✨│ai-help

[inland swift]

I tried that

It didn't help

Also I can't run smth

Let me show what I can't run

[proper shale]

u didn't restart runtime and fell on the same clone repo error

[toxic maple]

the first output with train feature index: (2349, 768),60
training
adding
Successful Index Construction，added_IVF60_Flat_nprobe_1_stinos3_v2.index

INFO:stinos3:loaded pretrained pretrained_v2/f0G40k.pth
<All keys matched successfully>
INFO:stinos3:loaded pretrained pretrained_v2/f0D40k.pth
<All keys matched successfully>

then this in cmd

and then a load of text

about matplotlib loaded modules

[inland swift]

Idk how to restart bruh

[toxic maple]

ImportError: cannot import name 'api' from 'matplotlib' (C:\Users\XXXX\AppData\Roaming\Python\Python39\site-packages\matplotlib_init.py)

[inland swift]

I went to phone to make it easier

[toxic maple]

With this at the end

[proper shale]

yeah no that just makes it worse

never train on phone

[proper shale]

huhh

[inland swift]

But actually how i do restart

[proper shale]

top right, has all the performance stats, click on the drop-down, there should be some options to restart

[inland swift]

I don't understand where this top right thing

[toxic maple]

I manually installed this package with pip, but it still doesn't work.

[rare goblet]

Ayo? @toxic maple level 2 !!!

[proper shale]

I'm afraid I won't be able to solve the issue then

[inland swift]

I'm looking at top right and I don't see anything there

I'm looking at where's my pfp

[proper shale]

The performance thing!!!!

T4

[inland swift]

T4

[proper shale]

There's a dropdown right near that

[inland swift]

Ok hold up

[toxic maple]

?

[inland swift]

Say from the memory which i need to choose

1-8

[inland swift]

Oh ok thanks

Done

What's next

I need to re-do all stuff?

[proper shale]

now run all dependencies step at once

[inland swift]

Alright

[toxic maple]

I've tried installing it via PIP manually, but it didn't work

[proper shale]

@sand iris do ya know this

[inland swift]

Btw

Thanks dor existing

For*

[proper shale]

mb, im not that good at python

[proper shale]

...you're welcome i guess

[inland swift]

By that i mean you help alot

[inland swift]

Useful etc

[proper shale]

thank you

i try my best

[inland swift]

Np you are good understanding person

I mean you didn't gave up on teaching me

And i like the way how u explain the stuff
It's chill

[pure scroll]

you mean i did gave up?!!

[inland swift]

No

You had some other job irl I think

[inland swift]

Your was busy etc i think

I mean no one does sit all day here only to help people

People has own mind and stuff

[pure scroll]

🐐

[inland swift]

?

[pure scroll]

you are the goat!

[inland swift]

Thanks

I'm understanding person so you didn't gave up

Maybe your was doing smth else while your was not in discord

[brittle wing]

Why is this not done yet

[analog python]

might seem like a dumb question but how/where do I turn my mp3 audiofile to pth and index files

[inland swift]

Top reasons why i didn't swear when i was angry:
1: because it's weak thing to do
2: haram
3: always trying to be positive even when i'm angry

[rare goblet]

Ayo? @inland swift level 7 !!!

[fast swan]

Be engineer: swear everyday.

[pure scroll]

what do this translate?

[fast swan]

حسبي الله ونعم الوكيل

[inland swift]

Idk arabic language

[fast swan]

Goodnight

[pure scroll]

you said haram so i thought you might know

[inland swift]

Are you muslim?

[fast swan]

Literally good night

[pure scroll]

really?

[fast swan]

Yes

[inland swift]

I'm just muslim

[fast swan]

Arabian even

[inland swift]

I wanna dm you

Can i dm

[fast swan]

Go ahead

I'm lonely

[molten pecan]

The only way to get index and pth files is through training a model.
Voice models are trained with an audio file

[sand iris]

idk can't check rn

[molten pecan]

Hello gdr

[pure scroll]

julia i got my autoreact

lol

clem

[left oasis]

help please

[rare goblet]

Ayo? @left oasis level 4 !!!

[sand iris]

Hello

[half cove]

W

[half cove]

Should say added index

[left oasis]

Is that okay?

[half cove]

Yea

[left oasis]

oh

[molten pecan]

When you finished training the model, was there a .index file named "added_"?

[left oasis]

thanks

[molten pecan]

Yea that

[toxic maple]

Fired, do you have any idea?

[analog python]

I see, what's the simplest way to do that?

[sand iris]

can you share more logs around the error?

[toxic maple]

File "C:\Users\xxxx\AppData\Roaming\Python\Python39\site-packages\matplotlib\projections_init_.py", line 4, in <module>
from mpl_toolkits.mplot3d import Axes3D
File "C:\Users\xxxx\Documents\RVC\Mangio-RVC-v23.7.0\runtime\lib\site-packages\mpl_toolkits\mplot3d_init_.py", line 1, in <module>
from .axes3d import Axes3D
File "C:\Users\xxxx\Documents\RVC\Mangio-RVC-v23.7.0\runtime\lib\site-packages\mpl_toolkits\mplot3d\axes3d.py", line 22, in <module>
from matplotlib import _api, cbook, _docstring, _preprocess_data
ImportError: cannot import name 'api' from 'matplotlib' (C:\Users\xxxx\AppData\Roaming\Python\Python39\site-packages\matplotlib_init.py)

[worn stag]

Can anyone help me? I've downloaded an AI model, but I don't know how to use it. Do I use it with RVC v2 in a classic way, and do I need to use Yandex Disk to upload it?

[sand iris]

This seems to be a Frankenstein'ed install of Mangio

are you sure you unzipped everything?

[toxic maple]

Yes, everything is unzipped

[sand iris]

why the hell is it looking for the system install of python

when it's using python provided by mangio ?

[toxic maple]

idk

thats why I'm here (:

[analog python]

The Mangio-RVC-Fork don't work, this what it says, any idea how to fix it?

To create a public link, set share=True in launch().
2024-01-02 18:04:21 | ERROR | root | Task exception was never retrieved
future: <Task finished name='evz68h83ifp_17' coro=<Queue.process_events() done, defined at C:\Users\Jani\Desktop\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\queueing.py:343> exception=1 validation error for PredictBody
event_id
Field required [type=missing, input_value={'data': ['48k', True, 'v...on_hash': 'evz68h83ifp'}, input_type=dict]
For further information visit https://errors.pydantic.dev/2.5/v/missing>
Traceback (most recent call last):
File "C:\Users\Jani\Desktop\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\queueing.py", line 347, in process_events
client_awake = await self.gather_event_data(event)
File "C:\Users\Jani\Desktop\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\queueing.py", line 220, in gather_event_data
data, client_awake = await self.get_message(event, timeout=receive_timeout)
File "C:\Users\Jani\Desktop\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\queueing.py", line 456, in get_message
return PredictBody(**data), True
File "C:\Users\Jani\Desktop\Mangio-RVC-v23.7.0\runtime\lib\site-packages\pydantic\main.py", line 164, in init
pydantic_self.pydantic_validator.validate_python(data, self_instance=pydantic_self)
pydantic_core._pydantic_core.ValidationError: 1 validation error for PredictBody
event_id
Field required [type=missing, input_value={'data': ['48k', True, 'v...on_hash': 'evz68h83ifp'}, input_type=dict]
For further information visit https://errors.pydantic.dev/2.5/v/missing

[solar jacinth]

@glad zealot

mr.hina san

mr. hina san

[molten pecan]

Yes, you must upload the model to RVC. I recommend Mangio

[solar jacinth]

Does kaggle mangio tweak can do vocal separtion ?

[worn stag]

Mangio ?

[molten pecan]

It is a fork of RVC

Do you know what that means?

