✨│ai-help

[high night]

-RVC

[brittle wing]

r..

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation to other languages

🇧🇷 Brasil (PT-BR)

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

🇫🇷 France

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

🇩🇪 Germany

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

🇮🇹 Italy

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

🇯🇵 Japan

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

🇵🇱 Poland

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

[naive fossil]

yh

[brittle wing]

oh

[strange kestrel]

lemme guess

[brittle wing]

https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/releases/tag/updated1006v2

pretty sure that's the latest one

[naive fossil]

love u thanks

[strange kestrel]

lmao

[brittle wing]

[strange kestrel]

anyway

how dare you just get here

your name is checkmate

this term is as old as the game

and the game is older than the water i'm drinking right now

[brittle wing]

now why the hell is my thing silent- the sound file (i tried using both wav and mp3)
i've exported using every sort of algorithm
it's dead silent

[high night]

are you a youtuber

[brittle wing]

yes kled has a kledillion subscribers

[strange kestrel]

technically yes

[high night]

u just joined discord

YOUR AN ALT

[brittle wing]

perhaps

[strange kestrel]

who are you joojoba

[brittle wing]

no i just forgot my login info for my other account that's in this server

[strange kestrel]

nah you didnt

[high night]

ALT OF @brittle wing IS KLED

[brittle wing]

true

[strange kestrel]

you're a liar, a cheat, a devil, a whore

[brittle wing]

no

[strange kestrel]

HOLY SHIT HE KNOWS

wait, you're not me?

[brittle wing]

i have gained sentience

[strange kestrel]

AI TRYING TO TRAIN AI

[brittle wing]

and now my purpose is to find out WHY TF IT*S SILENT

[strange kestrel]

IT'S

IT'S THE TERMINATOR

[brittle wing]

misinput

[high night]

tf u mean by silent lol

[brittle wing]

one sec i will screnshot

[high night]

what

u will screenshot audio?

[strange kestrel]

ALRIGHT CHECKMATE

[brittle wing]

THIS

excported audio

is completely dead silent

[strange kestrel]

mf just screenshotted audio

[brittle wing]

yes

i'm beyond logic

[high night]

bro

i cannot hear it

sorry

[brittle wing]

exactly

neither can i

[high night]

my ears are deaf

[strange kestrel]

wrigjaeoirgerigeirge

[brittle wing]

because its SILENT

[high night]

EITHER THE MODEL IS THE PROBLEM

OR THE DATASET

[brittle wing]

..I TRIED ALL

[high night]

ON WHICH IT IS TRAINED

[brittle wing]

I TRIED PM I TRIED CREPE I TRIED THIS I TRIED THAT

[high night]

TREY DIFFERENT MODEL

[brittle wing]

I TRIED OTHER MODELS

[strange kestrel]

OK THEN

[high night]

AND SEE

[strange kestrel]

SIR

[high night]

TRY DIFFERENT MODEL

[brittle wing]

I TRIED LIKE SHASHUMGA200 OR WHATEVER

[strange kestrel]

TRY UMM

[high night]

FROM #1159289700490694666

[strange kestrel]

did you try restarting it

like

[high night]

THEN YOUR SPEAKER IS TRASH

[strange kestrel]

replug it

[brittle wing]

THESE

THE XI JINPING ONES

I TRIED THOSE THAT DOESNT WORK

[strange kestrel]

bro

no way

[brittle wing]

I AHVE HEADPONHES

[brittle wing]

look i've installed htis rvc shit 4 times now okay i don't wanna redownload the models and insert them so i use the preinstalled ones

[high night]

https://discord.com/channels/1159260121998827560/1166264350298869820

[rare goblet]

@brittle wing has reached level 2.

[high night]

USE THIS https://discord.com/channels/1159260121998827560/1166264350298869820

AND LMK

[strange kestrel]

hmmmmm

checkmate

[brittle wing]

OKAY

ONE SECOND

[high night]

WHAT

[brittle wing]

i win

[high night]

IMMA GO PLAY CRICKET

[brittle wing]

okay.. installing...

[high night]

SEE U ALTS LATER

[brittle wing]

touch that grass

[brittle wing]

harassment

[high night]

I DO EVERYDAY

[brittle wing]

Good

[high night]

SPORTS IS MY LIFE

[brittle wing]

OKAY

ILL TRY

IF THIS WORKS

i'm oleaving the seerver and

crinyg

[strange kestrel]

checkmate

so

[brittle wing]

NO

[strange kestrel]

i have a bunch of pth files in the weights folder

[brittle wing]

ITS QUIET

ITS STILL QUIET

[strange kestrel]

what happens if i use the latest one

[brittle wing]

OY ULIES

OW WHAT F UHTE FICK

HOLY SHIT

im deaf now

[high night]

@strange kestrel I WILL REPLY TO YOU LATER I AM GOING TO PLAY CRICKET RN

[brittle wing]

in the middle of the audio there was this loud ass beep

[strange kestrel]

wait

WAIT

[high night]

It's the best

Latest is best

[strange kestrel]

ISNT A PTH FILE SUPPOSED TO HAVE ONLY UMMM

LIKE

[brittle wing]

you lie checkmate

[high night]

@brittle wing your speaker is trash bro

[brittle wing]

i have EHADPOHNES

OKAY FINE

ILL DO SPEAKER

[strange kestrel]

1:06pm is when the training finished

[high night]

Top most

[strange kestrel]

man

[high night]

@brittle wing try something else

[brittle wing]

If my speakers dont work checkmate

[strange kestrel]

you can marry my sister anytime you want

[high night]

BRO LEMME GO PLAY CRICKET

[strange kestrel]

no go play cricket

[brittle wing]

your cricket golf club will break

in half

[strange kestrel]

lMAO

[brittle wing]

and your ball will explode

[high night]

[strange kestrel]

BRO'S ACCOUNT IS BARELY 2 HOURS OLD AND HE'S ALREADY MAKING ENEMIES

[high night]

Wtf

[brittle wing]

YOUR

CRICKET BALL

AINT THAT CRICKET

[high night]

Oh...

[brittle wing]

.

[strange kestrel]

lmao

LMAAO

OH MY GOD

[brittle wing]

almost got a ban

[strange kestrel]

[brittle wing]

.........

silence speaks louder than any noise could

[brittle wing]

didn twork

[strange kestrel]

my model is ready

moment of truth

[brittle wing]

gets declined

[strange kestrel]

AAAAAAAAAAAAAA

IT JUST

WONT

[brittle wing]

[lone vine]

when i download my model i only have the index file

[strange kestrel]

are you using applio

i had this happen just a few minutes ago

check your weights folder, i had the final pth file be saved there

@lone vine

[lone vine]

ye im using applio

[strange kestrel]

.

