(It may take a while sometimes because NVIDIA's download servers are dogwater)

[hidden drum]

Is this a good fan speed idk

i didnt change it but thats the default its going at

[hidden drum]

@brittle wing

[brittle wing]

I'm not on about that mate, the fan speed is fine ig tho

[hidden drum]

|█████████████ | 1082.4 MB 3.3 MB/s eta 0:08:11 its almost done lol

[brittle wing]

Lucky bugger

it took me like an hour

[hidden drum]

just a half more pain to suffer on the loading screen

[hidden drum]

bc im using bell its the best wifi in canada

but not sure if it has to do with anything im doing rn tho XD

[brittle wing]

I'm using like 750mbps download internet

It's the NVIDIA servers though on their end, not yours.

Sometimes their download servers just suck.

[hidden drum]

True

im running 22.22 download speed on a 2.4 ghz internet bc im inside the room and my wifi router is outside the living room LMFAO

[brittle wing]

alrighty whilst that's doing, lemme zip some files up for you that you will need to shove in the assets folder for inferring to work.

[hidden drum]

This is my phone but idk why my pc is this bad

idk its this bad on my pc

@brittle wing

[brittle wing]

Just uploading ssomething for you to download.

[hidden drum]

@brittle wing wdym?

[brittle wing]

The assets folder inside the zipped archive just shove into the RVC folder

[brittle wing]

There's a few pretrains, rmvpe and hubert you need for inferring (and training) to work.

They don't come in the mainline package and you have to get them yourself, but since I have them I'll just send em.

[hidden drum]

why is my pc internet like that and my phone speed is damn fast

[brittle wing]

Possibly because your connection is poor to the PC?

[hidden drum]

is this good ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
numba 0.56.4 requires numpy<1.24,>=1.18, but you have numpy 1.24.1 which is incompatible.
huggingface-hub 0.20.1 requires fsspec>=2023.5.0, but you have fsspec 2023.4.0 which is incompatible.
Successfully installed MarkupSafe-2.1.3 certifi-2022.12.7 charset-normalizer-2.1.1 filelock-3.9.0 fsspec-2023.4.0 idna-3.4 jinja2-3.1.2 mpmath-1.2.1 networkx-3.0rc1 numpy-1.24.1 pillow-9.3.0 requests-2.28.1 sympy-1.11.1 torch-2.3.0.dev20231224+cu118 torchaudio-2.2.0.dev20231224+cu118 torchvision-0.18.0.dev20231224+cu118 typing-extensions-4.8.0 urllib3-1.26.13
WARNING: You are using pip version 21.2.4; however, version 23.3.2 is available.
You should consider upgrading via the 'C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\Scripts\python.exe -m pip install --upgrade pip' command.

or naw

[brittle wing]

It's fine

[hidden drum]

okay thanks

so what do i do

now

[brittle wing]

Try running RVC using the infer-web.py command I gave you earlier to see if it runs with the GPU detection

if so then use CTRL+C to cancel the script temporarily and shove in the files I will send you into the RVC folder.

Then you should be good to go after downloading some models.

[glad zealot]

what they doin?

[brittle wing]

I am guiding them to install RVC locally.

Mainline specifically.

[glad zealot]

ohhh

[brittle wing]

https://we.tl/t-PzQVjy1iYa @hidden drum alrighty shove those files into your RVC folder if you can confirm it works then should be sweeto

[glad zealot]

ye no idea i use mainline but online

[brittle wing]

(I've verified MD5 hash of the file downloaded so I can be certain it's safe btw)

[hidden drum]

K thanks and also on pc what does it mean abt link speed/recieve/transmit it shows for me 72/72 MBPS

[brittle wing]

Is it running?

[brittle wing]

Not going to lie I have no clue lol

[hidden drum]

No im just checking my network connection bc my pc network connection is trash asf

[bleak radish]

i need help

[frozen iron]

does anyone have a remote idea what might be causing these errors when trying to extract features when training in applio-rvc? 😭
ive tried conda and manual, system disk and not
even a clean windows install with all the prerequisites needed

[bleak radish]

i need help

help

help help

[brittle wing]

I'm halfway drunk so I'm just tryna get you good to go before I become completely wankered to be transparent with ya.

😛

[bleak radish]

is anyone helping me

helpppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp

[frozen iron]

dude tell your problem first

if anyone knows we'll help

[bleak radish]

i need help finding something

im trying to find a duke dennis ai vocie

[brittle wing]

BTW @hidden drum just for the record, when running RVC again in future you will have to enable the venv environment again to get it to run

[brittle wing]

Use @shy zealot in #🔍│find-models.

[bleak radish]

https://tenor.com/view/duke-dennis-type-shit-type-beat-gif-27562644

[brittle wing]

If there's a model there it should be there.

[bleak radish]

thats him

[hidden drum]

Ah shit okay

[brittle wing]

It's only just .\venv\Scripts\activate and python .\infer-web.py anyways so it's fine xD

[hidden drum]

When they update it or when i dont use it for a few years again i have to run the evironment again

[brittle wing]

not like it's that complicated tbh

[brittle wing]

As long as you do not mess with the venv or reinstall windows for example if you come back to it later it should just run perfectly normally. :P

[hidden drum]

Lol 😂 tbh rug roh but like 👍

[rare goblet]

Ayo? @hidden drum level 5 !!!

[brittle wing]

When an update happens though, just re-do the steps n stuff but it was not as hard as you thought it'd be.