[worn stag]

i don't know

[glad zealot]

AAAA

hi

[solar jacinth]

yayyyy mr. hina san is here

[glad zealot]

yes but goodluck putting the audio files

thats why i used it for training only

[molten pecan]

RVC has quite a few "forks". A fork of RVC is a modified copy of the original software, where they add their own features and modifications

The most popular forks are Mangio, Applio, RVC Studio, AICoverGen, etc

And the base, original and unmodified version of RVC we call it "Mainline RVC"

[worn stag]

like ai tranink?

[glad zealot]

ngl never tried that

but it should work

[solar jacinth]

nope , im trying to separate vocal and instrumentals

[molten pecan]

Do it with UVR

The vocal isolation tool of RVC is a little broken in my experience

[worn stag]

I think I understand, so what is the website address of this "mangio"?

[rare goblet]

Ayo? @worn stag level 1 !!!

[solar jacinth]

my laptop's gonna explode

[solar jacinth]

inference works pretty well

thanks mr . hina san

[solar jacinth]

Really ?

[molten pecan]

Check the links collection post on #1159513888199540817

[sand iris]

[molten pecan]

It doesn't in mine 😭

[glad zealot]

or mdx colab

lemme find it

https://colab.research.google.com/github/jarredou/MVSEP-MDX23-Colab_v2/blob/v2.3/MVSep-MDX23-Colab.ipynb

[solar jacinth]

I will try it

[glad zealot]

note that takes forever for datasets

[solar jacinth]

ohh

is it slower on colab and thats whay it works on colab ?

[glad zealot]

no idea why but it takes longer than the first version

since idk how to set it up properly lol

https://colab.research.google.com/github/NaJeongMo/Colab-for-MDX_B/blob/main/MDX-Net_Colab.ipynb

heres the other one

[solar jacinth]

yrrrrr love you 3000

hina is the best

plus your big fan on github

[glad zealot]

agreed

i used it to do my datasets ngl

[brazen sorrel]

I'm having issues regarding to Ilaria_RVC after duplicating the space, what I only see is blank and nothing

any help is greatly appreciated

[molten pecan]

Did you try reloading the page?

[violet heron]

Set the space to public

It’s because cross site cookies are disabled so private spaces won’t show

So set it to public to solve that issue

[proven hill]

NEW MOMMY VOICE MODEL!!!

Hi cutiepiessss you asked for a better mommy voice model for the voice changer and i listened!!!

Grab it here before it goes: https://ko-fi.com/s/98d841bf59

Also visit my shop for other voices like e-girl and v-tubers ones (free included!!)
https://ko-fi.com/ilariaowo/shop

i love youuuuuu :LC_WhiteCat5:

[brazen sorrel]

Thanks! I'll test it out tomorrow since I'm already in my bed

[rare goblet]

Ayo? @brazen sorrel level 1 !!!

[brazen sorrel]

I already did a couple of times

[fiery bramble]

help, the zip file seems to be fine and all the settings seem normal by the tutorial but it never finished processing the audio, I was 800 seconds in at one point and still it did not stop.

i restarted the gui multiple times and when i entered the same settings the processing never stopped

[brittle wing]

You don't give it a ZIP file locally, just a folder with all of your audio files in it for the model you want to make currently.

[fiery bramble]

oh so you dont zip it?

[brittle wing]

Correct, that's only for the cloud based shit.

[fiery bramble]

ah okay

how long should the processing take?

[brittle wing]

Depends on how many audio files you give it, and the total length of it.

[rare goblet]

Ayo? @brittle wing level 7 !!!

[brittle wing]

More audio files the faster it'll go. Still depends on your hardware too though.

[half cove]

Length matters more than the amount of files

[fiery bramble]

1 audio file, 39 seconds...

[rare goblet]

Ayo? @fiery bramble level 6 !!!

[fiery bramble]

its still processing

700 seconds in

[brittle wing]

The amount of files causes it to thread, otherwise if it's just one it'll take longer than 400 files if both are the exact same length.

[fiery bramble]

i mean i can process all sounds separately but i would have to make a new folder for that

wouldnt take long but that is a possibility

[proud elbow]

since you run it locally, you should input a dataset folder with at least an audio file (wav recommended to have no issues)

[half cove]

"matters more" i never said it didn't matter

[brittle wing]

You literally, have 1 folder that you point at, you don't put a / at the end of it, just ends when the folder name is hit. If you put the / after it'll end up looking at something that doesn't exist.

[fiery bramble]

does .flac cause issues?

i can do wav too but its just im wondering

[brittle wing]

And still no, bc if both are always the same length, and one is just one file, while the other is hundreds, you'll wait 2-3x as long for no reason.

[cold raptor]

so what is ur gpu?

[half cove]

That's one example where more files would matter I'm talking about most cases

[brittle wing]

Unless you are running on a Hard Drive, it really shouldn't take long. Flacs are fine as well.

[proud elbow]

also check the path and the file names not to have spaces and maybe special characters

[rare goblet]

Ayo? @proud elbow level 16 !!!

[fiery bramble]

OH my file folder and file have spaces....

[brittle wing]

Yes in EVERY case, more files = faster.

[half cove]

🤔

[fiery bramble]

ok i have started processing again without spaces

[brittle wing]

Do you want me to make your life easier because I can, instead of having to copy the folder path to your audio files all the time, make a folder inside of the base part of your install, if it's called AUDIO, then all you type is AUDIO in the path to training folder, since it will always start where you launched the bat from. Which, is in the rvc folder.

[cold raptor]

spaces are not a problem If the name was too long or something error would immediately appear

[fiery bramble]

i just do shift + right click on the folder and then copy folder as path, remove the two " " at the end and thats all

[brittle wing]

No one is questioning the fact that blud is running on a 4core CPU. So I will now point that out. It's going to take time.

[fiery bramble]

i dont even know how many cores my cpu has one second

[brittle wing]

its in the UI. unless mangio does something stupid and limits it to half

[fiery bramble]

YUP i have 4 cores...

my pc aint new

[proud elbow]

use 2 cpu processes, not full bar in order not to risk hanging or crashing the system

(esp in feature extraction)

[fiery bramble]

we will see about that but thanks for the suggestion

[covert nacelle]

Can someone go in voice with me and help me to fix my voice changer, idk why, it works like shit

[craggy dagger]

I'm having trouble finding a guide or general information on how to get audio to audio ai to sound better. Like an in depth guide to the settings.

Basically I have all the horsepower in the world and don't need to train a new ai, I just need to learn how to make any ai sound better by learning what to tweak.

Using rvc v2 mangio if that matters.

[proud elbow]

u mean for existing models? they may have limitation depending on their own quality, or try make and train a higher quality one

[brittle wing]

#1159513888199540817 read through the one on mainline, should give you an idea on how to make models the correct way, aka cleaning your dataset etc.

[craggy dagger]

I think I might have two separate issues.

One is sometimes the Model will break down, and in other spots on the same render it's fine. So that might be the model itself? Once I'm home from work I can post tiny clips of the type of artifacts/weirdness I'm getting. Hopefully that can have someone point me to what's going on specifically. Because I'm willing to bet it's user error. I.e. my fault. Lol

[rare goblet]

Ayo? @craggy dagger level 1 !!!

[half cove]

They did say that they don't want to train their own ai

[craggy dagger]

Oh I don't mind training my own. I just haven't needed to as of yet.

[brittle wing]

🧌 information is information, guides channel is still a place to be anyways.

[half cove]

Maybe this but i haven't checked if its good https://discord.com/channels/1159260121998827560/1160164377006194728

[craggy dagger]

Second issue is 30 minute renders where the ai only needs to convert about 20 seconds worth of audio. (Audio book and every track is the same length for sake of my sanity in editing)

[sand iris]

what's the GPU being used ?

