✨│ai-help

[gritty lily]

it doesnt let me

launch

when i type in

start_http.bat

[sand iris]

what does it say

can you screenshot it

[gritty lily]

ok

Mb idk how to screenshot

@sand iris it says that

Hm

[sand iris]

ok did you really do the same thing I did ?

[gritty lily]

Yeah

I typed in the search cmd

Right

[sand iris]

because that cmd isn't where voice-changer was extracted

[gritty lily]

Uhhhh

[sand iris]

I just don't have time to help at the moment

[gritty lily]

Ok

[cerulean breach]

how can i make my own voice model?

[tender trout]

what does rmvpe 900 epochs mean exactly? Which setting is that?

[serene citrus]

how long does it take (you know in average, or roughly) to convert vocals to your voice model in Ilaria RVC ?

[proper shale]

Depends on how long your audio is

~4 minutes took around 500-600 seconds for me

[proper shale]

RVMPE is a pitch extractor - epochs are a full cycle of the dataset being trained, so this means that the entire dataset has been trained 900 times.

[tender trout]

oh i see

[rare goblet]

Ayo? @tender trout level 1 !!!

[tender trout]

SO when Im putting in settings, I should change it to RVMPE?

I'm using mmvc

unless yall know a better one

[proper shale]

u can use other pitch detectors

like crepe tiny or something

but rvmpe is the fastest so go for it

[serene citrus]

yeah i waited for like 300 seconds, and it just hit me with a big fat nasty "error"

[rare goblet]

Ayo? @serene citrus level 2 !!!

[proven acorn]

Which one is the best Local RVC?

[proper shale]

Have you duplicated the space or nah?

this happens normally when you don't

[proper shale]

-local

[azure marsh]

Suggestions for @proven acorn

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 Huggingface

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

[proper shale]

Applio, Mangio or the Original RVC will work just fine

[proven acorn]

nevermind

same answer

thanks though

[proven acorn]

Which one would you recommded

Currently using the original rvc

[proper shale]

Original RVC/Mangio are less prone to errors so yeah I'd recommend those

[proven acorn]

Alright

Do you know how long it would take for it to render a 2 minute long audio file?

[proper shale]

Depends on your GPU really

[proven acorn]

Ah alright

Thanks

[proper shale]

No problem

[proven acorn]

I'm not sure if this is related to RVC but

[verbal abyss]

help

[proven acorn]

Is there a website to convert youtube videos to mp3

That's on this discord

[verbal abyss]

runtime disconnection

😭

[proper shale]

Not really, but let me recommend a good one - cobalt.tools

no ads, select any file type

[proper shale]

on what colab notebook

if it's rvc disconnected and you didn't export model, you lost your model

[verbal abyss]

easy gui

[rare goblet]

Ayo? @verbal abyss level 1 !!!

[verbal abyss]

i think

[proper shale]

oh should be saved on your drive then

[latent kettle]

Use RVCDISCONNECTED

[proper shale]

if you were even training a model

[proven acorn]

Thanks again

[proper shale]

and yeah rvc disconnected is slightly better

[verbal abyss]

i don't train i make covers

[proper shale]

np ;)

[proven acorn]

I can't wait to do an ai cover of smells like teen spirit and perhaps other nirvana songs

[proper shale]

oh

uh

[proven acorn]

that was quick

nose reaction

anyways

[proper shale]

our bot haha

[proven acorn]

thanks again again

what word does that react to

[proper shale]

smells

[proven acorn]

ah

thanks

again again again

[verbal abyss]

i followed the steps on that yt vid but it doesn't work

[proper shale]

hahaha np again

[proper shale]

old notebook, old tutorial

just use AICoverGen

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• 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

Credits to Faze Masta and Antasma for compiling these links.

[proper shale]

oop wt

wait

[proven acorn]

Is RVCbeta good?

[rare goblet]

Ayo? @proven acorn level 3 !!!

[proven acorn]

Because I got it in the same link as the original RVC

Sorry to ask again

[verbal abyss]

is it easy

[glad zealot]

gui ver doesnt show up sadge

[latent kettle]

Yes it is

[proper shale]

yeah i guess

https://colab.research.google.com/github/hinabl/AICoverGen-Colab/blob/main/Hina_Mod_AICoverGen_colab.ipynb

[proper shale]

right??? lol

will ask raytracer to add it

[glad zealot]

voice changer doesnt show up too

[proper shale]

yeah should be

[proven acorn]

Thanks again again again again

alright i will comeback if i have anymore questions

[verbal abyss]

so uh should i just execute everything step by step?

[glad zealot]

you can just do a run all

[latent kettle]

Just follow the steps and keep going

[ruby lake]

need help. i use kraggle version and i keep run into "No such file or directory: '/kaggle/working/voice-changer/server'
/kaggle/working

[glad zealot]

which version ar eyou using? the link

[proper shale]

probably very outdated

[ruby lake]

https://www.kaggle.com/code/rafacasari/wokada-voice-changer

[verbal abyss]

i already seperated the vocals on the song i'm gonna use so uhh should i just skip this part?

[glad zealot]

yup that doesnt work anymore

[ruby lake]

damn.

[glad zealot]

use thishttps://www.kaggle.com/code/hinabl/public-w-okada-voice-changer

it should work

last time i tried it at least.. which was last week

[ruby lake]

ohhh

[proper shale]

do it, you'll use the YouTube version of the song

[finite galleon]

You need the hubert base models

[proper shale]

aicg will separate the vocals n stuff

[finite galleon]

Run it

[glad zealot]

just run all the cell

[verbal abyss]

ok

[glad zealot]

dont skip nay

any

[verbal abyss]

ok is 2 minutes normal?

[rare goblet]

Ayo? @verbal abyss level 2 !!!

[proper shale]

yup

took me 3 to open yesterday, so ye

[verbal abyss]

ty

[glad zealot]

dont mind the timer

[verbal abyss]

will this take like a few mins again

[glad zealot]

yup

depends on your internet speed most of the time

[verbal abyss]

ok

[ruby lake]

holy shit it works thanks man.

[rare goblet]

Ayo? @ruby lake level 1 !!!

[glad zealot]

noice

it didnt do the bug on you

[last moat]

there is no japanese man's voice models?

[night stone]

I'm putting the index file with the .pth file in the "weights" directory, but RVC always says "Index not used". Am I putting it in the wrong place?

[rare goblet]

Ayo? @night stone level 1 !!!