[lone vine]

oh ye it's in my weights folder

there is like 200 pth files here wtf

[strange kestrel]

this is what mine looks like

-ish

i was told it was the very top one

[lone vine]

ye mine is similar

[rare goblet]

@lone vine has reached level 2.

[strange kestrel]

because there are no funny numbers and whatnot

[rare goblet]

@strange kestrel has reached level 5.

[lone vine]

oh alr

[strange kestrel]

thank you bot, very cool, i am purple now

alright so i just took the pth file and added it to the index file in my folder

[lone vine]

this model i made sound so shit

[strange kestrel]

still testing my model, creating a cover right now

hopefully it sounds great with only 9 minutes of dataset

[lone vine]

for some reason the only sentences being said are the ones said in the dataset

[strange kestrel]

sounds horrible

my poor laptop worked 15 hours for a broken model

[lone vine]

how many epochs did you do

[strange kestrel]

300

it sounds like

like

[lone vine]

15 hours for a 300 epochs model

damn

[high night]

Back

[strange kestrel]

https://www.youtube.com/watch?v=n8GN7F1quCM

listen to that guy's voice

it sounds just like him

high pitched and fucking hilarious

im not even mad

this is just funny

[jolly cliff]

Een nederlander of wa

Dutch guy?

[strange kestrel]

im gonna change the uhhhh pitch

[astral jungle]

Helo

[high night]

Ok im back

@strange kestrel

I'm back now

[strange kestrel]

welcome back checkmate

1 minute left on your timer

are you gonna pull a magnus carlsen or something on me now

[high night]

R prappaganda

[strange kestrel]

ayo my voice doesn't sound that bad

i changed some settings on replay

now it sounds good

only one problem though

the music is very loud, louder than the voice

[high night]

@strange kestrel

Was your dataset good

[strange kestrel]

@high night

define good

[high night]

clear, without artifacts and music

[small zodiac]

bro can anyone help with this rvc shit in the gc

😭

[strange kestrel]

it's just me talking in a normal way for almost 10 minutes straight trying to make a range of different expressions and there was no outside audio

[high night]

I would but I'm currently eating good

[high night]

How many epochs

[strange kestrel]

300

it's just somewhat quiet, that's all

[small zodiac]

eat bad then

[strange kestrel]

i changed some unga bunga settings on replay and now it (almost) sounds lik me

[high night]

I think you overtrained it

[high night]

[high night]

Tf is replay

[strange kestrel]

the uh thingy app thing to make thing covers thing

is there a better way

gimme your suggestions

[high night]

Definitely a better way

But you need good ass pc

[astral jungle]

Hello

[rare goblet]

@astral jungle has reached level 7.

[strange kestrel]

is an rtx3050ti sufficient

[high night]

Hello bro @astral jungle

Bro

I hate seeing 1050,2050,3050

2050 probably don't exist but

[strange kestrel]

bro i'm a budget broke ass university student

[high night]

My eyes hurt

[strange kestrel]

2050 exists btw

[high night]

But yeah it's good

[strange kestrel]

alright

what do i do

tell me and i'll do it father

[high night]

-rvc

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation to other languages

🇧🇷 Brasil (PT-BR)

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

🇫🇷 France

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

🇩🇪 Germany

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

🇮🇹 Italy

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

🇯🇵 Japan

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

🇵🇱 Poland

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

[high night]

Probably english one

[strange kestrel]

alright alright

[high night]

@strange kestrel after that check ur dms

[strange kestrel]

ooo kinky

[high night]

@strange kestrel

[rare goblet]

@high night has reached level 21.

[high night]

Ahahahahahha

[strange kestrel]

oui

[high night]

Idk where tf mangio rvc fork went

There was a tutorial for it

But now it isn't there

-mangio

Bruh

[strange kestrel]

mangio is dead

[high night]

@strange kestrel Guide: https://docs.google.com/document/d/1lIAK4Y0ylash_1M2UTTL_tfA3_mEzP0D2kjX2A3rfSY/edit?usp=sharing
Guide is very insightful and will resolve most of your questions.
Made by @oblique rain 👾

This should do the trick

[strange kestrel]

thanks dad

paperspace

[high night]

Sorry my brotha lollen ape for the ping

[strange kestrel]

i gotta pay

[high night]

@strange kestrel use mangio

Or applio

[astral jungle]

What do you usually use to train the model?

[high night]

Rvc

[astral jungle]

Okay

Where are you from?

[high night]

India

[timber crescent]

no way

[strange kestrel]

[astral jungle]

Oh

[strange kestrel]

dont look at me

[timber crescent]

its not dead

you prank

[strange kestrel]

it's alive

IT'S ALIIIIIIIIIIIIIVE

[timber crescent]

sure

[strange kestrel]

IT BEING DEAD IS JUST A FABLE

A STORY TO SCARE CHILDREN

[timber crescent]

k me..

[high night]

@timber crescent ur name is familiar

[timber crescent]

yea where you see me

[high night]

Discord

[timber crescent]

okay okay

you see me on old ai hub

[high night]

Nah bruh @venom is watching us

[crude steppe]

Hi, how can I convert multiple audio files ? Like with a queue.

[astral jungle]

Convert from mp4 to mp3??

[crude steppe]

no, with the ai

in the Model Inference tab

[astral jungle]

That means you want AI Cover , right?

[crude steppe]

yes

but I have 731 wav file and I don't want to do all of them mannualy

[astral jungle]

How many seconds per file?

[crude steppe]

from less than 1 to 6

[rare goblet]

@crude steppe has reached level 1.

[crude steppe]

the total size of the folder is only 100mb

[high night]

Just merge them

[astral jungle]

Do you have the model AI yet??

[crude steppe]

yes

[astral jungle]

So let's merge and let AI Cover again

[crude steppe]

But how can I unmerge them after that ?

[astral jungle]

I haven't considered it either

[crude steppe]

oh

maybe I can automate it in python ?

[candid sparrow]

hello how come i get this when i try training my own ai ? loss_disc=nan, loss_gen=nan, loss_fm=nan,loss_mel=nan, loss_kl=4.779

[rare goblet]

@candid sparrow has reached level 1.

[naive fossil]

why do i have only those algo?

[proper shale]

check weights

[proper shale]

older GUI

probably

idk

[strange kestrel]

i did

before you told me

thank you mordekaiser

https://tenor.com/view/brazil-youre-going-to-brazil-your-going-to-brazil-ur-going-to-brazil-gif-17868304

[proper shale]

😭

[strange kestrel]

also

[naive fossil]

those as been linked to me

[strange kestrel]

just as a reminder

a chess term complimented you

[naive fossil]

where i find the last version of rvc?

[proper shale]

ik haha

[maiden sluice]

Does anyone has a gradio link ^_^ ?? 😦

[lusty willow]

I can't seem to get Colab working (I'm a subscriber).