[craggy dagger]

4090

[fiery bramble]

it has been 1300 seconds i dont know if my 39 second dataset is supposed to be processed already or my cpu is THAT slow

[sand iris]

I believe it would crash if it failed to initialize it

[fiery bramble]

my cpu is not THAT slow

[civic meteor]

So basically i just got a new pc and finally installed the mangio-crepe 23.7 for training but when i try to train a model i get this can anyone help?

[sand iris]

Does it work fine on shorter inferences ?

[tired scroll]

hi rvc can't find the file "mute" in logs what should i do?

[sand iris]

4GB of vram is unfortunately too little to train on

[civic meteor]

so what should i do

[sand iris]

Use google colab or somehow get another GPU to train with

[civic meteor]

oh ok

thanks

[craggy dagger]

Was this to me or?

[sand iris]

ya

[tired scroll]

how can i send a screenshot ?

[civic meteor]

i mean my pc does not have 4GB tho

[half cove]

Level 1

[craggy dagger]

So it works with long runs as well. I have a 30 minute track of narration that is great the whole way through.

But I'll test the other models that have been giving me issues in shorter audio files without the silence. Maybe that is messing up the Ai.

[sand iris]

the GPU has though

[tired scroll]

Vram is Gpu's ram

[sand iris]

Only thing I can think of is that it doesn't have enough VRAM to buffer the entire audio, so it falls back to system memory and slows down the thing considerably given a certain point

I think they chunk the inference to avoid that though, so it's only a bad guess from me

[fiery bramble]

1600 seconds

[proud elbow]

then try split the input audio

[craggy dagger]

To be clear my workload looks like

All tracks are 30 minutes.

Track 1: Narrator talking
Track 2: character
Track 3 character
Track 4: character

I render all of them separately. The narrator is just fine, but the characters who are mostly silence in the tracks have weird ducking and Artifacts.

[thin slate]

Hii, i'm trying trying to use the google collab and it disconnect:

Runtime disconnected
Your runtime has been disconnected due to executing code that is disallowed in our free of charge tier. Colab subsidizes millions of users and prioritizes interactive programming sessions while disallowing some types of usage as outlined in the FAQ. If you believe this message is in error, file an appeal. Please include any relevant context about your usage.
Your compute unit balance is 0. Purchase more


To connect to a new runtime, click the connect button below.

But i didn't could use more than 1 minute, anyone can help me?

[sand iris]

Oh I thought you were having issues with processing times

[half cove]

If it doesn't have enough vram doesn't it just stop

[tired scroll]

FileNotFoundError: [WinError 2] Sistem belirtilen dosyayı bulamıyor: 'C:\Users\Samet\Desktop\RVC-beta-v2-0618/logs/mute/0_gt_wavs/mute40k.wav'

I think i accidentaly deleted something that i shouldn't...

[rare goblet]

Ayo? @tired scroll level 1 !!!

[sand iris]

some driver versions fallback to system memory somehow

[craggy dagger]

Naw. Artifacts and such, so I want to learn what I need to tweak.

If the shorter audio files with no long periods of silence don't fix it up I'll try training a few models and if the problem persists I'll post about it again.

[sand iris]

Oh my bad, sorry for getting you sidetracked 😅

[craggy dagger]

You're good.

[sand iris]

Which colab notebook were you running?
Some are disallowed by google (any of them that display on a separate web page are)

[craggy dagger]

But as a general rule, if I don't care about render times and just want the best output quality, what should my settings look like?

[thin slate]

What do you mean, which collab notebook?

[sand iris]

colab is the google platform, on it you can run any "notebook" people made

There are a lot of different ones for this tech

[sand iris]

I really don't know much when it comes to quality, hope someone else can help though

[thin slate]

I think T4 GPU, is that?

[rare goblet]

Ayo? @thin slate level 1 !!!

[half cove]

That's the gpu being used

Just send the link of the notebook

[sand iris]

Just share the link of the page you loaded lol

[thin slate]

https://colab.research.google.com/github/w-okada/voice-changer/blob/master/Realtime_Voice_Changer_on_Colab.ipynb

[sand iris]

Oh yeah

[half cove]

Use the hina modified one

[sand iris]

that one's detected & actively blocked by google

[thin slate]

i saw notebook settings and clicked on it 😂

[thin slate]

lol

[sand iris]

Seeing the error message you got, you might need to use another google account

[thin slate]

where is?

[sand iris]

and use the one pinned in #🔍│help-w-okada (hina's modified)

[thin slate]

oohh ok

[sand iris]

It's not blocked... yet, but it really should be lol

abuse it while it still works

[half cove]

https://colab.research.google.com/github/hinabl/voice-changer-colab/blob/master/Hina_Modifid_Realtime_Voice_Changer_on_Colab.ipynb

Do note that WEB UIs are technically still banned on colab

The colab might not work in the future due to Google finally taking it down, and there is a chance that your account might also get banned in the process

So use at your own risk (or use a throwable alt)

[thin slate]

lol ok i will KKKK

[half cove]

Yes i am too lazy that I just copy and pasted

[thin slate]

that's ok

[quartz fog]

Hey Guy's, What's the best Easiest way to make the voice ai thing? since Colab is banned?

[rare goblet]

Ayo? @quartz fog level 1 !!!

[thin slate]

thank you guys

i was trying sooooo much

[sand iris]

-colab

[azure marsh]

Suggestions for @quartz fog

☁️ 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

[sand iris]

all of those aren't banned !

(yet )

[quartz fog]

(lets see how long they stay) xd

Since Colab was like

how long active?

some months

I dunno Tbh

I just randomly saw that collab does not exist anymore

[thin slate]

so what happens if I get banned? I will just can't use colab, or other things too?

[sand iris]

only colab

[thin slate]

ok fine, ty 🙌 😊

[quartz fog]

well how can you get banned lol

[sand iris]

"oh you are running this which is against our TOS"

[sand iris]

Then you get hit with this

[quartz fog]

oh lol

[sand iris]

I guess you can re-extract it from the zip

[tired scroll]

yep, i'm trying that rn

[sand iris]

bump

[thin slate]

it's understandable, bad people could use this and do boring things

But also I just wanted to play a little so I wouldn't mind 😀

[quartz fog]

Is there a Tutorial for Doing it on Colab?

I just use this For my FNF Voices

they kiiiinda actually work

[half cove]

https://discord.com/channels/1159260121998827560/1160926856548200531

[quartz fog]

Ahh Thank you bud!

[toxic maple]

Do you have any solution for this by any chance?

[full falcon]

Guys does it affect the ai if when i run rvc it shows cuda 0 even if i have nvidia gpu do i need to install cuda ?

[quartz fog]

I don't see the drive Folder?

[half cove]

Did you connect to Google drive?

[quartz fog]

it didn'T show me a massege like usual

is that a problem?

[plush compass]

What does this photo during training tell us?

[half cove]

Are you using ai cover gen no ui?

[quartz fog]

uhhh

[rare goblet]

Ayo? @quartz fog level 2 !!!

[quartz fog]

ai cover?

[half cove]

What tutorial are you using

[half cove]

Wrong thing

[quartz fog]

This one https://docs.google.com/document/d/e/2PACX-1vThk7Qo7yCWNVbxOmahl2R8_Jgi6TFuMBUIi-PWre_HIN0lFTq-dr37Rh5iJlGgYb_vFapXMHt2W8Kp/pub

was in the original thing

[half cove]

-overtrain

[azure marsh]

Hey, @plush compass!

👇 Here are some resources to help you identify if your model is overtraining

How do I know if my model is overtaining?

All-In-One Guide on how to make a good model

This guide explains how the D and G files works and much more: https://rentry.org/RVC_making-models

Credits: LUSBERT

Automated Overtraining Detection (AOD)

Will be available soon in #1159513888199540817

Credits: grvyscale

[half cove]

If your trying to use your own audio instead of a YouTube video make sure that you have ran the mount drive option

After that refresh the page

Not the page

The uhh

File Explorer or whatever its called

[quartz fog]

OOhhhhhhhhhhh

that makes sense

xD

thanks buddy!