[glad zealot]

index should be in the logs directory iirc

[rotund zephyr]

is there a thing in faq on how to make the voice less robotic?

in the colab version

[glad zealot]

https://rentry.co/W-Okada-FAQ

[night stone]

Ah, there it is. Thank you!

I've been testing settings for several hours now, and now matter what I do, it always has the same problem: The voice sounds like it's talking into a room fan. You know, like we did as kids. I can't find a way around it. Any tips?

[glad zealot]

different model

[rotund zephyr]

Also is there any way i can adjust the epochs?

or is epochs and chunk the same thing?

[proper shale]

No, unless you made the model

[proper shale]

No

[serene solstice]

this worth continuing or is it dead?

[azure shoal]

Using Applio, I tried to run the preprocess step on 293 audio files and i never got the "End Preprocess" message as stated by the Applio tutorial despite waiting for hours. Is this because 293 files are too much?

[lethal sand]

How to make a voice model?

[azure shoal]

https://discord.com/channels/1159260121998827560/1159636677178953802

[proper shale]

D seems fine but mel and G are rising

[proper shale]

how long are these a

[serene solstice]

everything is sorta back to falling?

[proper shale]

that was a collapse

lets see if it keeps training like that

[serene solstice]

Wish it was more obvious where things go wrong. This is like my 3rd attempt at this person and it keeps turning out robotic with a mel no lower than 20 😭. This is the most epochs I've gone for so far too.

[azure shoal]

probably 200 minutes in total, i just took like 120 minutes worth of audio files and the preprocessing finished correctly. trying again with more files and it's getting the problem again

[proper shale]

honestly i don't think you need 200 mins

120 mins would already be way too excessive actually

[rotund zephyr]

is 131k extra hella heavy to run but better output?

[proper shale]

anything after 32k would be eccessit

excessive

16k is good enough

extra uses way too much cpu

[rotund zephyr]

i think i got the order confused, i thought 4096 was the resource heavy one lmfao, thanks

[proper shale]

oh lol

np

if you have any more questions lmk

[rotund zephyr]

well, is there noise cancel on this thing?

[rare goblet]

Ayo? @rotund zephyr level 3 !!!

[rotund zephyr]

i mean the colab version

[proper shale]

yea use sup2

[rotund zephyr]

bruh i didnt even see it said noise:

lool

and also whats the recommended S thresh does it matter or nah

[proper shale]

it depends on how noisy ur place is

if it's a bit noisy you can increase it

[rotund zephyr]

all i have is a fan making noises so i think ill keep it default

[azure shoal]

thanks, does changing the sample rate to 48k affect anything?

[proper shale]

i meant extra on that ... anyway, 48k is okay if ur dataset allows you, but it brings in more noise and doesn't change much quality

[azure shoal]

noted. thank you!

[rare goblet]

Ayo? @azure shoal level 3 !!!

[proper shale]

np! ;)

[serene solstice]

what is the goal mel for a good model?

[proper shale]

as low as posst

possible*

btw did you try to train index?

[serene solstice]

me? I trained index before started training

[proper shale]

just wondering if you faced a "no-feature-todo" log in feature extraction

[serene solstice]

💀 any hope at this point or should I just start over

[rare goblet]

Ayo? @serene solstice level 3 !!!

[rotund zephyr]

yo how do i fix the voice stutter

[shell tundra]

My vc is stuttering and not working with a 4070 but my friends is working fine with a 3060 and we have the same settings and everything

[proper shale]

eh maybe you should trim silences to like 0.04

[serene solstice]

I did

I have about 18min across 7 files of 0.004 trimmed audio 😭

it's acapella singing dereverb deecho denoise

[thin pumice]

how to put the voice in Ilaria Converter

[shell tundra]

My stuff is tweaking I need help lollll

[rotund zephyr]

you guys have different gpus so maybe try to play around with the chunks and extras

also whats the recommended index if a model has one\

[thin pumice]

how to use AI Cover

[shell tundra]

That’s what I thought

But they are running the like 90-16k real time vc with no issue

Mine can’t even run like 200-16k

[shell tundra]

How you check that

[rotund zephyr]

well when u download a model its gonna have two files one with index and pth

[rotund zephyr]

well idk it might be your mic settings

[shell tundra]

Yea you put them both in and upload it

[shell tundra]

The mic affects that?

[rotund zephyr]

yeah, go to sound control panel, click your default mic and then properties

then go to advanced and make sure its not over 24bits

mine is 16 bits 48k hz

[shell tundra]

Where’s the sound control panel at again

[rotund zephyr]

look up sound settings on your search bar and it should be the first thing that pops up

and then go to top right and click sound control panel

[azure shoal]

I cut the files down into two sets of 55-60 minutes total, with the first set mean purely talking clips and the second set being singing clips. The second set still never reaches "Preprocessing finished"

why is this so hard 😭

[shell tundra]

Ok and change the vcable to 16 or the mic im using

[rotund zephyr]

i only changed my mic and it worked

[shell tundra]

Cause my mic is only 1 channel 16bit 48000 dvd quality

[rotund zephyr]

if it doesnt work then maybe try to change the cable too

[shell tundra]

The vcable is 24

[rotund zephyr]

is it fixed now

[full marsh]

if there any video enchancing colab....like fps or quality?

[shell tundra]

Lemme check

[rare goblet]

Vote Booster: Vote now for a 10% boost. https://arcane.bot/vote

[azure shoal]

Are the singing clips suppose to be less than a certain length? i just slightly reduced mine by removing sliences so all of them are like 3+minutes still

[shell tundra]

Nope doesn’t look like it’s

Is there a recommended ram? I have only 16gigs of ddr5 and whatever is with my gpu

[rotund zephyr]

nah

okay maybe atleast have like 6 gigs but thats about it

[shell tundra]

Bro then I don’t know and it’s really annoying

[rare goblet]

Ayo? @shell tundra level 4 !!!

[rotund zephyr]

if nothing really works i recommend just switching to the colab version tbh, its much better for my pc

[shell tundra]

I don’t know how to set it up

[rotund zephyr]

i will send you a link

[shell tundra]

Yea I tried to follow that but I got lost

Only was able to set up rvc thanks to some guy

[rotund zephyr]

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

its simple really

[fluid dove]

hi i need help when i try to get the juice wrld ai it gives me always this error what can i do :ModuleNotFoundError Traceback (most recent call last)
<ipython-input-1-d115e1f88205> 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.

[shell tundra]

Ong?