I keep being greeted by the message "Unable to convert function return value to a Python type!"

Little help?

[astral jungle]

Helo

[cobalt oyster]

i am unable to preprocess data

@proper shale

[scenic gull]

is there a guide on how to use mangio

[wispy burrow]

What are you using?
What is the error?

[loud berry]

https://colab.research.google.com/drive/1r4IRL0UA7JEoZ0ZK8PKfMyTIBHKpyhcw#scrollTo=0o9rMOIk1EEF

[loud berry]

why is this not working?

[cobalt oyster]

@wispy burrow I am using
Easy👨- 10/17 (Inference)
Easy👨- 10/17 (Inference)_Notebook unstarred

[rose kindle]

i recommend rvc v2 disconnected

[loud berry]

that one is sophisticated

[cobalt oyster]

It is stuck at starting and ends and shows tick but the audio is not preprocessing I have done many models but this the first time I am facing this issue

[rose kindle]

is ok

[cobalt oyster]

@loud berry

same error for u too

[wispy burrow]

Sounds like a Colab.
Is it this thing?
EasyGUI by Rejekts https://colab.research.google.com/drive/1r4IRL0UA7JEoZ0ZK8PKfMyTIBHKpyhcw

[loud berry]

yes it was the best I was using it for almost 6 months but wth happened to it

[cobalt oyster]

yes

@wispy burrow

[wispy burrow]

Did you zip your Dataset?

[cobalt oyster]

No

[wispy burrow]

You need to zip your Dataset for Colabs.

[scenic gull]

hey are you allowed to ping a mod for help

[loud berry]

It's not required when it comes to this colab

[scenic gull]

if anyone knows how to use mangio i would appreciate the help

[cobalt oyster]

I will try zipping @wispy burrow I will be back after trying

[rare goblet]

@cobalt oyster has reached level 3.

[covert cipher]

I trained my model but covers I made are really VIRTUAL-like. I don't understand the difference my model is broken or the settings I'm picking for ai cover is wrong, is there someone to help for this?

[rare goblet]

@covert cipher has reached level 1.

[wispy burrow]

Training with RVC v2
https://docs.aihub.wtf/guide-to-create-a-model/model-training-rvc

[slate pawn]

I need help with local v2 training

I can't send a screenshot but the process data isn't working

saying the directory name is invalid

[wispy burrow]

You can't post screenshots just yet.
Get level 1 first.

[slate pawn]

dang I lost all my levels somehow

[rare goblet]

@slate pawn has reached level 1.

[slate pawn]

BINGO

here it is

[wispy burrow]

It's a different AI Hub.

[slate pawn]

for real??

No wonder things feel different here

I thought things just changed due to my inactivity lol

[wispy burrow]

The old one got deleted.

[wispy burrow]

Can you show the cmd?

[cobalt oyster]

@wispy burrow I tried zipping not working after helping him could u help me brother

[slate pawn]

this one?

[wispy burrow]

Oh, I think I see it.
You can't zip your Dataset on local Mangio.

[slate pawn]

oohh

makes sense, it's always the zip

[loud berry]

wth does this mean

[wispy burrow]

Maybe post a screenshot of the error?

[astral jungle]

Like me

[cobalt oyster]

ok I will send

[slate pawn]

Hey could you tell me about the new training methods? is Applio still working?

[loud berry]

I tried to contact the creator of the notebook but to no avail

[cobalt oyster]

It is showing starting and ending

[wispy burrow]

New training methods?
Applio IF you can get it installed and working properly.
But the the installer is horrible still.

[cobalt oyster]

like this

[astral jungle]

Bruh

[slate pawn]

Nothing changed huh

[astral jungle]

I tried it

[cobalt oyster]

@loud berry same error for u ?

is it working

[astral jungle]

The result is still error

[loud berry]

not working

[astral jungle]

https://colab.research.google.com/drive/1khsjzEigBogtkA6ZNL3gNXZqmjUAKkM1?authuser=1#scrollTo=pjlwm8SOwWVI

[astral jungle]

@loud berry

Try this link

[cobalt oyster]

ok

do we have to zip file

[slate pawn]

Just to check, Am I good to go to feature extractions now?

[astral jungle]

Create zip and upload to dataset

[wispy burrow]

It's says your Dataset is empty?
And what's that last sentence?

[cobalt oyster]

ok thanks for help @astral jungle

[loud berry]

almost similar, I think they cloned it

[loud berry]

really

[crude steppe]

Hello, is there any way to force the ai to keep the same lenght ? Idk why but it make it a bit shorter.

[slate pawn]

that's actually the last thing it says

[rare goblet]

@slate pawn has reached level 2.

[slate pawn]

and ended

took a couple seconds

[wispy burrow]

Wait for the cmd to finish.

[scenic gull]

im working on huggingface, does it usually take long to convert a song

its been processing for a while

[slate pawn]

yeah but isn't it finished? it said end preprocess and it's not loading anymore

[wispy burrow]

Not all threads were at 100% yet in the screenshot.

[astral jungle]

In the zip file is the voice of the character you want to train

[slate pawn]

Yeah I figured, That's why I came to check because even when not all were 100%, The preprocess ended

[loud berry]

I must do it? (my heart tells me no to)

[slate pawn]

Though I think that if I tried processing data again the results will change

changed by a lot

it also ended preprocess here

[astral jungle]

What do you fear?

[loud berry]

failure ofc

[rare goblet]

@loud berry has reached level 3.

[scenic gull]

does anyone know how to use either huggingface online rvc or mangio local rvc?

please anyone

i did everything the guide tells me to do and it wont work

[slate pawn]

Do you fear the consequences of your actions

[loud berry]

no failure related to self-embarresment

I've made over 20 models

[cobalt oyster]

@astral jungle same issue brother

[astral jungle]

Bruh

[wispy burrow]

Did you share those on the old AI Hub?

[loud berry]

no I didn't, just a Telegram group that I've created

[astral jungle]

I've had the error since last night

[cobalt oyster]

same for me

I have the same issue since last night

[slate pawn]

I'm being pranked at this rate

[astral jungle]

That's right, I planned to train the model when I went to sleep, but when I woke up, it didn't work out

[cobalt oyster]

Is there any other way to train

I hope the issue get fixed soon

[wispy burrow]

Maybe lower the CPU processes?

[astral jungle]

I hope so

[wispy burrow]

RVC Disconnected

[slate pawn]

does the number of CPU processes change the quality of the model?

[cobalt oyster]

I will try @wispy burrow Thanks for spending time helping me I will try if there is issue I will msg here

[wispy burrow]

It shouldn't lower quality but will take a bit longer to process.