[sand iris]

if it doesn't complain about missing cuda, it works

[sand iris]

From the top of my head right now, no

Removing matplotlib from the global python install could help I guess

But if anything relies on it, they'll stop working

[full falcon]

but is it better with cuda ?

[sand iris]

Yes

[toxic maple]

I’ll try, thanks

[warm fiber]

for training, does volume matter?

[rare goblet]

Ayo? @warm fiber level 1 !!!

[inland swift]

do someone want help with rvc

i learned how to use it properly

[toxic maple]

It didn't work, but I got this error, could this be the problem?: Note: you can still call torch.view_as_real on the complex output to recover the old return format. (Triggered internally at C:\actions-runner_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen\native\SpectralOps.cpp:867.)
return _VF.stft(input, n_fft, hop_length, win_length, window, # type: ignore[attr-defined]
C:\Users\xxxx\Documents\RVC\Mangio-RVC-v23.7.0\runtime\lib\site-packages\torch\functional.py:641: UserWarning: stft with return_complex=False is deprecated. In a future pytorch release, stft will return complex tensors for all inputs, and return_complex=False will raise an error.

[sand iris]

that shouldn't be an error

[toxic maple]

Have you seen this error before?

[rare goblet]

Ayo? @toxic maple level 3 !!!

[sand iris]

yeah, all the time when training if I remember right

just says "oh the code here does things which will be removed in the future, be careful"

[quartz fog]

well i think the Song output isn't working?

everything is checked

uh

i refrashed everything on the Data thing

still not showin up

any idea what could be the issue @sand iris ?

Becouse the thing is everythings also checked

[sand iris]

idk

[quartz fog]

and i cannot see whats wrong

[vapid jasper]

Anyone know the fix ? (Applio)

  File "C:\Users\Dators\Desktop\Applio-RVC-Fork\infer-web.py", line 26, in <module>
    import fairseq
ModuleNotFoundError: No module named 'fairseq'
Press any key to continue . . .```|

[quartz fog]

i refrash The webside and try again

still not working

wth

ok got this error

what the hack

ive putted in Luigi

ANd not fucking Gradio

who even is that

[brittle wing]

Things not installing properly challenge very easy.

[neat cosmos]

I downloaded Mangio-RVC and am following a video tutorial that uses TensorBoard but when I follow the link to get TensorBoard it appears to be broken. Is there a working link to get TensorBoard currently?

[quartz fog]

but how in the world does that makes sense it gives me A error of a Way different module Name ive installed xD

[brittle wing]

pins

[brittle wing]

It didnt install properly, it can happen

[quartz fog]

Ohh

alright i try again

[brittle wing]

You can manually edit whatever the cell is and add "!pip install gradio" i believe is the command

[quartz fog]

where do i have to put it in?

[rare goblet]

Ayo? @quartz fog level 3 !!!

[junior patio]

-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? @junior patio level 1 !!!

[quartz fog]

@brittle wing ?

[junior patio]

-rvc

[azure marsh]

Documentation

📚 All-In-One English documentation

https://docs.aihub.wtf/

❔ English FAQ

https://docs.aihub.wtf/faq

🎄 Credits: AI HUB FRANCE

✨ More guides

Full AI Voice Model Training Guide (Local)

Link: YouTube
Credits: Christopher Villanueva

Model training with Mainline RVC

Link: Rentry
credits: Raven (ravencutie21)

AICoverGen Colab Guide

Link: Google Docs
Credits: Eddy (Spanish Helper)

Create a model with RVC disconnected (colab)

Link: Google Docs
Credits: Angetyde

How To Make an AI Cover With Ilaria RVC

Link: Rentry
Credits: 👽 Julia (ailen2091)

[brittle wing]

1sec ill open the colab

[quartz fog]

Alright :)

(i'm doin this since 2 hours now just for making that thing into another voice xD)

[brittle wing]

If i'm not mistaken it's just this, you add to the top of dependencies and re run it

[quartz fog]

i don't see that one

"dependencies"

[velvet robin]

I have a question.
On a scale of 1-10, how easy is it to make voice models?

[brittle wing]

what one are you using, bc this is from RVC Disonnected.

[quartz fog]

uhm

https://colab.research.google.com/ that one?

[brittle wing]

#1159513888199540817 and look at the "mainline rvc" guide. gives you more than enough information to figure out how to do it, the hard part is just cleaning your dataset

[quartz fog]

The English Version of the AiCoverGen

[brittle wing]

Oh covergen

[velvet robin]

Okay. That's understandable. I was just curious.

[brittle wing]

no webui?

[quartz fog]

mhm

[quartz fog]

From this the seccond one from the top

[brittle wing]

same thing just on the first most cell

[quartz fog]

Ahhhhhh

[brittle wing]

any time it tells you, that you are missing something, you add the thing its missing

[quartz fog]

alright hold on let me try it

[grand crest]

hello, i last used models many months ago
can you please tell me what is the best colab link for training a model

[junior patio]

Why do the hugging face versions don't show my model in the interface when It's says Download successful?

Wait Is there a way to zip a file in My drive?

[violet heron]

Right click, compress

[violet heron]

Click refresh

Or reload

Depending on which HuggingFace your using

[junior patio]

I'm suing RVC V2

[junior patio]

There is no Compress

[quartz fog]

@brittle wing yo got this error:

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. lida 0.0.10 requires kaleido, which is not installed. tensorflow-probability 0.22.0 requires typing-extensions<4.6.0, but you have typing-extensions 4.9.0 which is incompatible. torchdata 0.7.0 requires torch==2.1.0, but you have torch 2.0.1+cu118 which is incompatible. torchtext 0.16.0 requires torch==2.1.0, but you have torch 2.0.1+cu118 which is incompatible. torchvision 0.16.0+cu121 requires torch==2.1.0, but you have torch 2.0.1+cu118 which is incompatible. Hit:1 https://cloud.r-project.org/bin/linux/ubuntu jammy-cran40/ InRelease Hit:2 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64 InRelease Hit:3 http://security.ubuntu.com/ubuntu jammy-security InRelease Hit:4 http://archive.ubuntu.com/ubuntu jammy InRelease Hit:5 http://archive.ubuntu.com/ubuntu jammy-updates InRelease Hit:6 http://archive.ubuntu.com/ubuntu jammy-backports InRelease Hit:7 https://ppa.launchpadcontent.net/c2d4u.team/c2d4u4.0+/ubuntu jammy InRelease Hit:8 https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy InRelease Hit:9 https://ppa.launchpadcontent.net/graphics-drivers/ppa/ubuntu jammy InRelease Hit:10 https://ppa.launchpadcontent.net/ubuntugis/ppa/ubuntu jammy InRelease Reading package lists... Done Building dependency tree... Done Reading state information... Done 24 packages can be upgraded. Run 'apt list --upgradable' to see them.

[violet heron]

Click refresh

Or reload

Right below the choose a voice

[junior patio]

I know that

But the name I put in not there

Evenn when I search the name

[viscid moss]

Wait, my version of CoverGen isn't working?

-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

[violet heron]

Was there any spaces in the name

[junior patio]

nope

[rare goblet]

Ayo? @junior patio level 2 !!!

[quartz fog]

used the AICoverGen-NoWebUI [English] one @viscid moss

[viscid moss]

ya, that's my version

[violet heron]

On inference

Right below choose a voice or next to it

[quartz fog]

but wy th shows it me that erorr D:

[violet heron]

There is a refresh button

Or reload

[viscid moss]

Checking rn....

[junior patio]

you mean this right?