[rotund zephyr]

first step

you click that

actually

before you click that you gotta make sure you have T4 gpu runtime

[shell tundra]

What’s that

[rotund zephyr]

to check that just look at top left

till u see runtime

[rare goblet]

Ayo? @rotund zephyr level 4 !!!

[rotund zephyr]

and then click change runtime

do not change runtime if its already at python 3 and T4 gpu

[fluid dove]

THIS IS THE MODEL :https://huggingface.co/cheeseydipped5/Juicewrld999forlife/resolve/main/Juice WRLD.zip?download=true ?

[shell tundra]

Of what

[rotund zephyr]

are u using the colab version?

[fluid dove]

YES

[rotund zephyr]

the colab version, the link i sent u

[rotund zephyr]

you're probably using the wrong version, try this link https://colab.research.google.com/github/hinabl/voice-changer-colab/blob/master/Hina_Modified_Realtime_Voice_Changer_on_Colab.ipynb

[fluid dove]

thank you ill try

[rare goblet]

Ayo? @fluid dove level 1 !!!

[proper shale]

I'd trim both down to 30 mins ngl

[rotund zephyr]

also make sure u go to runtime and click disconnect and delete runtime

[proper shale]

You don't need that much data unless the person has so much variations

[shell tundra]

It’s on t4 gpu

Python 3

[rotund zephyr]

alright you should make a ngrok account

pretty simple just login using your google acc

its right here

but dont click play on any of these yety

if u did just close and reopen the link

[shell tundra]

Don’t click play on what I just went to make an account

[rotund zephyr]

ok once u make one go to your authtoken

its in left side

copy it then paste it on TOKEN:

and uh you should definitely make sure you're on servers closest to you cuz its probably gonna lag

if you're in the wrong regions

[azure shoal]

i did another test. Three singing clips at 15min, 4.5min, and 4 min respectively.

15min only finished within a few minutes
15min + 4.5min finished with a few minutes
All three is currently running and probably won't finish preprocessing (~23 minutes total length)

yet 55mins of pure talking clips finished within a few minutes, with the longest clip being 29 sec... i have ruled out the number of audio clips and the length of audio clips as causing issues.. not sure what to try next

[shell tundra]

Where’s token at

I have the authtoken don’t know where to paste it

[rotund zephyr]

right under the instruction

[shell tundra]

Oh ok

[shell tundra]

Clear console checked ?

Yea I did that

Should I press play button

[rotund zephyr]

not yet, press play on the first step

which is clone repository and install dependencies

[shell tundra]

Yea

Ok I’ll press play

[rotund zephyr]

should be done within like 3 mins

[shell tundra]

Ok I’ll be back then

[shell tundra]

It’s done

Now what

[rotund zephyr]

now run ngrok

[shell tundra]

Kk

Ok it says “server available at”

Then a link

[rotund zephyr]

click on it then visit site

[shell tundra]

I’m getting an error

Not secure says it sent an invalid response

[rotund zephyr]

can u send an ss

[rotund zephyr]

ah damn that does happen sometimes

[shell tundra]

Gotta redo it ?

[rotund zephyr]

maybe try to paste the link on a different browser?

looks like you're using explorer

[shell tundra]

Yea links from discord open to that

Same thing

[rotund zephyr]

oof

okay, redo it

close the colab tab

but this time you go to runtime

and press disconnect and delete runtime

and then do the same thing

[shell tundra]

Close the whole thing, bring it back and do all that

[rotund zephyr]

btw you can run the first step and then immediately run ngrok but u have to wait for both to finish

[rotund zephyr]

close the tab ye

[shell tundra]

Wait I have to download all that stuff again?

[rotund zephyr]

yeah it just takes a few mins tho

[shell tundra]

Ok

[rotund zephyr]

it doesnt save anyways i do the same thing when i boot it up again

[shell tundra]

Ok

Ok I did it ran the first step first then ngrok

Am waiting

[shell tundra]

Same thing

Can’t provide secure connection

[sand iris]

expected

there's a way to bypass that

[shell tundra]

How so

[sand iris]

I don't have a website on hand to give the exact steps

but if on the error page you click something along the lines of "learn more", there's then a button that allows you to proceed

[rotund zephyr]

oh damn i didnt even know that

[shell tundra]

That’s what I thought but there’s no learn more

[rotund zephyr]

also sorry for late response i was away

[shell tundra]

Is all it says

[shell tundra]

It’s cool

I just want this to work

Idk how my friends was working so good when my gpu is better and they didn’t have that cuda stuff

[rotund zephyr]

what did u download? was the one with direct?

directM

cuz thats for amd users

[shell tundra]

Nah it was from like a guide

[rare goblet]

Ayo? @shell tundra level 5 !!!

[shell tundra]

And I used the nvidia one

[rotund zephyr]

hm ok

[shell tundra]

https://rentry.co/VoiceChangerGuide#gpu-chart-for-known-working-chunkextra

this

[rotund zephyr]

well this is the only thing thats holding you back now, maybe try a different browser other than google and explorer, like opera gx or brave

[shell tundra]

i dont have those

lemme download them

[rotund zephyr]

alright

[shell tundra]

Same thing

[rotund zephyr]

maybe try brave, if that doesnt work then it may have something to do with firewall

[shell tundra]

Prob firewall

How we check

[sand iris]

wait no

[crimson sand]

Hey I'm trying to run Ilaria RVC locally and it's erroring about config/32k.json missing, any idea what I can do to fix this?

[sand iris]

Is your windows up to date ? And is your clock set to the right time ?

[shell tundra]

Yea

Times right

Should be up to date

But I do remember something wanting to update but I put remind in 3 days b

[rotund zephyr]

ooof maybe update that then

[shell tundra]

Nvm

Think we’re up to date

[rotund zephyr]

we might need a mods/helper for this one lol im only helping based on what i know/experienced

[shell tundra]

No worries

[rotund zephyr]

personally i never experienced that error so idk

[shell tundra]

Ok side but

Note

When I downloaded vcable and if I used the setup 64 instead of the regular setup would that matter

[rotund zephyr]

whats your current vcable setup?

is it 32bit???????