[hidden drum]

K thanks it gave me error is it bc i have to put the asset in the folder first lol 😂

[brittle wing]

Mainline updates aren't often though so not a thing to worry about for a while.

[brittle wing]

What error?

[hidden drum]

When i run the python infer web cmd

[brittle wing]

Yeah but what does it say?

[hidden drum]

Traceback (most recent call last):
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer-web.py", line 8, in <module>
from infer.modules.vc.modules import VC
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer\modules\vc\modules.py", line 11, in <module>
from infer.lib.audio import load_audio, wav2
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer\lib\audio.py", line 4, in <module>
import librosa
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa_init_.py", line 209, in <module>
from . import core
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\core_init_.py", line 5, in <module>
from .convert import * # pylint: disable=wildcard-import
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\core\convert.py", line 7, in <module>
from . import notation
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\core\notation.py", line 8, in <module>
from ..util.exceptions import ParameterError
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\util_init_.py", line 77, in <module>
from .utils import * # pylint: disable=wildcard-import

it says this

[brittle wing]

You executed this command in the venv right?

Just making sure.

[hidden drum]

yup

[brittle wing]

If so, try pip uninstall numpy and pip uninstall numba, apparently this does something?

oh boy do I love python errors

ohh okay numba makes sense because it failed at initialization of a numba package

[hidden drum]

did it but it says no numpy in cmd

[brittle wing]

huh

[hidden drum]

ModuleNotFoundError: No module named 'numpy'

[brittle wing]

oH IM DUMB

install***

install numpy and numba

:LSDJFLPK

LOL

[hidden drum]

how do i do that?

[brittle wing]

same commands but replace "uninstall" with install lol

pip install numpy and pip install numba

[bleak radish]

i need help

[rare goblet]

Ayo? @bleak radish level 1 !!!

[bleak radish]

donloading it

it says Windows protected your PC

[hidden drum]

okay did it what i do next @brittle wing

[bleak radish]

help

[hidden drum]

@brittle wing what do i do now

do i put assest in folder now

charlotte what do i do now

Traceback (most recent call last):
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer-web.py", line 8, in <module>
from infer.modules.vc.modules import VC
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer\modules\vc\modules.py", line 11, in <module>
from infer.lib.audio import load_audio, wav2
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer\lib\audio.py", line 4, in <module>
import librosa
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa_init_.py", line 209, in <module>
from . import core
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\core_init_.py", line 5, in <module>
from .convert import * # pylint: disable=wildcard-import
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\core\convert.py", line 7, in <module>
from . import notation
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\core\notation.py", line 8, in <module>
from ..util.exceptions import ParameterError
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\util_init_.py", line 77, in <module>
from .utils import * # pylint: disable=wildcard-import
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-

what do i do now 😦

@brittle wing

[sand iris]

what have y'all done up until now ?

[hidden drum]

idk but im what do i fix for this FileNotFoundError: [WinError 3] The system cannot find the path specified: 'assets/uvr5_weights'

[sand iris]

oh you fixed the previous thing ? great

wait

[hidden drum]

and when i add the assests it dosent work

[sand iris]

charlotte's zip archive didn't have the uvr weights?

[hidden drum]

nope

[hidden drum]

do you have one that has uvr weights

[brittle wing]

For me it worked without them lol

[sand iris]

I do have it, I can't upload it atm

[brittle wing]

Sorry for lack of reply I’m busy atm

Family n stuff

[hidden drum]

Traceback (most recent call last):
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer-web.py", line 8, in <module>
from infer.modules.vc.modules import VC
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer\modules\vc\modules.py", line 11, in <module>
from infer.lib.audio import load_audio, wav2
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer\lib\audio.py", line 4, in <module>
import librosa
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa__init.py", line 209, in <module>
from . import core
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\core__init.py", line 5, in <module>
from .convert import * # pylint: disable=wildcard-import
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\core\convert.py", line 7, in <module>
from . import notation
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\core\notation.py", line 8, in <module>
from ..util.exceptions import ParameterError
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\venv\lib\site-packages\librosa\util__init__.py", line 77, in <module>
from .utils import * # pylint: disable=wildcard-import
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI- it shows this tho

i have this error

[sand iris]

lovely

[hidden drum]

indeed

[sand iris]

AAAAAAA

how to bork stuff 101

numpy will make other packages bitch

[brittle wing]

It worked fine for me daroche 😭

[hidden drum]

PS C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main> python .\infer-web.py
2023-12-24 08:49:54 | INFO | configs.config | Found GPU NVIDIA GeForce RTX 3060
is_half:True, device:cuda:0
2023-12-24 08:49:55 | INFO | main | Use Language: en_US
Traceback (most recent call last):
File "C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-main\Retrieval-based-Voice-Conversion-WebUI-main\infer-web.py", line 143, in <module>
for name in os.listdir(weight_uvr5_root):
FileNotFoundError: [WinError 3] The system cannot find the path specified: 'assets/uvr5_weights'

[sand iris]

can you .\tools\dlmodels.bat

it'll download all the assets

[hidden drum]

which do i download it took me to github page

[sand iris]

Huh

OH

[hidden drum]

its says press key or something

[sand iris]

yeah my bad, the script is worse than I had anticipated

[hidden drum]

what do i download

[sand iris]

Okay, I think a manual install is leading nowhere

Unless continue with the manual install by downloading the -win archive at https://github.com/aria2/aria2/releases/tag/release-1.36.0

and unzip it in the RVC folder

OR

Just reinstall everything from scratch with one of the RVC releases rather than setup manually

[hidden drum]

wdym abt that

which model do i download anyways

😭

[sand iris]

setup is as simple as

• download file
• unzip file
• click run

What y'all have been doing is a manual install which is way more trouble and work

[hidden drum]

oh okay

[sand iris]

https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/releases
Just grab the file from the "for nvidia gpu users" link

That's pretty much the intended way for non-developers to use RVC on windows

[hidden drum]

Okay

[hidden drum]

C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-updated1006v2\Retrieval-based-Voice-Conversion-WebUI-updated1006v2>runtime\python.exe infer-web.py --pycmd runtime\python.exe --port 7897
The system cannot find the path specified.