[viscid moss]

I tried it yesterday and it worked very well

[junior patio]

I refreshed thousand times

[toxic maple]

So this isnt the problem why I cant make custom pth files?

[junior patio]

The name isn't there

[quartz fog]

this is a custom command i tiped in from ain

[junior patio]

There are just 4 other models that I didn't put

[violet heron]

Show a full screenshot

[quartz fog]

becouse before i got a erro that "Gradio" isn'T installed

[viscid moss]

What's exactly the problem?

[violet heron]

I’ll tell you what to click

[viscid moss]

oh, ok

[sand iris]

Takes a lot of time to generate one

[quartz fog]

yup

[quartz fog]

hm

[junior patio]

still the same

the name I used is this

[violet heron]

Did you duplicate the space

[junior patio]

what is even that

[violet heron]

3 dots i the top bar, duplicate this space

[quartz fog]

what am i supposed to do now

[junior patio]

what do you mean by top bar to be exact

[quartz fog]

since it shows me the erorr

[junior patio]

which top bar

[viscid moss]

I'm checking the problem rn. U can use another colab or HF space meanwhile

[violet heron]

On HuggingFace

3 little dots in a button

Click it

[quartz fog]

hell naw i am not using hf space

this sht is needing like 30 mins to load

[junior patio]

oh right

[violet heron]

Oops

[quartz fog]

The easy gui one

[violet heron]

Wrong person

Lol

Talking to @junior patio

[junior patio]

I'm building one

[quartz fog]

(eddy said the same thing)

(he said i should use HF space)

i'm trying this thing now since 2 hours and 30 mins lol

[viscid moss]

[quartz fog]

Just for one thing

xd

[viscid moss]

Installing

Lemme test xd

[quartz fog]

ok

[craggy dagger]

Just fyi, I got home and tried short clips instead of the long silence with the audio clips dispersed and it works great. So yeah, silence really messes the AI up.

[rare goblet]

Ayo? @craggy dagger level 2 !!!

[junior patio]

Bruh I made it private but still it's not showing up

Am I uploading the wrong file format?

Cuz I read somewhere else that it should be a zip file

But I'm uploading it in pth

[violet heron]

The model?

[junior patio]

yep

[violet heron]

Model should be public

Otherwise it won’t work

[junior patio]

you mean the my drive link?

[violet heron]

Yeah

[junior patio]

It is public

[violet heron]

Also it might not work since your using a drive link

When it says Huggingface.co link:

[neat cosmos]

what does it mean when trying to run the training if my output information is like this? 训练结束, 您可查看控制台训练日志或实验文件夹下的train.log ?

[junior patio]

It's normal

[junior patio]

Now there is an error when I use hugging face link

[viscid moss]

It's working 👀

[quartz fog]

wth

i retry it

[neat cosmos]

In this video i found for a tutorial, the guy shows that during the training the terminal should show a lot of INFO: with logs and such. my terminal just gets stuck here

E:\Audio\AI-Voices\Mangio-RVC-v23.7.0\runtime\lib\site-packages\torch\functional.py:641: UserWarning: stft with return_complex=False is deprecated. In a future pytorch release, stft will return
complex tensors for all inputs, and return_complex=False will raise an error.
Note: you can still call torch.view_as_real on the complex output to recover the old return format. (Triggered internally at C:\actions-runner_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen\native\SpectralOps.cpp:867.)
return _VF.stft(input, n_fft, hop_length, win_length, window, # type: ignore[attr-defined]
INFO:torch.nn.parallel.distributed:Reducer buckets have been rebuilt in this iteration.

[rare goblet]

Ayo? @neat cosmos level 1 !!!

[junior patio]

wait my bad

[viscid moss]

How many songs do you want to make?

[junior patio]

But still nothing comes up

[quartz fog]

1 and it's not a song its for a fnf chromatic

[viscid moss]

send me the model link and yt link

I'll do it for u

[quartz fog]

Thanks but i still wanna try it myself again

if its not working than i send you ^^

I just want to be sure if it was just an one time error or not

[quartz fog]

My one is like this

where have you putted in !pip install gradio

[viscid moss]

I don't have that

Gradio isn't nedded

[quartz fog]

for me it is

[rare goblet]

Ayo? @quartz fog level 4 !!!

[quartz fog]

i get everytime a error that "Gradio isn't installed"

when i do the File thing

[sand iris]

what's your GPU ?

[quartz fog]

The Generate Cover one

[neat cosmos]

its a 1070

[quartz fog]

With the Content drive ect.

[viscid moss]

Try this:
Go to this part >>
Environment or something maybe

[sand iris]

How long have you waited rn ? Also how much time worth of audio files did you give it as dataset ?

[quartz fog]

I cannot speak Spanish

[viscid moss]

and select Disconnect and delete time or smth like that

[quartz fog]

idk where that is lol

[sand iris]

runtime > disconnect and delete runtime

yeah

[viscid moss]

ty

[quartz fog]

i just see that

[neat cosmos]

8:27 of audio

[quartz fog]

ok i try it again

[lunar iron]

Hi! I just found out that RVC GUI is outdated, and now I'm not sure which one I should download lol
which fork would be the easiest to use that doesn't take up too much space?

[neat cosmos]

anyone know what this error means?
RuntimeError: The expanded size of the tensor (1025) must match the existing size (513) at non-singleton dimension 0. Target sizes: [1025, 368]. Tensor sizes: [513, 368]

[viscid moss]

ok lmk

[violet heron]

Original RVC

[lunar iron]

Thank you!

[violet heron]

-local

[azure marsh]

All Working Local Forks 🖥️

• Applio-RVC-Fork, by IA Hispano GitHub
• Mangio-RVC-Fork, by Mangio621 Huggingface
• RVC Studio, by SayanoAI Huggingface
• AICoverGen, by SociallyIneptWeeb GitHub
• Replay, by Replay Team Website
• Original RVC, by the RVC-Project team GitHub

Need some help? 🤔

You can find more info on the #1159513888199540817 channel. If you can't find your answer, feel free to ask for help in #✨│ai-help. Credits to Faze Masta and Antasma for compiling these links.

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

[neat cosmos]

Im continuously getting this error when trying to do the training. any help would be greatly appreciated. 🙂 #✨│ai-help message

[unreal nest]

Hello is anyone here so i want to add my own voice to the select model maker thing how do i do that

[violet heron]

Can you clarify?

[unreal nest]

i cant really send an image

lemme dm you]

[unreal nest]

so did you see it

[rare goblet]

Ayo? @unreal nest level 1 !!!

[unreal nest]

white one did i use then

[meager silo]

-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

[quartz fog]

Damn it needs really god damn long @viscid moss

The uh Clone and install thing

[viscid moss]

ya

15 minutes exactly

But it's not my fault

Is that Colab sh1t

[quartz fog]

Worked now :O

[half cove]

[viscid moss]

https://tenor.com/view/imagination-spongebob-squarepants-dreams-magic-gif-12725683

[full falcon]

Can someone help me install cuda cus it always tells me i don't have microsoft visual studio , is it mandatory to have it ?

[quick jungle]

https://cdn.discordapp.com/attachments/1160576561636966491/1165372018414342284/warning.gif

[rare goblet]

Ayo? @full falcon level 1 !!!

[unreal nest]

can someone help me boot up applio

anyone?

[brittle wing]

What does applio even offer that makes it worth using over the official rvc.

[half cove]

Some people are too lazy to do stuff like tts separately

[wise moss]

my laptop blue screened while doing the feature extraction and it’s at 100% but won’t restart

[rare goblet]

Ayo? @wise moss level 1 !!!