[shell tundra]

No I changed it 16

It was at 24

But that didn’t do anything

[rotund zephyr]

no i meant 34x

theres 34x and 64x when you download it

what did u download

32*

i meant 32 lol

32 and 64

[shell tundra]

Vcabledriverpack43

Is the file

I used the bottom one pretty sure

Setup64

[rotund zephyr]

dang

you vcable seems to be correctly setup

i thought you downloaded a x32 version somehow

[shell tundra]

Hm

Like when I have rvc in sight it works great, but when I start the game and have it running in the back it starts to get choppy

But I can still run the game and just have rvc over top the game and talk and it works perfectly just the moment it goes under tabs

[rotund zephyr]

thats really weird

[shell tundra]

If that makes sense

[rotund zephyr]

hmmm

[shell tundra]

My gpu usage while both the game and rvc is at 20%

[rotund zephyr]

do you have monitor set to none

[shell tundra]

My memory and cpu is a little higher but when I compared it to my friend who has a 3060 we had the same percentage usage but they were running theirs perfectly

[shell tundra]

No

[rotund zephyr]

ok so you set monitor with your default headphones?

monitors purpose is so that you can hear yourself

to see if the voice changer works

[shell tundra]

Yea

It’s set to default

My headphones

And I had it on when I was playing

[rotund zephyr]

also what about ingame? does the other person you're talking to ingame hear the same thing you're hearing?

[shell tundra]

If that’s bad

[rotund zephyr]

like if its choppy on your side is it choppy on their side as well?

[shell tundra]

They said it’s laggy as well

[rotund zephyr]

when you talk how much does the res go up to?

[shell tundra]

Uhm

Without running the game

I’ll check both

[rotund zephyr]

okay

[shell tundra]

Without the game running its 70-100

Ms

At 112 chunk - 16kextra

With the game running in a mid size lobby

It’s like 120-135 I’d say

BUT

That’s with the tab actively open on top the game

When I play the game for a bit and open rvc and look lemme check

[rotund zephyr]

what if its just running in the background

[azure shoal]

I am at a roadblock. Is it okay to use only talking clips to train a voice model for doing covers?

[shell tundra]

I’m opening it and it’s like 5000

[rotund zephyr]

how much does it go up to

[shell tundra]

Then it drops down to 120-135 when I keep it open actively

[rotund zephyr]

5k res sounds rlly bad damn

and idk why it works so good when actively open

[shell tundra]

Yea fr

Idk why bro

I wanna fix it

So bad

[crimson sand]

Sorry this is a stupid question but where do I upload pth files to mangio ui

[rotund zephyr]

i assumed you already restarted the app so idk anything else

[glad zealot]

in the weights folder

i think it should be in logs for mangio

logs>weights

[crimson sand]

yeah i tried both its not picking them up

oh

okay

[glad zealot]

make sure to hit the refresh list button or whatever it was

[crimson sand]

yeah i did

[rotund zephyr]

whats your f0 det

for nvidia i think crepe or harvest is goodf

[glad zealot]

rmvpe

use it

if its available

[rotund zephyr]

or rmvpe onnx yeah

[shell tundra]

It’s rnvpe

Onyx

[rotund zephyr]

his problem is that it works fine when the app is open ontop of the game hes playing but when its running in the background the res goes skyrocket like 5kms

[crimson sand]

yeah neither of these worked, the inferencing voice dropdown is empty

[shell tundra]

This is my friends setup but his doesn’t even stutter in games

[brittle wing]

Hey, does anyone know how the rvc works? I know you guys may not have the time to guide me through all of it, but I have a .wav file of multiple voice lines of the P5 character Arsene, and I need to convert it to a dataset. How does that work, given I could choose any of the rvc links of my choice?

[glad zealot]

pretty sure that will happen with any f0

[rotund zephyr]

wait didnt u say u had 4070

[shell tundra]

This is mine

[rotund zephyr]

oh ok

[shell tundra]

That’s my friends, who doesn’t have this issue

[rotund zephyr]

think you can provide a better ss? i cant see anything

[glad zealot]

is this local?

[shell tundra]

What’s that mean

[copper pebble]

jus put all those audios in a folder and zip it

[glad zealot]

running on your computer

[brittle wing]

So like, use the multiple files and NOT and the one where I combined them all into one clip?

[shell tundra]

Uh yea

[rare goblet]

Ayo? @shell tundra level 6 !!!

[shell tundra]

Pretty sure

[glad zealot]

either is fine

[shell tundra]

My bad

[copper pebble]

u could also use it

[brittle wing]

Ohhhh, yeah I'm trying out the RVC Disconnected.

[brittle wing]

Oh, so I can use that one, okay.

[copper pebble]

good choice

[copper pebble]

yh

[glad zealot]

mine is using lots but its gonna be the same since the program will cut it up anyways

[brittle wing]

Just running the testing stuff right now before I go in the actual link. The dependencies, the whole shebang.

[shell tundra]

I can’t screenshot on my pc it’s not working pretty sure, is this good enough

[glad zealot]

all settings seems to be fine ngl

[shell tundra]

Is this to me?

[copper pebble]

u following the guide right?

[glad zealot]

yes on taht screenshot.. or is that screenshot unrelated?

[brittle wing]

Wait, where is it? Should've realized that.

[shell tundra]

I never sent that

[copper pebble]

give me a sec ill link it here

[glad zealot]

kewk

[copper pebble]

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

heres it

[rotund zephyr]

nothing wrong here, i still dont know why it lags the moment you run it on background

[brittle wing]

Thanks!

[glad zealot]

prob a bad idea but try setting the voice changer gui to high performance in task manager

[copper pebble]

yw

[shell tundra]

I don’t know either, is it possible to pin it above the game and just make the window super small

[glad zealot]

high priority i mean

[rotund zephyr]

yeah try that

[shell tundra]

This

Ok I’ll try

[glad zealot]

yes but pretty sure you need 3rd party app for that

not really native to windows..

[latent cypress]

If I want to train my model to 300 epochs, can I do 300 epochs in 1 training? Or should I do 100 epochs per session to avoid overtraining?

