✨│ai-help

[west rampart]

thank you for your reply!

ended up using the deiteris one you linked im not sure if thats the current best. have a 4070 ti super, win 10, want to use a voice changer. figured it myself but unsure if its the optimal way. using rmvpe as the F0 detector all the others seemed worse lmk if thats right, 5 seconds extra and 100ms of chunk (any lower dips to yellow).

[low shard]

Are you looking to do e girl trolling/catfishsing or are you trans / doing roleplay?

[simple scaffold]

In the Cable Driver Installer, there was a Delete Driver button, so I did that. Or should I use Windows App?

[flint vapor]

Delete driver button, if is still there use then windows app

[alpine onyx]

No. I'm going to host a fighting game tournament soon so I was looking to do voice over for the trailer next to some male characters, since I couldn't find anybody that would be willing to do the female part. Since most of my female friends don't want their voices out in the open

[low shard]

Great, try the models in the suggested models section in the guide, try other models and play with the pitch

Also remember rvc isn't perfect

[simple scaffold]

Where exactly is the Cable driver located in Windows App? I'm from Russia, and everything is in Russian, so I'm trying to figure out where it is. I couldn't find the Cable in the regular Applications settings tab.

[low shard]

It can't do screams, laughing, or anything that isn't natural speech

[alpine onyx]

That I know I'm just moreso "frustrated" at the fact that the samples sound so good T-T

[flint vapor]

I don't know if is named different in russian, you can try from panel control, but if you just do delete drivers should be fine

[simple scaffold]

Okay, I'm going to restart my PC now, because it asked me to restart after uninstalling. I'll let you know how to install VAC Lite

[flint vapor]

Sure

[low shard]

You also may need some voice acting, play with the pitch, and just try till if you find a good one, there isn't anything much you can do because your settings are good

[alpine onyx]

Ok the recommended model to try on the guide is arguably the worst so far lmao

but yeah, stuff like laughs or breaths ( like a frustrated exhale ) sound super choppy and robot like

is there nothing else I could do in my settings to kinda... help with that ?

[flint vapor]

Yes they are not made for laugh and breaths, just for normal talking

[low shard]

There's nothing you can do about it, it's an RVC limitation,
Also why would you need that in the trailer 😭

[alpine onyx]

The character is the classical ojousama type character

HO~HO~HO

you know the drill

😭

[low shard]

Not really but would be best to change the script or either find a female voice actor tbh

[alpine onyx]

fair

[low shard]

Because no matter what you do, you will never get non natural sounds like laughing good

[alpine onyx]

How do other people where it sounds super crisp do it btw? Do they just feed a text script into it so the AI just reads it with the voice ?

was thinking about just doing that too

[low shard]

Wdym?

[woeful basin]

this is how it sounds now i cant hear a difference plz help

[flint vapor]

Is only 2 sec long i hear good

[alpine onyx]

Like some of the samples on weights.gg as an example or some of the samples in #1175430844685484042

[woeful basin]

i did cut it off eariler and i changed these

[low shard]

Like there's some crisp laugh ?

[flint vapor]

extra put 2.7 chunk you can use a bit lower is too high

[woeful basin]

its also really quiet

[flint vapor]

Can i see all the settings you have put?

If you can send a full screenshot

[alpine onyx]

nono them just speaking and it doesn't start cracking and going robot mode

[low shard]

That depends on how good is the model, do you have always crackling issues or only on discord? And also with EVERY model even when just normally talking?

[woeful basin]

i dont know what half of these even do

[alpine onyx]

Yes with every model even when just talkign normally. 1 word out of the entire sentence is being pronounced properly and everything else just crashes into a wall

[flint vapor]

But now which is your main error you want fix?

[low shard]

They are all explained in the guide btw, and you can also just hover your mouse over them

[alpine onyx]

ok I switched it to rmvpe_onnx and it's slightly better

[woeful basin]

it sounds odd i think and its quiet
this is also the one i downloadedhttps://www.weights.com/models/clm72ye2a16c0cctcphtywhxi

[flint vapor]

Is quiet because your monitor is on 100

[woeful basin]

so u can hear what it should be

[flint vapor]

Try increase

You are trying to listen yourself right?