[violet heron]

TTS, UVR built in, some other stuff

Green UI

[wise moss]

I have a nvidia GeForce rtx 3050 I’m not sure what I did 😭

It hasn’t changed though after like 5 minutes so I’m a little concerned

[half cove]

Uvrs built into mainline tho

[violet heron]

TTS and green ui

[half cove]

Ah yes

Green ui

Thr pinnacle of rvc

[lunar iron]

Hi, I seem to have a problem unzipping Original RVC, I keep getting an error that it cannot be extracted and cancels everything. Is there someone who knows on how to fix it?

[rare goblet]

Ayo? @lunar iron level 1 !!!

[violet heron]

Use 7zip

Or a different program

[craggy sonnet]

I’m new to this but I’m using Ilaria RVC on Chromebook and getting “runtime error” I’m confused it loads on my phone but I can’t upload any files but it wont even load on Chromebook

[rare goblet]

Ayo? @craggy sonnet level 1 !!!

[neat cosmos]

#✨│ai-help message Still hoping for some help with this error. Thanks!

[fiery bramble]

i have no idea why this isnt loading...

i saved the audio as a .wav in 48k sample rate

and replaced the old audio in the folder

[neat cosmos]

change speaker id to 0

[fiery bramble]

same issue

[proper shale]

maybe something went wrong? check ur logs

[neat cosmos]

My logs say this
RuntimeError: The expanded size of the tensor (1025) must match the existing size (513) at non-singleton dimension 0. Target sizes: [1025, 368]. Tensor sizes: [513, 368]

[proper shale]

oh that's

yeah honestly I'm unsure how you'd solve that

[neat cosmos]

when i change the sample rate before training and everything, sometimes it'll switch the numbers like so

32k -
RuntimeError: The expanded size of the tensor (513) must match the existing size (1025) at non-singleton dimension 0. Target sizes: [513, 368]. Tensor sizes: [1025, 368]

48k -
RuntimeError: The expanded size of the tensor (1025) must match the existing size (513) at non-singleton dimension 0. Target sizes: [1025, 368]. Tensor sizes: [513, 368]

40k -
RuntimeError: The expanded size of the tensor (1025) must match the existing size (513) at non-singleton dimension 0. Target sizes: [1025, 368]. Tensor sizes: [513, 368]

[rare goblet]

Ayo? @neat cosmos level 2 !!!

[neat cosmos]

for some reason 32k comes out flipped from the others

[grand crest]

hello, how do i package the .pth files into a model zip? do i just put them in a zip?

[pure scroll]

create a zip file that contains .pth and added.....index files

[grand crest]

i did that but the replay app says that it couldnt find the pth files in the zip

[rare goblet]

Ayo? @grand crest level 1 !!!

[grand crest]

i put them as model.pth and model.index
in model.zip

[proper shale]

did u use the G/D pths

[grand crest]

i saw in other zips that there was only one pth so i put the bigger one of the two, D pth

should i put both?

[proper shale]

nope

wrong pths

G/D are for training

the one that u can use is the one with your model's name on it

it's on the weights folder

[grand crest]

ohhh

thank you

[full falcon]

guys can anyone tell me how to get cuda to work for rvc i tried installing visual studio and cuda and pyton and anaconda and conda

[fiery bramble]

same problem

i have no idea what could cause the problem help

1700 seconds in and nothing

[pearl loom]

I am trying to add the link for the voice model from the voice model channel, any idea what I may be doing wrong

[rare goblet]

Ayo? @pearl loom level 2 !!!

[pearl loom]

I am in a duplicated tab

[proper shale]

try accessing that link to see if it works

[pearl loom]

When I put that into the search bar it did download a model winRAR Zip archive file

[pure scroll]

strange

[proper shale]

honestly i think this is an issue with how the model maker posted this model

they had the name with spaces

[pearl loom]

I apologize I am very new

[proper shale]

It's ok, :)

[pure scroll]

can you screenshot, what's inside

[pure scroll]

yup, try making a zip file with just, the 2 .pth files and just added__.index

if 2 .pth zip file gives error, try with 1.pth file and added___.index

the trained___.index file is of no use

[pearl loom]

So just delete the trained one

[pure scroll]

yes.

[pearl loom]

then those 3 in a new .zip

[pure scroll]

yup try now

[fiery bramble]

help it says that Unfortunately, there is no compatible GPU available to support your training. on my laptop. does this mean im gonna have to do the google colab stuff?

[hybrid spire]

I'm noticing on weights gg that there's a msg about "copyright claims" & removing the audio sample

Am I doing something wrong, or is it simply bc the samples posted on here don't transfer to the other website?

[violet heron]

Don’t transfer

[fiery bramble]

LMAO i got a chinese response which says "After the training is completed, you can view the console training log or train.log under the experiment folder" when i pressed the train model button

[pure scroll]

the local training is for good Nvidia gpu's

sadly you have to use colab

[fiery bramble]

problem is, i have a decent gpu on my pc but on my laptop the commands seem to be working...

[pure scroll]

use pc then

[fiery bramble]

ye but the processing data never stops

it just loads and loads and loads

im sticking with this because it somewhat works and i get the issue

[pure scroll]

what's the gpu?

[fiery bramble]

nvidia geforce gtx 1070

its okay but uh

i cant shove it into my laptop

soo i have to use google colab

i have no idea why on my pc it just loads and loads non stop

and i get why nobody can help cuz uh thats an odd error

theres just no way my cpu is that slow

[fiery bramble]

im getting somewhere

[pearl loom]

I must still be doing something wrong, same error

[pure scroll]

no brother

[rare goblet]

Ayo? @pure scroll level 18 !!!

[pure scroll]

.

this is the error @pearl loom

[pearl loom]

I had used it on a colab yesterday, but was informed that would only work a few times. Trying to use Ilaria_RVC now

[pure scroll]

try using this

https://www.tryreplay.io/

it also create ai covers

[pearl loom]

This one I can just download then?

[pure scroll]

yup its an app

you can port the voice models into the app

and use them to create ai covers

if you don't want to install any app

use this google colab

https://docs.google.com/document/d/e/2PACX-1vThk7Qo7yCWNVbxOmahl2R8_Jgi6TFuMBUIi-PWre_HIN0lFTq-dr37Rh5iJlGgYb_vFapXMHt2W8Kp/pub

[pearl loom]

Thanks a ton!

Downloading Replay (hopefully my computer doesn't get sick lol jk jk)

[topaz peak]

Is there a newer guide on just text-to-speech?

[fiery bramble]

man im gonna need help on this colab shit tommorow probably in like a vc or something

its starting to get confusing

i figured out the google drive links but now i need to understand how to actually save the models and train them

🙃

[topaz peak]

Anyone

[brittle wing]

I was using the software normally for a month, but now the voice no longer comes out in discord or other applications, I already checked and in the application my voice is "changing", but in games or discord it doesn't come out, I'm using the "cable "software, anyone have any idea what is?

[rare goblet]

Ayo? @brittle wing level 1 !!!

[pure scroll]

https://huggingface.co/spaces/TheStinger/Ilaria_TTS

[pure scroll]

you talking about w-okada voice changer?

if yes

here do this

[brittle wing]

I'll take a look! ty

[brittle wing]

yes ... its is, but I dont know why my voice not on discord more I'm using this on the software.
I already tried install again too but still the same

[pure scroll]

for the upper half, do these settings

[molten pecan]

wait where did you get this from?

what gitbook page is that?

[pure scroll]

idk i was thinking of making a guide once, so have everything downloaded

[molten pecan]

check dms

[brittle wing]

alright, I did but still the same like, my voice dont goes out in discord, understand?
If I put on INPUT headset I hear my voice, but with Cable for some reason are not going out in discord/games more

[pure scroll]

INPUT - Your Microphone
OUTPUT - Virtual Audio Cable
MONITOR - Headphones if you want to see if working

[brittle wing]

yes, I understand, I was using it until yesterday and it was working, but now my voice "with the software" doesn't come out

[pure scroll]

is it happening just for discord or everywhere?