C:\Users\jayce\Downloads\Retrieval-based-Voice-Conversion-WebUI-updated1006v2\Retrieval-based-Voice-Conversion-WebUI-updated1006v2>pause
Press any key to continue . . .

what does this mean

[sand iris]

Did you unzip it first ?

[hidden drum]

i did

[sand iris]

is everything else also unzipped?

[hidden drum]

yes it is

[sand iris]

What the hell is up with it

[sand iris]

You sure you did not grab the "source code" archive ?

[hidden drum]

no

i got urs

the one you link

[sand iris]

RVC1006Nvidia.7z ?

[hidden drum]

but the run when i click on it disappers

[hidden drum]

7z i didnt got that

[bleak radish]

same

same

[sand iris]

https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/RVC1006Nvidia.7z

[bleak radish]

thats what happend to me

every time i run it it dissappers

[hidden drum]

okay downloading it

[bleak radish]

its done downloading

[sand iris]

Ya that seems more like it lol. You got the source code last time lol

[bleak radish]

what now

[hidden drum]

oh okay

[sand iris]

Unzip, and just run go-web

[bleak radish]

how do i unzip

[hidden drum]

right click

unzip

[sand iris]

You on windows 10 or 11 ? updated windows 11 machines should be able to unzip

[bleak radish]

ok

do i run it

[hidden drum]

it shows extract it thats how it unzips the folder

[bleak radish]

wendos

10

[sand iris]

https://www.7-zip.org/a/7z2301-x64.msi

It requires 7zip then

[hidden drum]

im on windows 11

[hidden drum]

what abt for 11

also

[sand iris]

if it's updated I'm like 90% sure windows natively supports 7zip

else, just get it from where I linked up above

[hidden drum]

where do i see if it supports 7zip for me

[rare goblet]

Ayo? @hidden drum level 6 !!!

[sand iris]

if you can extract it, it's supported

else it isn't

[hidden drum]

oh okay

[bleak radish]

ok

it works

omg

g

[hidden drum]

mines like updated updated

[bleak radish]

g

[hidden drum]

how tf it download fast for you but not me 😭

[sand iris]

big files

[bleak radish]

fr

[hidden drum]

also im using winrar is it fine or use 7zip

[sand iris]

winrar also works yeah

[hidden drum]

okay

[bleak radish]

ok its done dowloading what now

[hidden drum]

so when it downloads do i run bat and it works

after

[sand iris]

yeah

[hidden drum]

brb i hope the msg dosent get full when i comeback lmfao

[sand iris]

I don't do model training or inference, I'm not gonna be really useful there lol

But I can help with any issues related to python

[hidden drum]

ah okay

[bleak radish]

uhhh

uhuhuh

im done wtih rvc

[hidden drum]

Why lol 😂

[bleak radish]

my thing keeps dissapering

help

[rare goblet]

Ayo? @bleak radish level 2 !!!

[hidden drum]

@sand iris why does my thing say unsupported please help

[sand iris]

screenshot?

I don't understand

[hidden drum]

it says like while extracting the file the file is unsupported

[sand iris]

Ah

So that's what it is. We've seen that before but didn't understand why

it's winrar

[hidden drum]

so what do i do

[sand iris]

can you try with 7zip instead ?

[hidden drum]

sure

[steel saffron]

can someone help me please

[hidden drum]

it says some files are currently being used to for 7zip @sand iris

[sand iris]

if you extracted in the same place yes, it's gonna tell you the files already exist

if that's what you meant

[sand iris]

with what ?

[steel saffron]

when i load my dataset i have this :Dataset folder already found. Wiping...
Temporary folder already found. Wiping...
Archive: /content/drive/MyDrive/RVC-Disconnected/zipfile.zip
creating: /content/temp_dataset/zipfile/character_name/
inflating: /content/temp_dataset/zipfile/character_name/data.wav
inflating: /content/temp_dataset/zipfile/character_name/Recording (4).mp3
inflating: /content/temp_dataset/zipfile/data.wav
inflating: /content/temp_dataset/zipfile/Recording (4).mp3
Sanitizing...
Dataset Type: Multispeaker

FileNotFoundError Traceback (most recent call last)
/usr/lib/python3.10/shutil.py in move(src, dst, copy_function)
815 try:
--> 816 os.rename(src, real_dst)
817 except OSError:

FileNotFoundError: [Errno 2] No such file or directory: '/content/temp_dataset/zipfile/character_name' -> '/content/dataset/character_name'

During handling of the above exception, another exception occurred:

FileNotFoundError Traceback (most recent call last)
3 frames
/usr/lib/python3.10/shutil.py in copyfile(src, dst, follow_symlinks)
252 os.symlink(os.readlink(src), dst)
253 else:
--> 254 with open(src, 'rb') as fsrc:
255 try:
256 with open(dst, 'wb') as fdst:

FileNotFoundError: [Errno 2] No such file or directory: '/content/temp_dataset/zipfile/character_name'

[hidden drum]

nvm

it works

what do i extract with it still show winrar

for extract

@sand iris

[sand iris]

how'd you get 7zip ?

did you get the .msi or .exe from their website?