[woeful basin]

hold on lemme open a game and see my average volume on my headphones\

[alpine onyx]

also @flint vapor is there a way not to hear the feedback from the realtime voice changer anymore?

[low shard]

Try to set extra to 2.0 in case it's cut off issues too, then test with models you know should sound good,

If not fixed, and the issue persists on discord, try https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#discord-crackle-fix

Or try https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#crackle-fix

[flint vapor]

Which feedback? You hear your own voice? or you don't want listen yourself?

[low shard]

You can just set the monitor to ur headphones to hear urself

[alpine onyx]

I don't want to hear my self via the realtime voice changer website

[woeful basin]

yeah the voice is quiet

[flint vapor]

So remove the monitor put none

Monitor option

[alpine onyx]

ah oki

thanks

[flint vapor]

Try play with volume settings

For input, output and monitor

Monitor if you listening yourself

[woeful basin]

its also a bit delayed

[pearl spruce]

Alguém sabe como fazer miniaturas de IA para o YouTube?

[flint vapor]

Is your chunk which chunk you have

[flint vapor]

English please

[pearl spruce]

Does anyone know how to make AI thumbnails for YouTube?

my bad

[flint vapor]

Is ok

[woeful basin]

its still this

[alpine onyx]

tried restarting and it doesn't wanna open anymore lmao

[flint vapor]

You can try use Sora of OpenAI other tools i really don't know😢

[simple scaffold]

So, I think I've installed everything correctly, but what's next?

[flint vapor]

Try put your chunks a bit lower i said before

[pearl spruce]

well can i sent u a example of a thumbnail and how you think he makes them?

[flint vapor]

i think you had the .exe on your process still active

Go on task manager

and see if is still open and close it

[flint vapor]

Do you have interface now of voice changer?

[woeful basin]

also is there a way to use this in games or on discord

[flint vapor]

Of course, if you installed vac lite

Just put output line 1 on voice changer

And in discord as input line 1

But what you using for if i can ask?

E girl troll? Catfish? Testing?

[simple scaffold]

Where can I get the newest version?

[flint vapor]

Nick send you the guide before

-wokada

[patent trellis]

💻 Local Realtime RVC

Guides for Programs that use RVC Models in Realtime for Calls/Games

• Wokada Deiteris Fork

Most suggested WebUI with the best general support for many platforms. GUIDE

• Vonovox

A Realtime Voice Changer with similar performance to Wokada Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. and without cloud options GUIDE

• Original Wokada

Not suggested nor maintained, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated

[flint vapor]

Follow the first guide

[low shard]

I already gave you it, wokada deiteris fork

[woeful basin]

no i got a miku voice changer and i wanna be miku on roblox with a miku character

as ive used voice changers like this before

with another program

[flint vapor]

Ok, try just regulate your chunks

Are so much high

Put a bit more lower

[low shard]

https://tenor.com/view/hatsune-miku-miku-hatsune-miku-hatsune-washing-machine-gif-4863029126409914383

[alpine onyx]

Can I use the gpu.batch file while the voice changer is active ?

[simple scaffold]

Here?

[flint vapor]

This is the guide

[alpine onyx]

or should I close it, enable it and than start the VC

[flint vapor]

Follow all

[low shard]

You have to read it, yes

[flint vapor]

Yes close it

And open it again

[flint vapor]

Don't click gpu clocks

[simple scaffold]

God, I'm such a вdumb. I don't know where to download it. XD

[flint vapor]

Just run the exe always

[woeful basin]

wdym

[flint vapor]

There is download link for which is your gpu

[flint vapor]

On left under

There are settings

for output put line 1

[alpine onyx]

nick was trying to make it so the crackling goes away and that was one of the suggestions to run force_gpu_clocks

[flint vapor]

Ohh ok sorry

I thought you want open voice changer

by going on there

[woeful basin]

wdym\

[flint vapor]

If you execute it as administrator?

[flint vapor]

Did you even installed vac lite? there is on the guide

[woeful basin]

what is vac lite

[flint vapor]

There is on the guide...

[alpine onyx]

that worked thanks

[hushed lotus]

Hello
Which sites are working for training?

[flint vapor]

Please follow every step

Of the guide

[woeful basin]

im sorry its hard to understand you

[flint vapor]

What you mean?