[brittle wing]

everywhere

[brittle wing]

I just installed adobe photoshop today, Idk if that interfer on something, but is the just "one thing" that I did different

[pure scroll]

i don't think photoshop would have done changes

[half cove]

output is your headphones

and input is the vac

you should put your mic as input

then your virtual audio cable as output

then on discord put your virtual audio cable as input

[cold raptor]

no

output is virtual cable and input mic

[half cove]

No

I was saying what they had it as

[cold raptor]

no

output virtual cable and input mic

[half cove]

Ehem

[half cove]

Ehem

[cold raptor]

https://media.discordapp.net/attachments/844467597855817728/1178482022042308719/1.gif?ex=65764de9&is=6563d8e9&hm=611324b2b692974944712a732ecd65d135484ce67802bf5031ae4854aa910418&

Ehem

[half cove]

Fek u

[cold raptor]

[topaz peak]

Why is it taking so long to build after copying the space?

[rare goblet]

Ayo? @topaz peak level 1 !!!

[topaz peak]

nvm

[pure scroll]

(ilaria is the person that created the space)

[topaz peak]

So i shouldn't have copied the space?

Was following the guide

[pure scroll]

duplicating the space will let you first in the queue

so don't worry, you did it correctly

[burnt saffron]

-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? @burnt saffron level 2 !!!

[violet heron]

Banned.

[stiff current]

when i try to extract the file my file app just crashes

[hybrid spire]

I forgot to bring this up earlier.

I recall with Uberduck (& I think maybe Fakeyou) that they had like blacklists of specific VA names to avoid in re-making.

Is there anything like that here? If so, I'll remove any of mine that coincide with such names ASAP

[violet heron]

Yeah if a artist requests not to have their voice on here ai hub will take it down

[hybrid spire]

I see, so is there a list for me to refer to, just to see if I have to remove any at this moment?

[violet heron]

No list

[hybrid spire]

So until any artists say otherwise, I can just keep my current submissions as is, then?

[violet heron]

Yeah

So you can upload them all unless they go out and say “ay I don’t want my voice cloned”

[hybrid spire]

Cool, thanks for clearing all that up

Btw I just submitted my 15th model now, so am I expecting to get the Model Master status afterwards?

[violet heron]

If your models meet standards if so you do /jointeam

[glass oar]

-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? @glass oar level 2 !!!

[toxic maple]

Yeah I know, but it stops after 70-100 seconds with that matplotlib error.

[sand iris]

Huh so it's back ?

[toxic maple]

yeah

[sand iris]

You removed it from the global python install right ?

[toxic maple]

yes, I uninstalled it via cmd

[sand iris]

what does python --version display in cmd ?

[toxic maple]

Python 3.9.13

[sand iris]

And the matplotlib error is the exact same ?

[toxic maple]

Yes

[sand iris]

can you re-share the new one ?

I'd be super confused if it's still looking for matplotlib on the system side

[toxic maple]

I'll show you a bit more of the logs

[hybrid spire]

I just completed the application questions on the /jointeam command.

Am I anticipating a direct message for the result, or does that show up in announcements?

And does the evaluation tend to take awhile?

[violet heron]

Shows in DMs

[toxic maple]

Process Process-1:
Traceback (most recent call last):
File "multiprocessing\process.py", line 315, in bootstrap
File "multiprocessing\process.py", line 108, in run
File "C:\Users\xxxx\Documents\RVC\Mangio-RVC-v23.7.0\train_nsf_sim_cache_sid_load_pretrain.py", line 225, in run
train_and_evaluate(
File "C:\Users\xxxx\Documents\RVC\Mangio-RVC-v23.7.0\train_nsf_sim_cache_sid_load_pretrain.py", line 507, in train_and_evaluate
"slice/mel_org": utils.plot_spectrogram_to_numpy(
File "C:\Users\xxxx\Documents\RVC\Mangio-RVC-v23.7.0\train\utils.py", line 220, in plot_spectrogram_to_numpy
import matplotlib.pylab as plt
File "C:\Users\xxxx\AppData\Roaming\Python\Python39\site-packages\matplotlib\pylab.py", line 35, in <module>
from matplotlib import cbook, mlab, pyplot as plt
File "C:\Users\xxxx\AppData\Roaming\Python\Python39\site-packages\matplotlib\pyplot.py", line 43, in <module>
from matplotlib.figure import Figure, figaspect
File "C:\Users\xxxx\AppData\Roaming\Python\Python39\site-packages\matplotlib\figure.py", line 18, in <module>
from matplotlib import docstring, projections
File "C:\Users\xxxx\AppData\Roaming\Python\Python39\site-packages\matplotlib\projections_init.py", line 4, in <module>
from mpl_toolkits.mplot3d import Axes3D
File "C:\Users\xxxx\Documents\RVC\Mangio-RVC-v23.7.0\runtime\lib\site-packages\mpl_toolkits\mplot3d_init_.py", line 1, in <module>
from .axes3d import Axes3D
File "C:\Users\xxxx\Documents\RVC\Mangio-RVC-v23.7.0\runtime\lib\site-packages\mpl_toolkits\mplot3d\axes3d.py", line 22, in <module>
from matplotlib import _api, cbook, _docstring, _preprocess_data
ImportError: cannot import name 'api' from 'matplotlib' (C:\Users\xxxx\AppData\Roaming\Python\Python39\site-packages\matplotlib_init.py)

[sand iris]

[toxic maple]

These are the sentences before the error

with at the end the actual error

[sand iris]

why the the hell is it using the system matplotlib

[toxic maple]

should I uninstall python?

[sand iris]

Do you use it for anything else ?

[toxic maple]

at this moment I'm not using it

But I can reinstall it when I need it again

[sand iris]

At this point it might be worth trying

But I really don't understand why it's still grabbing the globally installed matplotlib when you've uninstalled it

or why it's even prioritizing that in the first place

[proud elbow]

no need tbh, though you should never do global install (not within the venv)

[toxic maple]

Same error with mathplotlib

[toxic maple]

So you think its the installation part?

[sand iris]

that's the thing: it's a runtime install using global packages

[toxic maple]

and do you know how I can ensure that I install it correctly?

[proud elbow]

what I know is the original RVC can run well without global shits installed as listed in pip list (not within venv/runtime environment)

[sand iris]

Yeah, yet apparently this one install would rather prioritize globally installed packages instead of the local ones

somehow

and since the global matplotlib is too new for what the runtime packages expect, it just errors out

[proud elbow]

the Mangio 23.7.0 I have also runs without issues despite having python 3.9, 3.10, 3.11 installed

[sand iris]

[plush compass]

Are there any posts for these options? Or can anyone explain what these options are?

[glass oar]

Can I ask why does the output volume is way lower than thhe original volume? I almost can't hear it.

[untold ridge]

how do i install RVC in the first place.

its confusing to me

[plush compass]

Is there a reason to default the index setting to 0.75 instead of 1? If I set it to 1, will the mechanical sound become worse?

[toxic maple]

Yes that's right. But that's why it's strange that it don't work and why I have no idea how to fix this.

[proud elbow]

try delete the old (don't forget to keep your existing models) and redownload Mangio

[toxic maple]

I'll try that, thx

[proven hill]

HIGH QUALITY FEMALE VOICE MODELS

Are you searching girl voices like e-girl, mommy, v-tuber and more?
You find them here: https://ko-fi.com/ilariaowo/shop

And for paid items, if you DM me, you will get a 20% discount if you pay with

I also accept Steam gift cards as last resort

Grab yours today!!!

[toxic maple]

Same thing, still this matplotlib error

[proud elbow]

reinstall python too I guess, and make sure it's 3.9.8

[toxic maple]

Also the same result

Again this matplotlib error

[pure scroll]

voice changer?

w-okada?

[forest sluice]

What's the part you feel confused?