[hidden drum]

msi version

[sand iris]

It's installed right? not only downloaded

[hidden drum]

yeah

its installed

[bleak radish]

fuck

[hidden drum]

but still shows winrar

[bleak radish]

its npt working

[hidden drum]

@sand iris

[sand iris]

if you right click you can choose "open with"

[hidden drum]

nvm

[sand iris]

idk, I don't use colab

you did fill in everything right?
I think there's a field you forgot to change

[hidden drum]

i had to do show more options

its working fine so far

[steel saffron]

what can i use instead of colab

[hidden drum]

Fuck YESSSS IT WORKS @sand iris

thanks bro

[sand iris]

for training, other than locally, I don't think there are other free versions

[hidden drum]

your a legend

[sand iris]

Ya great

sorry this took a while lol

[hidden drum]

its fine lmfao

[hidden drum]

which is good to add on the album of juice wrld rental

like which ai person is good

kevin im guessing but theres so many

[sand iris]

Oh boy, yeah, there's a whole lot of choice

[hidden drum]

yup just dont abuse ai lol

[sand iris]

you can unless browse #1175430844685484042 , use the search bot in #🔍│find-models or https://weights.gg/

[hidden drum]

ima link the discord channel on my yt so ppl can know

[steel saffron]

you need an nvidia gpu or amd work ?

[rare goblet]

Ayo? @steel saffron level 1 !!!

[hidden drum]

how do import models in here

[sand iris]

to create voice models or use them ?

[steel saffron]

for create

[sand iris]

nvidia only

[steel saffron]

ahh thats why it dont work

[sand iris]

(unless you happen to use linux then AMD also works)

[hidden drum]

@sand iris how do i like import the ai models of ppl models in the file of rvc im using

[sand iris]

there should be a weights folder under assets\

the .pth models go there

and put .index files in the logs folder

[hidden drum]

K thanks

[steel saffron]

if i host a virtual machine with linux it would work ?

[sand iris]

(everything is found where you unzipped RVC)

[sand iris]

Unfortunately no

[steel saffron]

ahh

[hidden drum]

can i delete the sample pth or dont delete it

?

[sand iris]

don't

[hidden drum]

K

[hidden drum]

abt metadata file

it shows

[sand iris]

wdym

[hidden drum]

it shows a metadata file of a jason file

[hidden drum]

@sand iris where do i put the vocals of songs in the rvc folder

[sand iris]

Uh I think you give it the file path

[hidden drum]

nvm know how

[sand iris]

I don't think the vocals need to be anywhere specific

[tribal cosmos]

does anyone know what this error means??

[hidden drum]

@sand iris for training it says put the path of the training folder but idk where to put it tho

[sand iris]

it's the usual stuff, it's working as far as I'm seeing

[sand iris]

I don't train models, idk

[hidden drum]

and for the training which batch do i put it for bc i have a 2.0/8.0 GB of GPU memory

[molten pecan]

#📑│making-models

[sand iris]

What do you mean by 2/8 ?

Which GPU is it

[hidden drum]

3060

[sand iris]

Oh all good then lol, you scared me for a sec

[hidden drum]

@sand iris what do i put the batch for

?

[sand iris]

idk

6-8

If it errors, lower it

[hidden drum]

K

[hidden drum]

Gl on my pc overheating and feeling like a campfire

[sand iris]

overheating?

As long as it doesn't go 90°C+ it's rated for it

90 is the dangerous range and 100+ is the no go zone

[hidden drum]

when im training it feels like its overheating

[rare goblet]

Ayo? @hidden drum level 7 !!!

[bleak radish]

help

sosos

so

sos

oso

sos

[tribal cosmos]

broo... true, i stopped it 😭

[bleak radish]

i need help

my every time i run it it disapres

[violet heron]

Okada or RVC

If it’s w-Okada #🔍│help-w-okada message

[simple igloo]

Need some help with RVC, My Corsair Hs80 worked a few hours ago perfectly, worked in discord calls, could hear myself

[rare goblet]

Ayo? @simple igloo level 1 !!!

[simple igloo]

Then now, its broken

It like mutes my mic when I turn on RVC, cant hear anything and mic only works on discord if i set it to my normal one

I reinstalled VC Cable, but that didnt seem to help

Any open suggestions?

Just tested it, it completely broke my mic this time

I cant even hear myself with both input/output combined

Neither can discord detect any audio, nor can RVC detect any

It fixes when I restart my headset, but when I try to change anything on RVC it just mutes again

[violet heron]

That’s not RVC

That’s w-Okada

[simple igloo]

Its not realtime voice changer?

[violet heron]

Real-time voice changer is Okada

[simple igloo]

OH

[violet heron]

RVC is for making covers

[simple igloo]

Sorry for making that mistake then

[violet heron]

But RVC models work on Okada

[simple igloo]

Is the abbreviation for something else?

I've gotten help in this channel before

[molten pecan]

Aren't they both built with RVC?