[astral jungle]

Do you need Rvc Disconnected

[cobalt oyster]

Yes

I found it

[wispy burrow]

There's even a guide for RVC Disconnected.
#1160926856548200531

[cobalt oyster]

I will check

[slate pawn]

you're kidding me

[astral jungle]

https://colab.research.google.com/drive/1XIPCP9ken63S7M6b5ui1b36Cs17sP-NS?authuser=5#scrollTo=g1X5oANker6l

Here

[cobalt oyster]

Thanks

[loud berry]

how to contact notebook owners for personal purposes

[brittle wing]

How can i fix these errors i tried everything

i never got these images

[wispy burrow]

Don't have any spaces or special characters in your path to the RVC folder.

[brittle wing]

how do i fix it? i did nothing wrong

[wispy burrow]

Don't have any spaces or special characters in your path to the RVC folder.

[brittle wing]

show me a video of how you did it @wispy burrow

[wispy burrow]

The folder 'Mangio-RVC-main (3)' has a space and the special characters ()...
Remove those.

[brittle wing]

okay?

@wispy burrow it did not work

@wispy burrow how can i

[astral jungle]

Can you do it?

[brittle wing]

@astral jungle i dont know how this even happened....

[astral jungle]

What's the matter?

[cobalt oyster]

@astral jungle still trying kind of complicated

[rare goblet]

@cobalt oyster has reached level 4.

[astral jungle]

Which step are you on?

[brittle wing]

idk

[astral jungle]

@cobalt oyster

[brittle wing]

i think i just got scammed @astral jungle

[slim owl]

Look at this Easy👨- 10/17 (Inference): 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.
lida 0.0.10 requires python-multipart, which is not installed.
llmx 0.0.15a0 requires cohere, which is not installed.
llmx 0.0.15a0 requires openai, which is not installed.
llmx 0.0.15a0 requires tiktoken, which is not installed.
plotly 5.15.0 requires tenacity>=6.2.0, but you have tenacity 5.1.5 which is incompatible.
plotnine 0.12.3 requires numpy>=1.23.0, but you have numpy 1.22.0 which is incompatible.
tensorflow 2.14.0 requires numpy>=1.23.5, but you have numpy 1.22.0 which is incompatible.

[cobalt oyster]

@astral jungle loading data set

[astral jungle]

Who scammed

[rare goblet]

@astral jungle has reached level 8.

[astral jungle]

Is this step correct?

[cobalt oyster]

yes

[astral jungle]

Go to Google Drive and there will be a section called rvc disconnected

[proper shale]

avoid spaces in folder names

[brittle wing]

@proper shale i did that it still showed the same error message

[astral jungle]

@cobalt oyster

[proper shale]

Not sure then

[cobalt oyster]

@astral jungle Thanks a lot

Training started

😭 I have never seen a person helping me like this Thanks

once again

[astral jungle]

I feel happy to help you

[brittle wing]

is there a "RVC disconnected" tutorial on yoube beacuse i am super confused

[slate pawn]

could anyone tell me the difference of a 8 Batch size and 16 batch size?

[rose kindle]

say it!!

[brittle wing]

????

[astral jungle]

That's it, I'll go to sleep

[slate pawn]

I forgot which is which in higher qualities

[astral jungle]

Bye everyone

[rose kindle]

ok, so be careful, the more batch_size you add, the model becomes relatively faster but loses quality, and the less batch_size you add, the model becomes slower but better in a way, I would recommend 7-8 batch_size is the most efficient

[slate pawn]

Thanks man

[wispy burrow]

Apparently, the lower the Batch, the better the quality but will be slower to train.

[rose kindle]

you're welcome! 🙂

[rose kindle]

yup

[slate pawn]

wait so it's the training speed?

I thought it was the delay when using RVC

[rare goblet]

@slate pawn has reached level 3.

[rose kindle]

no no

as i said, i recommend 7-8 batch size

[slate pawn]

I don't mind having to train for longer if it gives a better quality

So how about I use 4?

[rose kindle]

that s my bro

[rose kindle]

perfect

better quality but long time to wait

but it s worth it

[slate pawn]

I was going to say 2 but I thought I might be going overboard

[slate pawn]

exactly

[rose kindle]

i am going to use 7 batch size anyway cause i tested a lot and does a great job every time

[slate pawn]

The key to great work is patience

[rose kindle]

2 is good as well

[rose kindle]

YES

[astral jungle]

Is it okay if I set 8 batch_size?

[rose kindle]

yep

[rose kindle]

it s more than good

[slate pawn]

I used to train too before but got inactive

All my knowledge in training models got erased

[astral jungle]

Okay

[rose kindle]

it s ok, you gonna recover your knowledge with us 🙂

[slate pawn]

lol you should've seen the art I used to make in the old AI hub

the old days

[broken quest]

Guys I still need help

Please

G:\RVC-MAngio\Mangio-RVC-v23.7.0_INFER_TRAIN\Mangio-RVC-v23.7.0>runtime\python.exe infer-web.py --pycmd runtime\python.exe --port 7897
Traceback (most recent call last):
File "G:\RVC-MAngio\Mangio-RVC-v23.7.0_INFER_TRAIN\Mangio-RVC-v23.7.0\infer-web.py", line 79, in <module>
conn = sqlite3.connect('TEMP/db:cachedb?mode=memory&cache=shared', check_same_thread=False)
sqlite3.OperationalError: unable to open database file

This is the error message when I try to open go-web.bat

[rose kindle]

at first, when i first start with the rvc i ve seen a lot of guys on youtube using 40 batch size cuz they said that is gonna train a lot faster, but ok i understand that you want faster but man you have to be patient cuz with training faster you can t recieve a very good quality let s be honest

[rare goblet]

@rose kindle has reached level 3.

[slate pawn]

Doing quick maths with wrong answers

[rose kindle]

exactly

u know how we say Haste makes waste.

[slate pawn]

I started training 100 epoch on batch size 2

Gonna see how long it takes

Perhaps next week

[broken quest]

Can no one help me?

[rose kindle]

at me with 2 batch size it takes about 4 hours 100 epoch

[broken quest]

Please

Can someone help me

What is going on?

[rose kindle]

i really don t know srry

[slate pawn]

could you send a ss

[broken quest]

a what?

[rose kindle]

XDD

[rose kindle]

screenshot

[slate pawn]

screenshot

[rose kindle]

🙂

[broken quest]

Getting it now

[rare goblet]

@broken quest has reached level 2.

[broken quest]

Thia

This

What does this error mean

[slate pawn]

what did you open?

[broken quest]

go-web.bat

[slate pawn]

did it not take you to mangio website?

[broken quest]

no

I just get this error message

[slate pawn]

and nothing happens when you press any key to continue?

[broken quest]

yes

It just exits out of the command prompt

[slate pawn]

okay let me check for a bit

this is your first time using local right?