[flint vapor]

is ok

For this there is the guide

-wokada

[patent trellis]

💻 Local Realtime RVC

Guides for Programs that use RVC Models in Realtime for Calls/Games

• Wokada Deiteris Fork

Most suggested WebUI with the best general support for many platforms. GUIDE

• Vonovox

A Realtime Voice Changer with similar performance to Wokada Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. and without cloud options GUIDE

• Original Wokada

Not suggested nor maintained, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated

[flint vapor]

https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/

[hushed lotus]

I wanna train models
Last time I tried was a year ago I don’t know which website is working rn

[low shard]

This is a General AI Server, AI has many fields

Elaborate:

• your PC GPU
• your operating system
• what you want to do
• what tutorial link are you using
• a screenshot of the program

[alpine onyx]

both of these made it sound worse and even more choppy and barely picked up

[flint vapor]

You can't train models on website like that

[low shard]

What models? What's your PC GPU first? Cloud should never be your first option

[flint vapor]

Maybe you mean on cloud?

[flint vapor]

I didn't understand you

[woeful basin]

?

[hushed lotus]

Exactly
Sorry for bad English😭

[flint vapor]

Which is your gpu first?

If you have good gpu you can run it on locally

[flint vapor]

https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/

[hushed lotus]

I tried locally before
My laptop is not strong enough

[flint vapor]

https://docs.aihub.gg/rvc/cloud/applio-kaggle/

Follow this guide

[hushed lotus]

Alright thanks for helping🫶

[simple scaffold]

He opened the interface in the browser ._.

[flint vapor]

Yes it is correct

[simple scaffold]

Shouldn't it open separately from the browser?

[flint vapor]

No that was mainline outdated okada

[simple scaffold]

It's a pity, but okay

[hushed lotus]

Is there any other cloud?
It doesn’t support my phone number

[flint vapor]

Don't worry you will get comfortable

[flint vapor]

https://docs.aihub.gg/rvc/cloud/applio-no-ui-colab/

[hushed lotus]

🫶🫶

[simple scaffold]

Um, where's the Microphone?

There's no microphone or virtual cable.

[flint vapor]

Select

On input

Output

You will have options

[simple scaffold]

Yes, I've already found it, thank you)

All of this works really well, thank you guys!

[low shard]

input: microphone
output: line 1
monitor: headphones to hear urself optionally

on wokada deiteris fork, you can **optionally **use more advanced settings for benefits:

• Advanced Settings -> Force FP32 mode: on (THIS IS OFF BY DEFAULT! Turning this on improves stability. Increases VRAM usage by 200 MB)
• Advanced Settings -> Disable JIT compilation: off for faster loading speed of the program, on for slightly better performance (10-15 ms) for Nvidia only)
• Reduce the delay on Windows via: https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#reduce-more-delay

[low shard]

wtf could you please try restarting your pc?

[flint vapor]

You welcome 😄

[alpine onyx]

Will try later

[hushed lotus]

Hi guys
What should I do with this error?

An error occurred extracting the index: need at least one array to concatenate
If you are running this code in a virtual environment, make sure you have enough GPU available to generate the Index file.

[low shard]

This is a General AI Server, AI has many fields

Elaborate:

• your PC GPU
• your operating system
• what you want to do
• what tutorial link are you using
• a screenshot of the program

!give-media-perms 1h @hushed lotus

[hushed lotus]

I’m using applio cloud version

[low shard]

cloud shouldn't be your first option, could you please elaborate the other things I asked you?

cloud has limited gpu time and is more restricted, you should first check the things I told ya

[hushed lotus]

My gpu is weak I can’t use local version

[low shard]

could you please check?

You can check your pc gpu on Windows via:

ctrl+shift+esc (task manager) -> Performance tab -> GPU

I'm asking because we have seen users with a capable gpu saying the same thing, or thinking it worked only on nvidia which isn't true

[low shard]

holy shit, nvm 😭

i'm guessing you want to do training right? the issue is you didn't select the T4 gpu in colab, you have to do runtime at the top right -> change runtime type -> T4

also, colab isn't the best option to train, it has random gpu time of max 4 hours, I would honestly suggest Applio kaggle https://docs.aihub.gg/rvc/cloud/applio-kaggle/ which gives 30 hours weekly, but you need to verify your phone number (if you got any issues with that, you can contact kaggle support)

[hushed lotus]

I tried kaggle but it doesn’t support my damn number😭

Is there any other ways to use kaggle?