[violet heron]

Okada is built with a bunch of different things

[molten pecan]

Oh shit really?

I see

[violet heron]

It has support for SoVits, RVC, Beatrice, and a few others

[molten pecan]

Yeah I still use so vits svc

How did you know?

[simple igloo]

I see

I've gotten it to work and let me monitor the voice, But discord isnt detecting my input, any advice? @violet heron

[deft locust]

i need the web site

[rare goblet]

Ayo? @simple igloo level 2 !!!

[simple igloo]

VB Control Panel says that its detecting input

and it matches with the delay of the ai voice

But discord isnt detecting ir

Ill try restarting discord

[violet heron]

For what

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation by country

🇧🇷 Brasil (PT-BR)

https://docs.aihub.wtf/v/brazil

🇪🇸 Spain

https://docs.aihub.wtf/v/spain/

🇫🇷 France

https://docs.aihub.wtf/v/france

🇩🇪 Germany

https://docs.aihub.wtf/v/germany

🇯🇵 Japan

https://docs.aihub.wtf/v/japan

🇰🇷 Korea

https://docs.aihub.wtf/v/korea

🇵🇱 Poland

https://docs.aihub.wtf/v/poland

🇷🇺 Russia

https://docs.aihub.wtf/v/russia

🇮🇹 Italy

https://docs.aihub.wtf/v/italy

[rancid pine]

Shit, there are still cracks, and the model sounds different than intended

[half cove]

#✨│ai-help

[elfin cairn]

hey im having an issue where i can hear my changed voice in the changer client demo but its playing my normal voice in obs and stuff

idk if this isthe right place please assist me if im wrong

[simple igloo]

do the same thing with discord

change mic input to whatever virtual thing ur using

OBS should pick up on that

[brittle wing]

anyone in the chat?

.

is there any tut for facefusion?

Is there a better alternitive to jukebox ai the software that can extend songs?

@molten pecan

[molten pecan]

No clue

[brittle wing]

this chat feels empty

[pliant remnant]

ModuleNotFoundError: No module named 'mega'

[rare goblet]

Ayo? @pliant remnant level 1 !!!

[pliant remnant]

anyone can help?

[proper shale]

try reuploading that model somewhere else

[flat nacelle]

hey im having trouble with rvc. whenever i train the model if gives me this error: " [WinError 3] The system cannot find the path specified: 'C:\Users\pat\Downloads\rvc\Mangio-RVC-v23.7.0/logs/mute/0_gt_wavs/mute40k.wav'

[old oracle]

-colab

[azure marsh]

☁️ Google Colabs

• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab

🤗 Hugginface Spaces

• Ilaria RVC, by thestingerx Hugginface Spaces
• RVC-HF, by r3gf Huggingface Spaces
• AICoverGen, by r3gf Huggingface Spaces
• RVC v2 Huggingface version, by Clebersla Huggingface Spaces
• Advanced RVC Inference, by neuclya Huggingface Spaces

🎄 More commands: -audio, -local, -overtrain, -realtime, -rvc, -help

[flat nacelle]

did you manage to fix this?

[thin bronze]

if i remember correctly I just deleted v1 and installed v2 instead, got no solution

[dusty topaz]

When I launch applio there is written no module named 'torch' can you help me please

[steep halo]

Is there a fork of mangio RVC that doesn't need fairseq? cause that god damn thing always fails to build for me, even if I have the build tools necessary and in the environment variable.

I don't wanna rely on colab when I have a good GPU to do the work

[sand iris]

Nope unfortunately

Can you screenshot your c++ build tools installation ?

you need two optional components to have it working

also note that python3.11 and newer will not work, also because of fairseq 😆

[flat nacelle]

when training, its been stuck at the message "Reducer buckets have been rebuilt in this iteration." for like 5 minutes. is this normal

[rare goblet]

Ayo? @flat nacelle level 1 !!!

[flat nacelle]

thanks

[steep halo]

I've been using 3.9. one sec lemme open the installer cause i might've went a bit overboard installing random shit i dont need cause i've nearly lost my sanity the last time i attempted to get it working.

Also question: Are the 2022 build tools the right ones for RVC?

[sand iris]

yeah, just make sure you have the "C++ Build Tools" selected, and also the two following optional componentx :

• MSVC v14*
• Windows 11 SDK (even on windows 10)

[steep halo]

uh can I dm you the screenshot and we can go from there?

[rare goblet]

Ayo? @steep halo level 1 !!!

[steep halo]

oh bruh

LMAO

now i can post images

this is what it looks like

I have this and... I have this

is it something wrong with my environment path? @sand iris

[sand iris]

huh yeah it should be working the way it currently is

[steep halo]

weird. perhaps the wrong build tools?

which ones specifically do I need?

[sand iris]

the ones you have right now

what's it failing on ?

[steep halo]

It fails building the wheel for Fairseq

everytime I try to install the requirements, it always fails on that

[sand iris]

really not sure what could be up with it

[steep halo]

theres a chance it might not be reading the build tools right. could you show me your environment variable paths for it?

[sand iris]

I don't have a windows machine on me rn

[steep halo]

welp that sucks

[sand iris]

last time I did try though, it was on a fresh win10 install, I narrowed down the deps to those two components I sent

[steep halo]

wait i just remembered I also have the 2019 build tools installed

should i get rid of them?