[broken quest]

yes

First time

[teal glacier]

for some reason it just doesn't download the runtime.zip for me in apllio what can I do? and if applio is still unreliable what alternatives I could use because the colab also doesn't work (I want to make a model)

[rose kindle]

i think bro, you just have to reinstall it

so, as a conclusion to what we discussed previously, batch_size must be as small as possible if you rely more on performance and you really need that model, if you want to do it for fun and you don't necessarily mind the quality but you want to it is done faster, then it would be advisable to use a larger batch_size.

[slate pawn]

I think one of the things you should avoid is having any spaces in the RVC folder leading to go-web

and avoid spaces in dataset folder too

[broken quest]

Let me try that

[broken quest]

But there weren't any spaces

[slate pawn]

in that case unsure

Consider reinstalling it as Venom suggested

Make sure it's the same latest version

https://docs.google.com/document/d/1KKKE7hoyGXMw-Lg0JWx16R8xz3OfxADjwEYJTqzDO1k/edit

I have this guide I kept on installing Mangio RVC local

Hope it helps you

[broken quest]

I'm going to try installing the folder directly into the main directory

[livid palm]

I just finished training my first model with Applio. I see that there are two pth files, D and G. Do I only need one of them or both of them? Last time I had a model I zipped up the index file with my pth, but now I have two pth files and I am not sure if I can put both of them with the index in the zip file

[slate pawn]

shouldn't you just put the pth with your model name?

[rare goblet]

@slate pawn has reached level 4.

[livid palm]

Still a little new to this, what do you mean by model name? Like a file with my model name? The index is my modelname.index

Never mind I have figured it out. Thanks tho

[rare goblet]

@livid palm has reached level 1.

[ancient latch]

Is it normal that applio installer doesn't do anything for a while after cloning the repo? The disclaimer says the install is going to take a significant amount of time, but I don't see that it's doing anything

GIT is installed

[viscid wedge]

i get a lot of errors at the bottom part of the app when i convert. how do i fix it?

[copper pebble]

ya doing cover??

[viscid wedge]

yeah

rvc gui

[ancient latch]

Still trying to get it to work. Applio Installer doesn't do anything so I installed everything manually: BuildTools, VC Redist and Python. Python is added to PATH.

When I try to run it, I get the not found error. The fix states that I should add runtime/scripts to PATH, but I don't have such directories in my Applio folder.

[rare goblet]

@ancient latch has reached level 1.

[crisp plover]

Hey how do i use RVC in realtime on discord for example ?

I open the RVC - GUI ( running go-realtime-gui ) and then loading my model and index

and start audio conversion, it is working but i just hear it playback on my desktop

[brittle wing]

How do I use models I downloaded on websites for phone

[broken quest]

File "G:\Mangio-RVC-v23.7.0\infer-web.py", line 79, in <module>
conn = sqlite3.connect('TEMP/db:cachedb?mode=memory&cache=shared', check_same_thread=False)
sqlite3.OperationalError: unable to open database file

I got the same message, even thugh its in the root folder with no spaces

Is the issue something with Python or my GPU

It says database file. Should I try to delete that?

[broken quest]

What do I do?

[covert bramble]

so im getting this error "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." how do i fix this?

[broken quest]

Nothing seems to work

[rare goblet]

@broken quest has reached level 3.

[broken quest]

And I'm starting to get frustrated.

[brittle wing]

.

[dim cape]

is anyone else having this problem that doesnt let u create the index?

not output, nor a single error in cmd

also uhh how to change applio language? im spanish but i prefer english

[rare goblet]

@dim cape has reached level 1.

[broken quest]

If it helps ant I have an Intel CPU

[ancient latch]

Let the installer run for over an hour. It did nothing other than clone the repo.

How would I get the runtime/scripts folders because they dont come with the github files?

[brittle wing]

@dim cape I'm having problem idk what websites to use for my models I downloaded here I downloaded a pomni one and I want to use it to make ai covers

[teal ferry]

Hi, i got a strange reading on one of my tensorboards. this is the loss/g/total graph. Training a voice around 8k - 10k steps into training, so between epoch 800-1000. Has anyone seen this before and can give insight on to what happened please let me know https://i.imgur.com/9VuwdLz.png

[vapid gust]

You can send a picture here

You don’t need to send a link

[teal ferry]

it didnt work

yeah it wont let me control v

[vapid gust]

Send in dms

To me

[dim cape]

You can use RVC, or Applio (Recommended).

You can get the local Applio version in /IAHispano/Applio-RVC-Fork

In GitHub.

You must know that isn't official and it can be changed to some kind of malware, which I obviously doubt, but can happen.

[sand iris]

What isn't official ? Applio ? We have a few contributors here

We know who to spank if there's an incident

[dim cape]

Oh, it's okay then, alr.

[ancient latch]

These are the issues I am facing when I select Option 1 Runtime (Nvidia Support) or Option 4 (Nvidia Support):

[dim cape]

It's probably not using the 3.9 version of Python

maybe you have another version in your computer.

I had 3.12 and I uninstalled it because I didn't really need it, I could stay with 3.9

proper grammar on top 100%

so professionel

[ancient latch]

I do in fact have 3.11 from another project. Uninstalling it to check if it gets me any further

Same errors as before

[dim cape]

well, try searching up how to install tqdm module in python, maybe pip install tqdm

[gentle finch]

Apologies to be a bother, But after making a model and Index, whenever I uploaded it to my RVC. I get This Code

Traceback (most recent call last):
File "restapi\MMVC_Rest_Fileuploader.py", line 92, in post_load_model
File "voice_changer\VoiceChangerManager.py", line 170, in loadModel
File "voice_changer\RVC\RVCModelSlotGenerator.py", line 37, in loadModel
File "voice_changer\RVC\RVCModelSlotGenerator.py", line 43, in _setInfoByPytorch
KeyError: 'config'

And it does not appear. I have tried going to older .Pth Models but that did not work either. Any Ideas?

[dim cape]

Haven't really heard of it, did u install all requirements?

[pastel marsh]

Hello guys, anyone had experience with Paperspace? Trying to buy 8$ subscription "Gradient Pro", even tryed with my friends with different cards and phone number, but its just doesnt work, its says something wrong with card, but all cards which are been used are absolutely fine(

Paperspace Support saying just try again, what the hell.. still not working 😄

Anyone can help..........?)

[dim cape]

uhh this is the help rvc channel

[rare goblet]

@dim cape has reached level 2.

[dim cape]

i dont know if thats actually related or smth

🤷‍♂️

[gentle finch]

I believed so? I got everything the guide was telling me

[dim cape]

that's strange, havent heard of paperspace when i was setting up rvc

[prime hemlock]

it's an alternative to running on colab

[gentle finch]

I'm not using paperspace?

[prime hemlock]

or local

[gentle finch]

Yeah, I'm usin local

[rare goblet]

@gentle finch has reached level 1.