[brittle wing]

loading rmvpe model
2023-12-11 16:24:11 | ERROR | root | Exception in callback _ProactorBasePipeTransport._call_connection_lost(None)
handle: <Handle _ProactorBasePipeTransport._call_connection_lost(None)>
Traceback (most recent call last):
File "asyncio\events.py", line 80, in _run
File "asyncio\proactor_events.py", line 162, in _call_connection_lost
ConnectionResetError: [WinError 10054] An existing connection was forcibly closed by the remote host
2023-12-11 16:26:05 | ERROR | root | Exception in callback _ProactorBasePipeTransport._call_connection_lost(None)
handle: <Handle _ProactorBasePipeTransport._call_connection_lost(None)>
Traceback (most recent call last):
File "asyncio\events.py", line 80, in _run
File "asyncio\proactor_events.py", line 162, in _call_connection_lost
ConnectionResetError: [WinError 10054] An existing connection was forcibly closed by the remote host
2023-12-11 16:26:06 | ERROR | root | Exception in callback _ProactorBasePipeTransport._call_connection_lost(None)
handle: <Handle _ProactorBasePipeTransport._call_connection_lost(None)>
Traceback (most recent call last):
File "asyncio\events.py", line 80, in _run
File "asyncio\proactor_events.py", line 162, in _call_connection_lost
ConnectionResetError: [WinError 10054] An existing connection was forcibly closed by the remote host
** uhh is this normal?**

[glad zealot]

nope try adding the model again

[latent cypress]

damn, my new message just dissapear

[shell tundra]

Yea idk if I can

Don’t see that anywhere

[glad zealot]

you can just leave it at 300

[glad zealot]

ye cause you need to get it from somewhere else

[glad zealot]

heres mine :D

[shell tundra]

Ok so I can’t

I put my gpu in max performance mode when I first did all this cause the guide said so

[glad zealot]

try searching for powertoys on the microsoft store

[shell tundra]

In my nvidia control panel

[crimson sand]

What is a feature search database and how do you generate one

it's not an index right

[latent cypress]

It won't be overtrained if I make 300 epochs in one training, right?

[glad zealot]

doesnt matter if you overtrain, you can just get the models and test them

[latent cypress]

oh okay, thanks

[glad zealot]

btw voice changer help in #🔍│help-w-okada

[shell tundra]

At this point I might just delete all the stuff and redo it

[hollow apex]

Hello,everyone.can I make AI Model in Mobile?

[glad zealot]

yes, by using colab

[brittle wing]

Hey everyone! I got my Arsene Voice Changer... thing... whatever you call it! Now, I need to test it, so what app do I use?

😄

However, I made the file private.

[brittle wing]

Is it this app on the picture @shell tundra uploaded?

[glad zealot]

@sand iris mew

[sand iris]

on it

thx

[shell tundra]

Yea I redownloaded everything and it still is broken

[rugged harbor]

im trying to use any type of voice, already set or custom and I get that error

[Voice Changer] VC PROCESSING EXCEPTION!!! CUDA error: the launch timed out and was terminated
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

CUDA error: the launch timed out and was terminated
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
Traceback (most recent call last):
File "voice_changer\VoiceChangerV2.py", line 238, in on_request
File "voice_changer\RVC\RVCr2.py", line 236, in inference
RuntimeError: CUDA error: the launch timed out and was terminated
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

[sand iris]

can you screenshot the window you're seeing the log from and DM it to me ?

There's usually more context before that issue

[rugged harbor]

alr

[uncut coral]

there is a google colab error that stops the voice charger

any solution ?

[glad zealot]

screenshot the error

[brittle wing]

Where can make models

i want to make a model for myself

[uncut coral]

there are two mistakes

[brittle wing]

Ahem

[uncut coral]

after using it 3 times

banned acount ngrok

solutions ?

[sand iris]

https://cdn.discordapp.com/emojis/1026989550628778005.gif?size=48&name=huhcat&quality=lossless

[uncut coral]

two error

Your runtime environment was taken offline because code that was not allowed on our free tier was executed. Colab subsidizes millions of users and prioritizes interactive programming sessions. In turn, it disables some types of use, as described in the FAQ. If you think it's a mistake, file an appeal. Include relevant information about the context of your use.
Your processing unit balance is 0. Buy more

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

[rare goblet]

Ayo? @uncut coral level 1 !!!

[proper shale]

old colab that got restricted

[uncut coral]

a ok

I already found a version of voice changer for processors and not graphics, thanks

[proper shale]

alright then

[proper shale]

uh

that still uses GPU

but sure

[uncut coral]

uses integrated GPU of processor in pc

[proper shale]

not really

just uses cpu if u don't have a gpu

[uncut coral]

ok thanks

[sand iris]

I mean, nvidia doesn't do integrated graphics so direcmtml is like the only way to run on those

[proper shale]

h

[azure shoal]

@proper shale

i did another test. Three singing clips at 15min, 4.5min, and 4 min respectively.

15min only finished within a few minutes
15min + 4.5min finished with a few minutes
All three is currently running and probably won't finish preprocessing (~23 minutes total length)

yet 55mins of pure talking clips finished within a few minutes, with the longest clip being 29 sec... i have ruled out the number of audio clips and the length of audio clips as causing issues.. not sure what to try next

based on what you just replied to me in #🧬│ai-chat , I think that i need to clean the singing clips more

[brittle wing]

can someone please give me the latest version of rvc?

mines outdated

[paper aurora]

can someone help idk what im doing and i cant convert audio

[oblique rain]

local?

[brittle wing]

yeah local

[oblique rain]

what are you using? colab? local? huggingface?

[paper aurora]

😭 idk

[oblique rain]

i mean, which site are you using right now

or program

[solar belfry]

I'm preprocessing my dataset on the RVC Easy collab, but it states that "Your dataset folder is empty", I never had this issue before. #@title 1.Preprocess Data
%cd /content/drive/MyDrive/project-main
model_name = 'voicemodel1' #@param {type:"string"}
#@markdown <small> Enter the path to your dataset folder (a folder with audios of the vocals you will train on), or if you want just upload the audios using the File Manager into the 'dataset' folder.
dataset_folder = '/content/drive/MyDrive/project-main/dataset/voicemodel1/' #@param {type:"string"}
while len(os.listdir(dataset_folder)) < 1:
input("Your dataset folder is empty.")
!mkdir -p ./logs/{model_name}
with open(f'./logs/{model_name}/preprocess.log','w') as f:
print("Starting...")
!python infer/modules/train/preprocess.py {dataset_folder} 40000 2 ./logs/{model_name} False 3.0 > /dev/null 2>&1
with open(f'./logs/{model_name}/preprocess.log','r') as f:
if 'end preprocess' in f.read():
clear_output()
display(Button(description="\u2714 Success", button_style="success"))
else:
print("Error preprocessing data... Make sure your dataset folder is correct.")

[paper aurora]

its localhost

[rare goblet]

Ayo? @paper aurora level 1 !!!

[paper aurora]

@oblique rain can we priv chat so i can send pictures

[oblique rain]

the latest update from mangio rvc is july 27 and applio september 17

[brittle wing]

how do i update?