[brittle wing]

My voice can’t be transferred to Discord

[sand iris]

I'll see once I'm back at my desk

this shit's weird

removing %Appdata%\..\Roaming\Python\Python39

Could potentially help

...But again I'm assuming you don't have anything else that requires python lol

[sand iris]

index rate is how much it tries to copy the original dataset

If you set it to 1 it should stick 1:1 to the original voice (which isn't good in case you need sounds that it doesn't know about)

[forest sluice]

Now I think one should make a system restore point in cases like these

[sand iris]

¯_(ツ)_/¯

Somehow uninstalling python 3.9 didn't get rid of the site packages

Or was it 3.9 that got removed

@toxic maple did you ever attempt to install "Applio" ? Also when you removed python, did you reinstall it afterwards?

[toxic maple]

No, I didn't try. When I removed python I tested if Mangio did work, and then I installed 3.9.8 and that didn't work either.

[rare goblet]

Ayo? @toxic maple level 4 !!!

[sand iris]

could it be that you had multiple python versions installed ?

so when you uninstalled one, the problematic one remained ?

[toxic maple]

Where can I check that, cause I think that could be the solution

In the "Control Panel" there are no other python programs active, but when I check python --version, python 3.9.13 is active

I uninstalled python from the windows settings -> Apps menu

I think that was the python 3.9.13 version

[toxic maple]

With that other python version uninstalled, still the same error: ImportError: cannot import name 'api' from 'matplotlib' (C:\Users\xxxx\AppData\Roaming\Python\Python39\site-packages\matplotlib_init.py)

[sand iris]

AAAAA

There is no other python version on the system ?

[toxic maple]

With the output information: 训练结束, 您可查看控制台训练日志或实验文件夹下的train.log

[sand iris]

Because yet again it's still complaining about something on the global python install

[toxic maple]

I don't know where to check this

cause I already checked control panel and windows settings

[sand iris]

python --version in cmd once more

just for good measure, it should tell you python is not installed or something along those lines

[toxic maple]

python isnt found

[prime harness]

help me guy why i cant upload this yuumi model

[toxic maple]

So thats good I think

[sand iris]

Does this folder exist ?

C\Users\xxxx\AppData\Roaming\Python\Python39\

[pure scroll]

replay

[toxic maple]

I can still see that folder

[prime harness]

how

[toxic maple]

Should I delete that folder?

[prime harness]

is this zip have problem

i just want a yuumi voice to play LOL game

[sand iris]

You can, just make sure there's nothing python related that uses it

Else you'll have to re-get the packages and stuff

[toxic maple]

I don't think so, all the programs I made with python I turned off.

[sand iris]

I still don't know why it's looking there instead of the Mangio folder though

[toxic maple]

wiat

[plush compass]

thx

[toxic maple]

When I want to delete the map it said: Map is used. So when I close Mangio, I could delete python

so its connected with eachother

[sand iris]

Yeah it's looking for stuff there somehow

if it's Mangio it's complaining about it's fine

[toxic maple]

I'm currently trying to make a pth file to test if it works

[silk aurora]

Hi guys, i started training a model on mangio and my cmd is doing this. Is this supposed to happen.

[toxic maple]

Its currently showing the epochs, so I think its working because all of this didnt show up before the uninstallation of python

[sand iris]

Seems like it's working now

Sorry this took a while lol

[toxic maple]

doesnt matter, I sincerely want to thank you very much for your help

Otherwise this still wouldn't work

[forest sluice]

You should upload to HF

https://rentry.co/fdg_guide_newer

[brittle wing]

How install rvc

[proper shale]

#1159513888199540817

-local here's some install links for forks too

[azure marsh]

All Working Local Forks 🖥️

• Applio-RVC-Fork, by IA Hispano GitHub
• Mangio-RVC-Fork, by Mangio621 Huggingface
• RVC Studio, by SayanoAI Huggingface
• AICoverGen, by SociallyIneptWeeb GitHub
• Replay, by Replay Team Website
• Original RVC, by the RVC-Project team GitHub

Need some help? 🤔

You can find more info on the #1159513888199540817 channel. If you can't find your answer, feel free to ask for help in #✨│ai-help. Credits to Faze Masta and Antasma for compiling these links.

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

[forest sluice]

Should have auto-reply

[blazing garnet]

I tried to continue training but it is running from start

@proper shale

[proper shale]

ur d/g files may not have saved properly

or u inserted different settings

[blazing garnet]

Rip

[blazing garnet]

I am getting this error

Wait so do I put step count 2333333?

@proper shale tell fast I need to train thissss

[proper shale]

yes

[blazing garnet]

Bruh why didn't they tell that in the guide

[proper shale]

The Colab says it tho

[blazing garnet]

Thx

[proper shale]

You're welcome

[blazing garnet]

Bruh i didn't understand what it was sayin

[blazing garnet]

Do you know if you leave and auto clicker on in the colab it can run for more than the average time

[proper shale]

I'm aware

But I don't like to risk it ngl

[hybrid spire]

So I got the msg saying they're not gonna give me a Model Master status.

Is it possible I could ask what could be the reason? Like if it had something to do w/ the model quality?

[blazing garnet]

Can I use the colab on phone?

Why did ur name change

Is the accopolipse coming

[proper shale]

Yeah, it's slightly clumsier tho

I use it only for like

Inference

Wouldn't train on it

[proper shale]

I got promoted

[blazing garnet]

Yeah I jus wanna convert

[blazing garnet]

W

[proper shale]

Oh yea it's alright then

-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

[blazing garnet]

Why didn't it save the 100 epochs one?

Btw what kinda GPU does google have it's literally running so many colabs

[proper shale]

oh cuz the latest was probably the 150bone

150 one*

[proper shale]

Tesla T4

[cold raptor]

Last one was 150

That's why

[blazing garnet]

K

[blazing garnet]

How do I upload my model on the ai hub? I am confused

And why does the ai hub only consist of voice changing lmao

[proper shale]

#1159514067187277865, qcbot

[blazing garnet]

Wait can I use hugging face like google colab but not get disconnected

[proper shale]

welp there's other interests

[proper shale]

yeah but not for training

[blazing garnet]

Do I have to upload it to hugging face then say the link?

What is demo and link in it???

[proper shale]

Yeah

[proper shale]

demo is a sample of ur model

[ionic delta]

any guides on how to

[dusty jewel]

I dont know in which channel to ask - how do I install a virtual audio cable?

[marsh jay]

hey the voice changer wont pass though my mic

ive installed the audio cable

[proper shale]

basically just click on "save to onnx" and reupload model with the onnx file

should be easy

[proper shale]

The input output settings should be like this in W-Okada:

Input - Your mic
Output - CABLE Input

And in another program:

Input - CABLE Output
Output: Your headphones

[marsh jay]

yea ive tried that

still wont work

[pure scroll]

try increasing in and out

at GAIN

[marsh jay]

yea it still wont work

[rare goblet]

Ayo? @marsh jay level 2 !!!

[pure scroll]

restart everything, pc too?

[proper shale]

change ur f0 to rvmpe too

[blazing garnet]

What

Like what demo?

[marsh jay]

this is my discord too

and discord is not picking it up

[proper shale]

A little sample of your model

An audio

[tribal sorrel]

Uh I Have A Problem with downloading RVC

[pure scroll]

what?

problem

[tribal sorrel]

So when I download it and start the program it just stays on the command prompt forever

no window defender thing pops up

so im confused

[pure scroll]

you are talking about w-okada right?

[tribal sorrel]

oh yeah

srry

[rare goblet]

Ayo? @tribal sorrel level 1 !!!

[pure scroll]

send a screenshot

[tribal sorrel]

alr

[marsh jay]

alright i just restarted

and it still dosent work

[prime harness]

can someone make me a yuumi voice for vocie ai. my head is about to go bald when i figure out how to create a yuumi voice