[dim cape]

oh kk

[ancient latch]

Fixed errors one by one, now I am getting somewhere when I choose Option 4. It's downloading stuff from huggingface right now

[pastel marsh]

I know) Its same as colab thing, just trying to buy subscription) to use RVC infer)

[gentle finch]

Figured it out, Grabed the wrong pth file

[broken quest]

It's still not working for me

[wispy burrow]

For 1.
Launch Go-applio-manager.
And then
[3] Download NVDIA Runtime.

[rare goblet]

@wispy burrow has reached level 17.

[broken quest]

Is it even possible to get it working on non-nvidia GPUs

If it helps, it is also on an external hard drive

[wispy burrow]

Training? No, not yet at least.
Inference can be used with AMD in Applio.

[sand iris]

Surprisingly intel ARC can train

[wispy burrow]

Hey, I remember seeing this one!

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English Documentation

https://us.aihubfrance.fr/

Credits: AI HUB France

🇺🇸 English FAQ

https://us.aihubfrance.fr/faq

Credits: AI HUB France

Guides

🍏 Applio-RVC-Fork: Guide

Link: Google Docs

Credits: IA Hispano

How to install RVC on windows

Link: YouTube

Credits: iroaK

🌐 Translations

• 🇫🇷 French Documentation
• 🇮🇹 Italian Guides
• 🇷🇺 Russian FAQ for RVC/Voice changer
• 🇪🇸 Español (-rvc es)
• 🇧🇷 PT-BR (-rvc br)

[mighty sinew]

im new to this stuff and im trying to make a voice over from a game character where do i train the voice?

or make the voice even

[wispy burrow]

You'll need RVC to train. There are several versions...

[wispy burrow]

First of all, you need to collect as much audio clips of the voice as possible for the Dataset.

[mighty sinew]

well i have about a minutes worth

the character is from a game called gang beasts it just grunts and makes noises i was gonna use it to make a song cover for a tiktok

[wispy burrow]

The recommended length is 10-50 minutes of audio without any silence or any unwanted sounds.

[mighty sinew]

oh damn

if im not looking for a very detailed replication will just a minute do

[wispy burrow]

It's a recommendation, it doesn't mean it can't work with only 1 minute.

[mighty sinew]

ah okay

[rare goblet]

@mighty sinew has reached level 1.

[mighty sinew]

which rvc would you recommend

[half cove]

rvc2

[wispy burrow]

If you have a decent NVIDIA GPU you can train locally.
I would suggest Mangio RVC.

[mighty sinew]

where is the link for the mangio rvc

or how would i download it

[wispy burrow]

Download Mangio-RVC-v23.7.0_INFER_TRAIN.7z from
https://huggingface.co/MangioRVC/Mangio-RVC-Huggingface/tree/main

[mighty sinew]

tyvm for the help

[wispy burrow]

Simply extract the zip.
And after that launch Go-web.bat.

And you should be good to go.

[mighty sinew]

which should i download

[wispy burrow]

Mangio-RVC-v23.7.0_INFER_TRAIN.7z

[mighty sinew]

okay thank you

[foggy totem]

is there a way to take the files from my local applio session, transfer it to a rvcdisconnected colab, and resume training from there?

[astral jungle]

Helooo

Good morning

[vapid gust]

For the dataset

Do we have to remove the sound?

Like the silence

Do we have to remove the silence

[brittle wing]

how to configure rvc

Hypothetical speaking, is it possible to train voice ai with your voice combined with another voice to create a whole new voice or is this complex for Ai?

[astral jungle]

It works

[glad zealot]

yes theres a feature in the voice changer that lets you merge 2 voices into a single one

not sure how to do it on applio tho but iirc you can also merge there but not sure

[cyan maple]

did you find a fix?

[potent root]

yeah it's called ckpt processing

[rare goblet]

@potent root has reached level 6.

[potent root]

i do it all the time

it's a great feature

[lusty willow]

Is anyone else having issues with EasyGUI? I've been trying to clone voices for 2 days now, but every time I run it; it promptly stops and flags up an error. This morning, I keep getting "ModuleNotFoundError: No module named 'pyngrok'". Any advice?

[ancient latch]

Did that already. Now I'm just pip installing everything that is reported missing

[slate pawn]

lol I still remember how it left everyone confused

[tough light]

Nope, just went ahead and found a different colab.

[rare goblet]

@tough light has reached level 2.

[tough light]

The one I'm using RN works just as good as the previous one.

[slate pawn]

anyone know about overtraining points?

[ancient latch]

Finally got it to run, selected a voice model, a acapella audio file and tried to convert, but it results in an error (AttributeError: 'NoneType' object has no attribute 'tobytes')

[ancient latch]

Open CMD and pip install (module) until you run out of missing modules

[rare goblet]

@ancient latch has reached level 2.

[placid light]

Hello.

[placid light]

a

[astral jungle]

Yo

[placid light]

Are there any training colabs I can use? The one the official AI hub website showed to me had errors during the training process.

[astral jungle]

You can use Colab

[placid light]

Can you give me one please? I really appreciate it.

[astral jungle]

RVC V2 Disconnected

https://colab.research.google.com/drive/1XIPCP9ken63S7M6b5ui1b36Cs17sP-NS#scrollTo=Yj-_npuA_HRB

[sand iris]

Pstt, there's this

-colab

[azure marsh]

Google Colabs

RVC

• AI-Cover-Gen-No-UI (inference only)
• RVC v2 Disconnected (training only)

🌐 Translated colabs

• PT-BR: -colabs_br

Audio Separation/Isolation

• MDX Net Colab
• UVR arch colab 5.0.2/c1.04
• KaraFan colab

[sand iris]

No one saw me skill issue there

[placid light]

Is Firefox a non-chromium browser?

[lusty willow]

Apologies, I'm still a bit of a newb at this...CMD is the window that opens when you click "show code", right?

[rare goblet]

@lusty willow has reached level 1.

[astral jungle]

I don't know, you should try running it on Firefox to know

[high night]

Yes

Chrome and Firefox were literally rivals at some point

[formal lintel]

Why can't i run mangio rvc? it keeps giving me this error "DLL load failed while importing _multiarray_umath: The specified module could not be found."

[placid light]

Hey, can you give me one example of a colab disconnect prevention method?

Since it's hard to find one nowadays.

[marsh sierra]

Can someone help me install mangio rvc in vc later today, since google colab just makes me pay for runtime. And kattle just doesnt work.

[sand iris]

There's colab notebooks that can be used with the free plan

The UI is not quite as fancy... but it works

[marsh sierra]

Ah ok

Could you link it in my dms perhaps?

Oh and does it include villagers?

[ancient latch]

In Windows, search for CMD and open it, that's the command prompt. If python is installed correctly, you will be able to type pip install (name of the module) and let it install

[hard gust]

if it looks like it after only 100-120 epochs, could it be overtraining this soon?