[oblique rain]

just install from their .bats

[paper aurora]

?

[oblique rain]

alright

[brittle wing]

Guys, I did it! I downlaod RVC, did the stuff, and now I have an Arsene Voice Changer that sounds mostly accurate! Does anyone know if I can use this on Roblox, if so, how?

[brittle wing]

what one is better?

Hey, did anyone see my previous message?

[rare goblet]

Ayo? @brittle wing level 3 !!!

[brittle wing]

No? Ok.

I now got this link, so what do I do: https://vb-audio.com/Cable/

[ionic yew]

you download it and install it

[brittle wing]

Ok. Installed it, now I just got to restart my computer.

[ionic yew]

yup

[azure shoal]

were you able to find a solution? I have encountered the same issue

[brittle wing]

Restarted computer! What's next?

[crimson sand]

i just used applio instead

[oblique rain]

Mangio-RVC is the standard GUI for RVC, if you just want to inference or train models;

Applio is more personalized and custom, with more stuff in it.

[crimson sand]

mangio hurts my head to use

[ionic yew]

open rvc realtime

[shell tundra]

Anyone ever had issues of choppy rvc even tho your components aren’t even at full usage Or like when you have rvc running on top of a game it’s res is 100-130 but the moment you put it below the games tab it hits 5000 res

[brittle wing]

It's taking a bit.

Okie! I'm in!

In the RVC Client!

[ionic yew]

now check discord if the vb cable output is in the microphone settings

[brittle wing]

Yep! I turned it on!

👍

[ionic yew]

alright

now do you have voice models to change into?

[brittle wing]

Yep! Got the Arsene one I created! 🥳

[ionic yew]

do you hear the rvc voice filter

[brittle wing]

Oh, need to press start!

Yep! The filter is on!

[solar belfry]

does anyone know why rvc collab is telling me my dataset folder is empty even if it isn't?

[ionic yew]

nice! now you should be ready and for Roblox I guess make sure the VB Cable output is the default microphone or set as it

[brittle wing]

Okie dokie! We did it! Yay!

[proper shale]

so basically, the full 23 mins just won't work

is that what you mean?

[azure shoal]

Yes, there's something wrong still with the singing clips

[ionic yew]

your welcome

[proper shale]

well honestly I'm unsure

[azure shoal]

it's okay, i'll keep trying

[proper shale]

could be codec mismatching, basically the stuff is exported incorrectly

but audacity users don't really have a problem with it

and it doesn't affect preprocessing, instead affects feature extraction

¯_(ツ)_/¯

[copper pebble]

oh you did it really happy to hear that

[sand iris]

No

https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/tree/main
is

[full tiger]

is there like a website where i can make ai covers, google banned gui

[ionic yew]

Ilaria RVC I guess?

[molten pecan]

https://rentry.org/ilarvc_inf_guide my guide :b

[full tiger]

thank you julia

[molten pecan]

yw!

[full tiger]

do i have to download it?

[molten pecan]

Nope. It's through a website. You can even do it on the phone

[oblique rain]

by standard I mean the one that everyone use (i think) w/mangio-crepe in it, ofc that is the default one, maybe it is more update than the others

[wise escarp]

Why is illaria RVC being so slow? i used RVC magio-crepe (i think) before, and it was much faster

idk why its being slower rn

(i'm using RMVPE right now, and its veeery slow, even when it's meant to be faster)

[rare goblet]

Ayo? @wise escarp level 1 !!!

[wise escarp]

[proper shale]

Because Ilaria RVC uses the CPU whilst the old Colabs used the GPU

That's how Huggingface is

[wise escarp]

ooh, i see

can i still access the old collabs tho?

just so i can use GPU to get it going faster

[molten pecan]

Did you duplicate the space first?

[wise escarp]

ye, i did

[wise escarp]

i did it two times actually

[molten pecan]

Uhh why two times?

Once was enough

[wise escarp]

cuz i thought it would get better idk 😭

[molten pecan]

Are you sure you are using the duplicated space? Or just the original one?

[wise escarp]

ye its the duplicated one

[molten pecan]

How do you know?

Does the URL have your username?

[wise escarp]

ye it does

i can send you an screenshot

[molten pecan]

I see

[lyric imp]

hi, can i still convert a voice with google collab? if so, can anyone give me the link?

[solar belfry]

why does rvc keep telling me my dataset folder is empty when it's not

[rare goblet]

Ayo? @solar belfry level 1 !!!

[oblique rain]

check if you put the correct path

[brittle wing]

can anyone please help me regarding how to upload a model I made to hugging face to get a link for ilaria rvc tts use

[finite galleon]

-huggingface

[azure marsh]

Suggestions for @brittle wing

huggingface.co upload Guides

English guide

Link: https://rentry.org/fdg_guide_new
Credits: FDG

Spanish Guide

Link: https://rentry.org/AI_Hub_guide_esp
Credits: Julia (ailen2091)

Hindi guide

Link: https://rentry.org/AI_hub_guide_hi
Credits: Enes

Turkish guide

Link: https://rentry.org/AI_hub_guide_tr
Credits: Enes

[brittle wing]

To create a brand new model repository, visit huggingface.co/new. Then follow these steps:

In the “Files and versions” tab, select “Add File” and specify “Upload File”:

From there, select a file from your computer to upload and leave a helpful commit message to know what you are uploading:

Afterwards, click Commit changes to upload your model to the Hub!

Inspect files and history

You can check your repository with all the recently added files!

thank you so much

Anytime!

[rare goblet]

Ayo? @brittle wing level 1 !!!

[brittle wing]

Need Anymore help?

[solar belfry]

I have check if I put the correct path and nothing changed. It's so weird because I never had this issue before and I don't know what I did that was different. Could it be the length of the dataset?

[brittle wing]

Who can help implement the model in the Python script for $?

[zealous shuttle]

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• 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

Credits to Faze Masta and Antasma for compiling these links.

[undone hamlet]

@violet heron

[violet heron]

Does it say DirectML

[undone hamlet]

where would it say

@violet heron

[violet heron]

Like beside “Real Time Voice Changer Client Demo”

[undone hamlet]

nope

[violet heron]

No I meant scroll up

Where it says that in big

[undone hamlet]

where?

[violet heron]

On the voice changer

It should say that in big

[violet heron]

Yep you downloaded the wrong one

https://huggingface.co/wok000/vcclient000/resolve/main/MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.17b.zip

[undone hamlet]

can you link the right one?