[sand iris]

I really can't think of any reasons why the old ones would conflict

[steep halo]

wtfffffff

[flat nacelle]

when training, its been stuck at the message "Reducer buckets have been rebuilt in this iteration." for like 5 minutes. is this normal

[steep halo]

well is there any branch of rvc that is up to date and doesn't use fairseq?

[rare goblet]

Ayo? @steep halo level 2 !!!

[sand iris]

yesn't

they all use fairseq, only diff is that you don't have to build it

https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/releases

[steep halo]

wait there's an all in one installation?

oh my goddddddddddddd thank you

[warm needle]

Please help, I keep trying to extract vocals/intrumentals using UVC, It takes me couple tries to successfuly extract, but it fails most of the time weirdly even though it was working fine couple months ago, i get a Make sure you've selected right GPU, and I do have selected the right one, i tried running from admin and making sure i selected the right gpu

what can i do?

[molten pecan]

Did you reinstall it already?

[warm needle]

ive just installed it couple days ago, should i still try reinstalling?

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation by country

🇧🇷 Brasil (PT-BR)

https://docs.aihub.wtf/v/brazil

🇪🇸 Spain

https://docs.aihub.wtf/v/spain/

🇫🇷 France

https://docs.aihub.wtf/v/france

🇩🇪 Germany

https://docs.aihub.wtf/v/germany

🇯🇵 Japan

https://docs.aihub.wtf/v/japan

🇰🇷 Korea

https://docs.aihub.wtf/v/korea

🇵🇱 Poland

https://docs.aihub.wtf/v/poland

🇷🇺 Russia

https://docs.aihub.wtf/v/russia

🇮🇹 Italy

https://docs.aihub.wtf/v/italy

[shy zealot]

**Method 1 🔎: ** The search accuracy is 100% 💕

Jschlatt

Upload the <t:1685523409>
Server uploaded: AI HUB (RIP)
Channel uploaded: AI HUB (RIP) Post
Profile Owner: analogspiderweb

Author

analogspiderweb

Discord User

analogspiderweb

Internal ID

121

Language

EN 🇺🇸

Type

RVC V2

Size

273 MB

Extractor

N/A

Epoch

500

Steps

0

Powered by ModelAI.net, Made with Love ❤️

[half cove]

@quiet glen #1159290666019471421

[mystic pagoda]

Is rvc still the best way to make ai covers or are there better alternatives/simpler?

[hard chasm]

Rvc is the way to go

[rare goblet]

Ayo? @hard chasm level 1 !!!

[half cove]

[buoyant saddle]

uhhh
how do I fix this?

[molten pecan]

@viscid moss @honest pond

[honest pond]

hmmm

you need to use cuda

[buoyant saddle]

? What do you mean?

[honest pond]

nvm

i think you can use cmd

[buoyant saddle]

okay? I don't exactly understand how to fix the problem..

[honest pond]

clic in your path and type cmd

[molten pecan]

CUDA is a technology developed by NVIDIA that uses the power of graphics cards to perform calculations that require great processing power. It is especially useful for AI tools, such as RVC.
If possible, we recommend using CUDA to optimize the process of training models.
CUDA is downloaded automatically as a part of the NVIDIA driver.

[buoyant saddle]

ok?

[molten pecan]

took that from the ai hub wiki

:3

[honest pond]

te actuerdas de como activar un env?

[molten pecan]

ni idea de que hablas

[buoyant saddle]

what now?

[honest pond]

weit

[buoyant saddle]

okay..

[honest pond]

eddy todo tuyo

[viscid moss]

idk how to install Applio, too much steps and errors

[honest pond]

no recuerdo como se activa el env

[viscid moss]

nah

[honest pond]

en cmd

[molten pecan]

too many*

[viscid moss]

yo peor

[honest pond]

tiene que activar el entorno

e instalarlo a partir de alli

[viscid moss]

Applio so hard to install

[honest pond]

try doing git pull

type git pull

[molten pecan]

@proper shale @copper pebble

[buoyant saddle]

okay?

[molten pecan]

sabes cuando se usa too many or too much?

[viscid moss]

nop

[honest pond]

you have this app @buoyant saddle ?

[buoyant saddle]

yes

[molten pecan]

too many see usa cuando el objeto en inglés se puede escribir en plural. too much se usa cuando el objeto solo existe sin una palabra en plural

• too many steps
• too much money

too many GPU's
too much Applio

[honest pond]

srry this trash is not working

@brittle wing

[buoyant saddle]

what do I do with this?

[honest pond]

no, normaly it works but they delate de activate scritp

[buoyant saddle]

I will pretend that I understood what you just said

[viscid moss]

https://tenor.com/view/anya-forger-taking-notes-oomfie-spy-x-family-memo-gif-8287618218761733944
tysm

[honest pond]

type conda env list

and sendme sc

[buoyant saddle]

on the anaconda prompt, correct?

[honest pond]

yes

[buoyant saddle]

\

[molten pecan]

I think he meant "no, normally it works, but the activation script is delayed because of it (Applio installer/Windows)"

I think

[honest pond]

ok good

conda activate your path\Applio-RVC-Fork\env

[buoyant saddle]

so I type this in to the anaconda prompt again?

[honest pond]

yes but your path

[buoyant saddle]

got it

[honest pond]

that is showing there

perfec

is seems like this?

[buoyant saddle]

OHH

OK YEA SURE

give me a sec

[honest pond]

now tipe pip install tensorboard

[buoyant saddle]

uhhh

did I Do it correctly?