[slate pawn]

dam I was just about to send mine lol

[slate pawn]

lower the smoothing a bit more so it can help point out the point

[hard gust]

it looks pretty much the same. The lowest point was during epoch #100

[slate pawn]

no need to lower the smooth that low

[proper shale]

nah its still goin

[proper shale]

140-150k seems just about right for you

check d total

[astral jungle]

Yo

@cobalt oyster How's it going

[cobalt oyster]

@astral jungle good

how's it going for u

[astral jungle]

It's good but I like EASY better

[lusty willow]

Ah, I gotcha. Unfortunately, I'm not using Windows. I'm on a Mac and using Google Colab, I don't have a powerful enough GPU for local installation.

[slate pawn]

Michael Jackson I missed you

[jagged sonnet]

Anyone use the AICoverGen WebUI?

[astral jungle]

Me

[slate pawn]

I think this is an overtraining point?

[proper shale]

na wait its still going

[jagged sonnet]

where do you find/get the intermediate audios?

[proper shale]

Haven't seen you in so long

[astral jungle]

YouTube

[mighty orchid]

Okay so, i tried using the Hyunjin voice model by @woeful cave "huggingface: https://huggingface.co/kaengpil/HanJisung/resolve/main/hanjisung.zip
if you use my model, please credit me @ bokpage on youtube and/or changbinpop on tiktok!" but it gives me this error message and i dont know what to do, before i got a different error message stating that something was circular(?) im really not a technology person so i need help

The error message:
ModuleNotFoundError Traceback (most recent call last)
<ipython-input-1-f28cdfd0b25c> in <cell line: 4>()
2 #@markdown Link the URL path to the model (Mega, Drive, etc.) and start the code
3
----> 4 from mega import Mega
5 import os
6 import shutil

ModuleNotFoundError: No module named 'mega'

---

NOTE: If your import is failing due to a missing package, you can
manually install dependencies using either !pip or !apt.

To view examples of installing some common dependencies, click the
"Open Examples" button below.

[jagged sonnet]

i mean the separated audios like the instrumentals since there's an option that you can check to keep the separated audios

[astral jungle]

That means you want to upload vocals without instruments so AI can sing them back?

[jagged sonnet]

I want all the separated vocals and instruments so I can fine tune them

[rare goblet]

@jagged sonnet has reached level 1.

[proper shale]

All the files are in the Colab file explorer

[rare goblet]

@proper shale has reached level 32.

[proper shale]

So uh yeah

[jagged sonnet]

hmmmmmm...i don't know what that is

[slate pawn]

i got a bit inactive so my training knowledge reached back to 0 lol

[proper shale]

It's in the left side of the UI in the Colab page, has a lil file explorer icon.

I think maybe audio output has everything

[proper shale]

Oh man haha

I spent a few weeks without training models so I can relate

[slate pawn]

bummer the old AI hub got deleted too

By the way is it true that the smaller the batch size the less quality it makes?

[proper shale]

less batch size is more training per epoch

so more accurate with less epochs

[slate pawn]

I keep reading and hearing about it but I never got a clear answer on if higher quality (less randomness) means lower batch size or higher batch size

[proper shale]

higher batch_size just means more epochs

but less training steps

the ideal batch_size needs to be found by you in every model

you gotta try out 4, 8, 12, 16

see what has the steadiest slope

in g/total

and you should be good

[slate pawn]

oh

[proper shale]

you can finetune too

[slate pawn]

Should I tell you about the current model I'm training on 2 batch size

[proper shale]

so like, 5 or 6 could be really good

[proper shale]

ooh more training per epoch, i like

[jagged sonnet]

where?

[proper shale]

Colab page

[slate pawn]

that took 7 hours to train 100 epochs lol

so is this info not true?

[proper shale]

where's this from

[jagged sonnet]

after clicking a bunch of random shmiz, i found the song output folder

and they are there, tnx

[proper shale]

np, glad to help! :)

[jagged sonnet]

are the downloaded files transfered elsewhere? i see a loading icon when i download them and then...nothing happens

[proper shale]

try going on another browser to download the files

[jagged sonnet]

nvm it was because of the automatic block when multiple downloads happen

[proper shale]

oh

[nova shard]

Why am I getting this error when downloading Applio?

FileNotFoundError: [Errno 2] No such file or directory: 'assets/pretrained/D32k.pth'

[idle musk]

What do de "passthru" button do

[echo creek]

When should I stop training based on the shapes of the graphs in Tensorboard?

[idle musk]

I dont think anyone's gon help us...

[rare goblet]

@idle musk has reached level 1.

[proper shale]

-overtrain

[azure marsh]

Hey, @echo creek!

👇 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

[pliant stream]

hello I need to know how to create a Lord Garmadon ai voice model! If you know how could you please help me?

[proper shale]

#1159513888199540817 can help

[pliant stream]

thanks

[shy brook]

hey does anyone know how to fix the error of pip not wanting to install anything?

I get a syntax error every time

[proper shale]

huh

i think u should download only one module per pip install

so like

pip install torch

then do the others

separately

[shy brook]

doing them separately gave the same error

I tried installing just torch and it still gave syntax error

[proper shale]

try it on powershell or something

[shy brook]

I'll try that once I'm home

[half cove]

https://tenor.com/view/thanos-memoji-gif-23490017

fuck

embed fail

[eager pelican]

haha

[hard gust]

Can you send me a link to the automatic overtraining detector?

[echo creek]

I don't understand even after reading.

Should I stop training?

[glad zealot]

that seems to still be fine

[rare goblet]

@echo creek has reached level 2.

[echo creek]

Oh, thank you. I thought I should stop now because someone said that when the graph starts to rise, you should stop training

[glad zealot]

it seems to not be movign to that might be good

you can just try the saved pth file on voice changer or covergen to check if its sound they way you want it to

[echo creek]

So far, having a higher number of epochs has sounded cleaner to me.

[glad zealot]

you can still test it now

i mean if thats local you can just test later

[echo creek]

Oh, so what you're saying is that I should try various experiments after training first.

[glad zealot]

you can just test the pth files saved and if you like how it already sounds like you can just stop it

[echo creek]

Ah, I see. Thank you!

[plain oyster]

is this the disconnected version?

[rare goblet]

@plain oyster has reached level 1.

[plain oyster]

can you send the links to these if you dont mind?