[violet heron]

Delete it and use this one

[undone hamlet]

does the version matter?

[violet heron]

Yeah

[craggy sonnet]

I’m using ilaria RVC on iOS and why are all audio files greyed out like I can’t upload them I’ve followed step by step (sorry I’m new to this)

[hollow trellis]

Hey can someone help me . When my friend talk's there feed back goes into my mic how do i fix this

[solar belfry]

What microphone do you use?

[molten pecan]

In which folder is your audio at?

[hollow trellis]

BENGOO G9700 Gaming Headset

[molten pecan]

I think iOS doesn't allow apps the access of some folders

[hollow trellis]

mmm i just put on Sup2 on and i don't know if that fix it

[solar belfry]

The microphone might've picked up the audio from the headphones.

[hollow trellis]

mm so question's for you does this RVC also works on VR i am gussing

[rare goblet]

Ayo? @hollow trellis level 1 !!!

[solar belfry]

There probably is, I'm not too familiar with RVC VR stuff though

[ionic yew]

is there a way to stop the weird stutters in this version of rvc

like i managed to get it to work on my main pc

[craggy sonnet]

I tried it in many different folders and in no folders and everything is greyed out I’m using safari so I’m trying to see if there’s a setting that’s not letting me upload files but I can’t find anything

[ionic yew]

nvm i increase the sample length

[molten pecan]

You allowed the files permission, right?

[molten pecan]

Use rmvpe

[west sluice]

Guys how many epochs and recording time will be needed for a reliable and high quality model for TTS and Okada?

[rare goblet]

Ayo? @west sluice level 1 !!!

[proper shale]

quality should be the main focus

15-30 mins should be good

train until it overtrains ig

[solar belfry]

Have you tried this program called voice.ai?

[proper shale]

...voice.ai is even worse 💀

plus it's paid so

[west sluice]

@proper shale TYSM for the tips

[proper shale]

np ;)

[west sluice]

[solar belfry]

Yeah like I said I'm not really familiar with the real time ai voice thing and it really shows oof

[hollow trellis]

i used Voicemod but it kinda booo boooo

i feel like this one 100% a lot better

[hushed cove]

@violet heron i downloaded python 3.8.9 and im still getting the same message in the go-web cmd prompt

[violet heron]

Did you put in in runtime?

[violet heron]

Bad

Don’t use

[solar belfry]

duly noted

[glad zealot]

Set a bot to reply to anything that says "voiceai|voice.ai" with "no"

[brittle wing]

-voicechanger

[rare goblet]

Ayo? @brittle wing level 1 !!!

[brittle wing]

how do i get the voice changer

lol

[violet heron]

Okada?

[brittle wing]

yes

[violet heron]

AMD or NVidia

[brittle wing]

nvidia

i have a 970

[violet heron]

GTX 970?

[brittle wing]

yes

[violet heron]

Not powerful enough to run it locally

But you can use the colab

[brittle wing]

alr then

[violet heron]

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

[brittle wing]

im sure thats easy enough to set up

[violet heron]

If you don’t know what to do read it

[shell tundra]

Wait mine looks like that too, that’s the download from the guide link you guys gave me

It’s the wrong one?

[violet heron]

You said Nvidia GPU right?

[shell tundra]

Yep 4070 Ti

[shell tundra]

^

[violet heron]

Yep

Wrong one

[shell tundra]

But I’m having an issue where

Basically when I have rvc running on top of the game it works fine with like 100-130 res but the moment I full screen the game to actually yk, play and click back on rvc it’s like 5000res and it’s choppy and starts to come down when I pull up the tab again

My friends with 3060 and even a guy with a 1070 weren’t having this issue so I really doubt it’s my gpu being over stressed

[shell tundra]

No way

This can’t be

[violet heron]

You got the directml one which is for AMD

[shell tundra]

This is what mine looks like

It’s wrong? Just to make sure

[brittle wing]

anyone know how to fix this?

[proven acorn]

Do the Local RVC .bat files effect my internet/WiFi router?

[safe pike]

can anyone send me a doc or something of the sorts on how to do live rvc?

[rare goblet]

Ayo? @safe pike level 1 !!!

[bold sinew]

i require ... assistance ....

i cant send an image bc i dont think im allowed to but heres an imgur link to what im having trouble with

https://imgur.com/a/hTjhsZB

i just wanna make her sing last christmas !!!

[bold sinew]

i just did this rn but go to cmd and input this > pip install faiss-cpu

[magic locust]

hey on the ai voice changer i cant get it to work i cant set it to harvest would anyone know why
Image

[violet heron]

AMD or Nvidia GPU?

[magic locust]

nvidea

3060

#🔍│help-w-okada

[brittle wing]

why my res 50k

pls help me

[violet heron]

GPU?

[brittle wing]

rx 470

pls help me

@violet heron

[violet heron]

Only 4GB of vRAM

That’s probably not going to cut it

[brittle wing]

😭

[rare goblet]

Ayo? @brittle wing level 1 !!!

[brittle wing]

pls bro

help me

[violet heron]

I am

Are you using real time or cover?

[brittle wing]

real time

@violet heron

pls if you can help me

help me

pls

[violet heron]

Your GPU is probably not good enough

[brittle wing]

thank you

[rare goblet]

Ayo? @brittle wing level 4 !!!

[violet heron]

You can try using the colab instead

[brittle wing]

what is this

[violet heron]

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

[brittle wing]

what gpu do you recommend

[violet heron]

Read

[violet heron]

Usually more vRAM = better for ai stuff

[brittle wing]

im

8 vram

gb

[brittle wing]

so any gpu with 12+ vram?

[violet heron]

Oh

What’s your chunk at

And did you you download the DirectML version

It could be that

[brittle wing]

onnx directml

and the with no model

work good

[violet heron]

Ok

[brittle wing]

300 res

if i use

model

100k res

[violet heron]

Did you convert your model to onnx

[brittle wing]

oh thats was a problem ?

how i convert

and if you help me

i will realy be thanks

to you

realy realy

how do i download dotenv? i tried pip install dotenv and it shows error

[violet heron]

Convert to onnx button

[brittle wing]

ok i will do

but rx 470 8gb

work ?

[violet heron]

Probably

[brittle wing]

i click to start https

[rare goblet]

Ayo? @brittle wing level 2 !!!

[brittle wing]

to work

right ?