[honest pond]

put the path " the path"

is because the spaces

[buoyant saddle]

okay?

[honest pond]

like this

te lo dejo a ti @molten pecan ?

[molten pecan]

No tengo ni idea acerca de esto

Apenas se abrir cmd

Yo soy helper por otros motivos, no por ser experta en RVC desafortunadamente xd

[buoyant saddle]

fixed it

[rare goblet]

Ayo? @buoyant saddle level 3 !!!

[buoyant saddle]

I said stop doing that

[honest pond]

ok now type

pip install tensorboard

[buoyant saddle]

what now?

[honest pond]

try to open the go-appilo

[buoyant saddle]

ok!

[honest pond]

if that dont work ping me

[viscid moss]

Raid, que está instalando exactamente?

[honest pond]

el transboard

[viscid moss]

Hmmmmm y es todo ese rollo para hacerlo, interesante

[honest pond]

el env es util pa no tener diferencias de versiones

[buoyant saddle]

uhhh

[honest pond]

ok you steel have the conda open?

[buoyant saddle]

I still have the anaconda prompt open, yes.

[honest pond]

ok type cd .assets\requirements

[buoyant saddle]

with the env thing activated?

[honest pond]

yes but you are not in the appilo path

type first cd and the path of appilo

[buoyant saddle]

may you show an example?

[buoyant saddle]

done

[honest pond]

ok now

.\assets\requeriments

[honest pond]

ok now

pip install -r requirements.txt

you have intel?

[buoyant saddle]

I have ryzen cpu, Nvidia GPU.

uhh
what did I do

[honest pond]

ok try pip install -r requirements-amd.txt

weit ...

[buoyant saddle]

should I press enter?

[honest pond]

amd is a gpu

xd?

pip install -r requirements-applio.txt

try better this

[honest pond]

type ls

[honest pond]

wtf

[buoyant saddle]

uhh
what do I do with this?

[honest pond]

download this and put in the folder

[buoyant saddle]

got it?

[honest pond]

copy them there

[buoyant saddle]

good news: One of them was pasted

[honest pond]

yes

remplace

[buoyant saddle]

got it

just did

[honest pond]

ok try again

[honest pond]

ok now try

[buoyant saddle]

why does this keep happening😭

[honest pond]

hmmmmmmmm

you try to open the go appilo whit option 1 or ???

[buoyant saddle]

option 1

I always do that option

[honest pond]

hmmmmmmm

@brittle wing tu le sabras mas

[buoyant saddle]

why does this happen😭
I should have never updated💀

[honest pond]

yeh

[buoyant saddle]

idk
I had like some rvc thing idk
and it said that "Applio is better" when looking at the new guides at that time, so I upgraded to that. Then, today I tried to use it, and it said "New Update!" or something idk. Got the .bat, and I thought everything was fine, then it said that it didn't work.. and here we are

[rare goblet]

Ayo? @buoyant saddle level 4 !!!

[buoyant saddle]

Arcane I said stop

[finite galleon]

mainline training doesnt have rmvpe or crepe right?

[honest pond]

to see is given a lot of support and if you are in their chat they help you and add things, the bad thing is that they update things and then it stops going.

[buoyant saddle]

I don't even adjust hop length bc Idk what that is

[finite galleon]

I might or may not tell someone to install mangio/applio since mainline doesnt have rmvpe training

[buoyant saddle]

wait does mainline.. well make the ai covers, or just models? BC I need both from it.
(this is a stupid question oml)

thank heavens

[glad zealot]

yup pretty much almost everything ai voice related in this server comes from mainline first

[viscid moss]

Mainline

[buoyant saddle]

ok

finally set up mainline

I hope that I can actually make a cover and not get any errors

FINALLY

IT WORKS

THANK YOU ALL
OMG

[glad zealot]

nice

[rare goblet]

Ayo? @cloud finch level 4 !!!

[flat nacelle]

if i want to continue to train my already existing model, what do i have to do?

[cloud finch]

Why when training did it make so many

[glad zealot]

dont mind it

those are needed to actually train

[flat nacelle]

https://docs.google.com/spreadsheets/d/1owfUtQuLW9ReiIwg6U9UkkDmPOTkuNHf0OKQtWu1iaI/edit#gid=1227575351 is this legit yall?

[vague sage]

Hey

Need help

[glad zealot]

no idea but those should be fine... tried searching my models and its not there tho

[vague sage]

@glad zealot can you help me?

[flat nacelle]

thanks. wondering if you have a good Donald Trump model. Also, if we want to continue training our model after we're done training it once, what do we do? Thanks in advance.

[glad zealot]

to continue training you need to have the generated G and D pth files and the original dataset iirc

[flat nacelle]

can i train an already existing model?

[glad zealot]

nope

as far as i know

[violet heron]

[latent kettle]

-colab

[azure marsh]

☁️ Google Colabs

• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab

🤗 Hugginface Spaces

• Ilaria RVC, by thestingerx Hugginface Spaces
• RVC-HF, by r3gf Huggingface Spaces
• AICoverGen, by r3gf Huggingface Spaces
• RVC v2 Huggingface version, by Clebersla Huggingface Spaces
• Advanced RVC Inference, by neuclya Huggingface Spaces

🎄 More commands: -audio, -local, -overtrain, -realtime, -rvc, -help

[rare goblet]

Ayo? @fossil glen level 15 !!!