[glad zealot]

its their own model so ask them not me

[plain oyster]

but you said you can test the pth file on voice changer and covergen

[glad zealot]

ya

[plain oyster]

are those sites or different softwares

[glad zealot]

voice changer is in the pin of #🔍│help-w-okada and covergen is in the pin of #✨│ai-help

[echo creek]

yes

[plain oyster]

thank you very much

[plain oyster]

i see it seems to be the only one worming at the moment

[echo creek]

I don't know either. I was just following the guide.😅

[plain oyster]

oh lol i am still new to this too

[astral jungle]

Yoooo

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English Documentation

https://us.aihubfrance.fr/

Credits: AI HUB France

🇺🇸 English FAQ

https://us.aihubfrance.fr/faq

Credits: AI HUB France

Guides

🍏 Applio-RVC-Fork: Guide

Link: Google Docs

Credits: IA Hispano

How to install RVC on windows

Link: YouTube

Credits: iroaK

🌐 Translations

• 🇫🇷 French Documentation
• 🇮🇹 Italian Guides
• 🇷🇺 Russian FAQ for RVC/Voice changer
• 🇪🇸 Español (-rvc es)
• 🇧🇷 PT-BR (-rvc br)

[marsh sierra]

can you link it?

[sand iris]

-colab

[azure marsh]

Google Colabs

RVC

• AI-Cover-Gen-No-UI (inference only)
• RVC v2 Disconnected (training only)

🌐 Translated colabs

• PT-BR: -colabs_br

Audio Separation/Isolation

• MDX Net Colab
• UVR arch colab 5.0.2/c1.04
• KaraFan colab

[marsh sierra]

audio isolation?

[viscid moss]

-uvr

[azure marsh]

Tag suggestion for @marsh sierra

Guides

📑 Perfecting Audio Isolation on Low-End Rigs: A Practical Guide (by Litsa The Dancer and Faze Masta)

https://docs.google.com/document/d/1HmkG9cmL8SLX7-vJcPT1-1KgUQtCrwXB8CicYmG4LW8/edit?usp=sharing

📑 Guide: Gathering and Isolating Audio (by SCRFilms)

https://docs.google.com/document/d/1wTJ_wutDqEtsA99BJOXDDGax25pPIDE84O5E2Rio5Qk/edit#heading=h.g896c5vbh0dv

📑 How to make a good model All-In-One guide by LUSBERT

https://rentry.org/RVC_making-models

▶️ Vocal mixing tutorial by Roomie

https://www.youtube.com/watch?v=rSxLF3kkBw0

Tools

Youtube Downloaders

• https://github.com/yt-dlp/yt-dlp
• https://github.com/dsymbol/yt-dlp-gui

Audio Separation/Isolation

• https://ultimatevocalremover.com/
• https://mvsep.com/
• 🆕 https://github.com/Captain-FLAM/KaraFan
• https://goyo.app/ (Note that this plugin is for clean ups only, like removing background noise, reverb, ambience or sustained sounds)')

Tip: Use -colab command if you're looking for audio separation/isolation colabs.

[sand iris]

which one ?

[marsh sierra]

the above

[sand iris]

There's multiple

[marsh sierra]

AI-Cover-Gen-No-UI

[proper shale]

Just use RVC HF or the one in this channel's pin

[sand iris]

It might be because you ran banned colabs before

[marsh sierra]

huh

[proper shale]

Nah actually that one is restricted

[sand iris]

It is ?

[proper shale]

Believe it or not

It is

[sand iris]

lmfao

[proper shale]

I've tried it

Loll

[viscid moss]

The bot's AI Cover Gen collab has a warning and that's why it disconnects

[marsh sierra]

oh ok

[viscid moss]

Try my fixed version (spanish)

[marsh sierra]

i hope mc villager there hehe

[viscid moss]

https://colab.research.google.com/drive/1qezaBvCn4eSG1U2L4a64tEHTynbIu8eN?usp=sharing

[marsh sierra]

idk how but ina mod aicovergen is working

[viscid moss]

It's exactly the same but in spanish

[marsh sierra]

ah ok

[viscid moss]

lol

[marsh sierra]

so its hina mod but fixed?

[rare goblet]

@marsh sierra has reached level 4.

[proper shale]

na its the non web ui but fixed

hina's mod will work fine so

[marsh sierra]

ah

[glad zealot]

somehow the gui works and the noui doesnt lol

that sounds kinda backwards

[proper shale]

lmao

[glad zealot]

easy gui almost working

imma try fixing it myself too

[elder cloak]

bro what is going on??? I'm using moistCr1tikal V2 and every time the audio is complete the original vocals are still attached!

[ancient latch]

Does anyone know how to get rid of this error?

[slate pawn]

Found it here, sorry for the late reply https://rentry.org/RVC_making-models

[brittle wing]

which guide link is the best on rvc cover

[brittle wing]

https://docs.aihub.wtf/vocal-cloning-guide/rvc-v2

[loud berry]

https://colab.research.google.com/drive/1r4IRL0UA7JEoZ0ZK8PKfMyTIBHKpyhcw#scrollTo=G0MEhFM19Vq6

why does it not extract features

[brittle wing]

there was like something easy i clicked a button then it would give me a gradio link

is that still up

[pliant creek]

vffjf

hjtjkfhghddddddgggggggggggggggggggggggggggggggggggggggggggggggggggggggggggggggggggg

[elder cloak]

Help! 😦

[pliant creek]

what

why help

[elder cloak]

☝️☝️☝️

[pliant creek]

oh

hello

jolly holly

[elder cloak]

Hi

[loud berry]

pleaseeeeeeeeeeeeeee

[pliant creek]

please=?

[rare goblet]

@pliant creek has reached level 1.

[pliant creek]

oh my old friend don't worry

mal heni martin

orospi çocuğu

geyikkkkkkkkkkkkkkkkkkkkk

fshddjsgfjd

ssfhdjhd

jolly hollly

[viscid moss]

try rvcDisconnected for train

[pliant creek]

why

why

you are is bot ?

[loud berry]

that one is sophisticated I couldn't comprehend

[pliant creek]

oh I worry

veryyyy impressive

hi jolly holly

how are you

[elder cloak]

I need help @viscid moss

[pliant creek]

ı hope you fine

it's a dog

bak abi bentürküm fakat yabancı dil biliyorum

[elder cloak]

@pliant creek Appreciate the comment but this is the help section, you should chat in general.

[pliant creek]

ok

goodbye

[uneven parrot]

aga nası yapacaz öğretebilin mi

[pliant creek]

neyi

[uneven parrot]

şu ai sesi

ses değiştiriyoz ya ai le

[pliant creek]

muhtemel level atlayarak

[uneven parrot]

lvl atlamak ne alaka aga

sende mi bilmiyon

[pliant creek]

bak bol bol chatleş sonra level atlican sonra ai açılcak

[viscid moss]

I remember the problem you have, but I don't know how to solve it

[uneven parrot]

dayı kısa yolu yok mu

[pliant creek]

malesef

[viscid moss]

When I make a cover, it always converts my entire voice without leaving traces of the original voice

[pliant creek]

gir muhabbet et 1 saat yada 2 saat sürer enfazla