@violet heron

i convert

say thats Looks like there's a bit of a problem.

@violet heron

how do i install dotenv?

@violet heronbro

talk with me pls

[violet heron]

start http

[urban rapids]

Heyo, I uh am unsure what to do here. Trying to load the Chika Takami RVC but they're all .pth files in a weights section, what do I do if there's no ZIP?

Idk if I worded that properly but I can't send pics

[hushed cove]

@violet heron runtime?

[urban rapids]

There's like

rvc_models/weights and it says Chika.pth and Chika2.pth I honestly hoped they were in a zip already for the simple version of the RVC converter but they're not

[violet heron]

the folder called runtime in your folder

[hushed cove]

in the rvc mangio folder?

[rare goblet]

Ayo? @hushed cove level 2 !!!

[urban rapids]

Wait am I allowed to put links here

[rare goblet]

Ayo? @urban rapids level 1 !!!

[hushed cove]

what do i have to put in there?

[urban rapids]

Okay so if they're .pth files how do I turn em into ZIPs

[violet heron]

python

[hushed cove]

the installer?

[violet heron]

you send them to a zip file

[violet heron]

nope your python files

[hushed cove]

ok hold up

[urban rapids]

I see so can I just name it anything or does it have to be specific

[violet heron]

name of the zip doesntg matter

[hushed cove]

are they all in a folder in program files x86?

[hushed cove]

i dont see any python folder

[violet heron]

think so if thats where you installed it to

[hushed cove]

i dont think it gave me an option to choose my destination

[urban rapids]

Gotcha. One more question since I'm new to this, aside from the .pth files, must there be specific files with the .pth ones in the zip?

[hushed cove]

oh, it was in AppData the whole time

[hushed cove]

do i have to copy Python or Python38?

[violet heron]

i would copy both

[hushed cove]

alright i copied Python into runtime

Python38 is still in Python

[violet heron]

ok

i have to sleep

[hushed cove]

anything else or am i all set?

[verbal abyss]

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• 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

Credits to Faze Masta and Antasma for compiling these links.

[brittle wing]

@violet heron

[hushed cove]

hes sleeping now

[brittle wing]

Can anyone

help me?

[hushed cove]

with what?

[ember crow]

what is it

[brittle wing]

if i convert

to onnx

dont want

[ember crow]

oh onnx

[rare goblet]

Ayo? @ember crow level 37 !!!

[brittle wing]

yea

pls help me

@ember crow

[ember crow]

never did onnx

[brittle wing]

why ?

@ember crow Why my res

is

100k

[brittle wing]

anyone know how to fix this?

[night stone]

In your opinions, what is the best RVC fork for non-realtime audio file conversion?

[lucid peak]

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• 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

Credits to Faze Masta and Antasma for compiling these links.

[brittle wing]

how do i disable playback?

i dont wanna hear myself

[verbal abyss]

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• 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

Credits to Faze Masta and Antasma for compiling these links.

[short pecan]

tensorboard it spitting this error on startup and isnt running, how to fix?

[mental grove]

Ive trained a model how can i get a Feature Index, from a exisiting model?

[short pecan]

should just be the train feature index button after your done cooking your model

[latent kettle]

Can someone explain what is
1. model_architecture
2. target_sample_rate
3. speaker_id
4. crepe_hope_length
5. pitch_guidance
6. total_epochs
7. batch_size
In simple words so I can understand it easily

[proper shale]

1. whether you use rvc v1 or v2
2. sample rate of the final model and its outputs
3. honestly not sure about it but just don't mess with it lol
4. hop length is basically for how long it will analyze pitch variations, longer means it'll average out
5. pitch guidance means it'll use the same pitch as the output
6. total epochs you wanna train it to
7. how many samples it will train per epoch, lower means more training, higher means faster training

[brittle wing]

why if someone talk with me in dis the app replay it

[proper shale]

mic bleed?

[brittle wing]

?

[proper shale]

when the microphone gets the sound from a speaker

[brittle wing]

yea replay it

[proper shale]

if u have earbuds and aren't using them, please do

[rare goblet]

Ayo? @brittle wing level 3 !!!

[brittle wing]

i already do this

[proper shale]

send a screenshot of your input/output settings in W-Okada and Discord then

[brittle wing]

how i go to w-okada

[proper shale]

should be in ur browser

[meager terrace]

Any idea what is this problem on easy Gui?

[brittle wing]

how do i fix this problem

[proper shale]

modules weren't installed correctly?

honestly just use mangio because it installs everything correctly lmfao

[rare goblet]

Ayo? @proper shale level 56 !!!

[proper shale]

-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 Huggingface

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

[brittle wing]

how do i install that module

i tried to download it but it shows an error messages

[brittle wing]

shit taking too long to use and i cant download the audio for some reason

[fossil glen]

please help @red kayak my Dms already enable

[lean mesa]

shouldn't be at 6 from what i know

[red kayak]

let me see ur settings

[red kayak]

scroll down a lil

[red kayak]

whha?

try again ig?

[half raptor]

how to add url in GUi step 2 downloaded from discord?

[red kayak]

i really dont know why

[red kayak]

ohh ur trying to use qwq chan

yeah she is blocked from discord

the bot cant send dms

[fossil glen]

noooo

[red kayak]

cuz discord dumb

[red kayak]

[fossil glen]

how can i use restore system?

[fossil glen]

can i use other bot for restore system?

[red kayak]

which one

[fossil glen]

only qwq chan?

[red kayak]

yeah qwq chan cant send dms

[fossil glen]

any solution?

[red kayak]

no

we gotta wait for yui to get a new bot out

[fossil glen]

[red kayak]

we aint getting deleted any time soon

[tame mica]

unless

[fossil glen]

[brittle wing]

i want good girl vc

[violet heron]

-egirl

[azure marsh]

Suggestion for @brittle wing

Searching for the perfect RVC e-girl voice model?

Look no further – we've got you covered! Our RVC e-girl voicemodel is in high demand, and we've streamlined the process to make it effortlessly accessible for you.

[runic lynx]

whats this

wait

whered the picture go

https://cdn.discordapp.com/attachments/899459536635719690/1184151944554631248/image.png?ex=658aee70&is=65787970&hm=d4a276acf3639dcb81aab2233c79dcfed697183da37aaa4cb07bbc304237822e&

[violet heron]

https://tenor.com/view/jesus-ballin-mars-bars-gif-19910027

Which colab are you using

[runic lynx]

hina modified rvc

oh wait nvm