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation by country

🇧🇷 Brasil (PT-BR)

https://docs.aihub.wtf/v/brazil

🇪🇸 Spain

https://docs.aihub.wtf/v/spain/

🇫🇷 France

https://docs.aihub.wtf/v/france

🇩🇪 Germany

https://docs.aihub.wtf/v/germany

🇯🇵 Japan

https://docs.aihub.wtf/v/japan

🇰🇷 Korea

https://docs.aihub.wtf/v/korea

🇵🇱 Poland

https://docs.aihub.wtf/v/poland

🇷🇺 Russia

https://docs.aihub.wtf/v/russia

🇮🇹 Italy

https://docs.aihub.wtf/v/italy

[elfin cairn]

yeah i am but its not working even tho i only have one mic

[brittle wing]

https://www.tiktok.com/@vee_vrchat/video/7307443351329049899

This vrchat user has a very convincing ai model
What model do they have?

[rare goblet]

Ayo? @brittle wing level 1 !!!

[brittle wing]

Nice

[jade knoll]

bro idk why whenever i open voicemeeter i cant hear my AI voice and then my ai voice doesnt work and it messes up my audio stuff like it was working before but idk what happened can someone help me please. and i think this problem is because of voicemeeter

i need help

[brittle wing]

you trying to use it for discord?

[jade knoll]

no for fortnite

[brittle wing]

oh okay

[brittle wing]

still need help?

[jade knoll]

yes

[faint bloom]

the download dont work

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation by country

🇧🇷 Brasil (PT-BR)

https://docs.aihub.wtf/v/brazil

🇪🇸 Spain

https://docs.aihub.wtf/v/spain/

🇫🇷 France

https://docs.aihub.wtf/v/france

🇩🇪 Germany

https://docs.aihub.wtf/v/germany

🇯🇵 Japan

https://docs.aihub.wtf/v/japan

🇰🇷 Korea

https://docs.aihub.wtf/v/korea

🇵🇱 Poland

https://docs.aihub.wtf/v/poland

🇷🇺 Russia

https://docs.aihub.wtf/v/russia

🇮🇹 Italy

https://docs.aihub.wtf/v/italy

[green parrot]

when I try to train a model, the PTH file doesnt get saved in the weights folder, anoyone know how to possibly fix that? feel free to @ me

[rare goblet]

Ayo? @green parrot level 1 !!!

[glad zealot]

it takes a while depending on your save epoch count

and dataset length prob

[green parrot]

the strange thing is that it is not generated even after the end of all the procedures...
the dataset only contains 8 files so I don't think that's the problem

[glad zealot]

weird

can you screenshot your setting for training if you can?

[green parrot]

sure, I will be back in a sec

[glad zealot]

oke

[green parrot]

here are my training tab settings

[glad zealot]

what gpu do u have btw

nvm it shows up there

@sand iris halp its local

[green parrot]

gtx 1660 super, should be 6GBs

[glad zealot]

btw is the weights folder in assets>weights?

[green parrot]

Exactly

[neat compass]

is there a way to do this on mac?

[glad zealot]

no idea, i use online

[glad zealot]

ya weird

[neat compass]

whats the website you use

[glad zealot]

kaggle

[neat compass]

can u send link

[sand iris]

How long did it train for ?

[neat compass]

oh wait i c

[green parrot]

I should try to reinstall it completely when I have time and see if the issue persists, maybe trying to read the logs for further error notifications

[green parrot]

about 3 minutes iirc

[glad zealot]

do you know how to use kaggle?

[sand iris]

I think that's to be expected then

training takes a long while

[neat compass]

never used it

[glad zealot]

welp i use kaggle for training

its similar to colab

[neat compass]

doeos it give you a index file and pth

[green parrot]

mhhh so it did not generate the PTH file because of the process not ending ?

[glad zealot]

yes

[neat compass]

do I do click to make new dataset?

[glad zealot]

screenshot of what im training rn

[neat compass]

do I click on make new dataset

[glad zealot]

yes

https://www.kaggle.com/code/hinabl/multinotebook-ai-public heres the one i use

you can just mess around to figure it out

[warm needle]

What can I do to get rid of the Robotic voice after generating the vocals with model? using Mangio RVC

[proud elbow]

batch size 16 could use around 11 GB vram, in order not to have heavy bottleneck with 6 GB vram, should instead go batch size 3

[green parrot]

I think ur right, I read the logs and the program was allocating more memory than the one available, so I lowered to 6 and It now seems to be progressing fine, thanks!

[rare goblet]

Ayo? @green parrot level 2 !!!

[hollow yew]

Yeah i did refreshed

[half cove]

did you duplicate the space

[hollow yew]

Still the same

[half cove]

click the 3 lines

[hollow yew]

How to do that?

[half cove]

click the three lines top right

[hollow yew]

And...?

[wary cedar]

@round fog

[hollow yew]

Holy shit this is complicated

[half cove]

it should say duplicate space with other options

[wary cedar]

yeahhhhh

it aint lookin too great for me haha

[half cove]

let me get on mobile rq

[round fog]

on the top leg t

left

you name ur model

[half cove]

click the 3 dots

not the 3 lines

[wary cedar]

done i put my path down and processed it.

[round fog]

oh you already pressed the button

okay

[wary cedar]

yuh

sample rate 40k

[round fog]

one sec

[wary cedar]

all good

take ur time

[round fog]

i’ll be back in 5

[wary cedar]

oki