[low shard]

where are you from?

you could try contacting https://www.kaggle.com/contact

[hushed lotus]

Iran

Usually they don’t support my country

[low shard]

google uses a phone verification on kaggle to no thave alt accs, without it it wouldn't work

please try contacting kaggle and see what they say

[low shard]

atleast worth a try

[flint vapor]

I don't know if he can use also VoiP for change country number but i don't know if is legal

And it costs some money

Also

[hushed lotus]

Yeah I’ll try colab first
If It doesn’t work I’ll try kaggle anyway with some random number

[flint vapor]

it will ask you verification

For this i was talk about voip, but i don't know if can be used

[hushed lotus]

I can buy number tho

[hushed lotus]

My dataset is zip file
Is that okay?

[flint vapor]

Oh yes, voip, but i don't know if is legal to do for kaggle i'm not sure their guidelines

[hushed lotus]

Ah alright I’ll check

[low shard]

btw there's also an applio ui colab, the user interface might make it easier for you https://docs.aihub.gg/rvc/cloud/applio-colab/ but your choice

[hushed lotus]

Thanks man🤝

[low shard]

you're welcome

[viral mason]

what's voip?

[viral mason]

why are collabs still recommended when Kaggle is just better in general because of the longer time limit for the account and the infinite use loophole

for applio at least

[low shard]

#✨│ai-help message

he's from iran

the phone verification is an issue, it seems

[viral mason]

ah

that's unfortunate

kaggle should probably add email verification

[flint vapor]

Virtual Number

[low shard]

will 99% never happen, emails are easier to get for free for an alt acc

[flint vapor]

Should mean Voice over internet protocol, but in reality are just temporarily number for services, especially for make an account if your number is not compatible

[low shard]

the whole point about number verification is not having alts to abuse the service

[west rampart]

what F0 setting do yall use?

[viral mason]

really cool message

it's the only reason I can freely make models at all

but I'm not gonna let my secret stay here imma delete it and whoever saw it before can use it as well

[low shard]

that's normal, but you need phone verification on both

[viral mason]

oh it's an intended feature?

[low shard]

#✨│ai-help message for context

RMVPE is the best for most robustness and quality, CREPE does similar job (but worse) but is more lightweight

Also, if you send a screenshot of ur entire settings, I can help u

[low shard]

no it's not intended you're not supposed to have alts on kaggle and the second they find you will be prob IP banned

[viral mason]

I don't have alts lol

[west rampart]

thought so

what about onnx?

[viral mason]

bad version

[low shard]

onnx is meant mostly for non nvidia gpus performance

[low shard]

no it's not bad

just meant for non nvidia 😭

[viral mason]

yes

bad

[west rampart]

this guy gets it

[low shard]

nope, bc not everyone has an nvidia gpu, like you

[low shard]

btw, want me to also check ur other settings ?

[viral mason]

it's the mid version of the thing that's better

[low shard]

i can also suggest optional ones

[west rampart]

cant attach images 💔

[viral mason]

I think bro needs image perms

[low shard]

!give-media-persm 1h @west rampart

now ?

smh typo

!give-media-perms 1h @west rampart

[west rampart]

ty

[viral mason]

"persm"

[viral mason]

why do u have two virtual cables?

[low shard]

uninstall vb audio cable from windows app settings, it gives issues such as crackling and randomly stop working on windows

get vac lite from the guide

also, extra kinda controls quality, but values over 2.7 can risk cutoff issues I would suggest lowering it to 2.7 if you don't want to risk cutoff issues at the end of your words, especially with some models

[west rampart]

wanted to route audio to process it

[low shard]

wdym? the input should be ur microphone, output should be line 1

[west rampart]

ill grab vac lite

[low shard]

are you using other post processing effects? we removed the realism page as it used to give various issues on users, saying if you followed that

[west rampart]

just voicemeeter stuff nothing major

was thinking about routing it through my daw tho

[low shard]

we had a page dedicated to that, but worked only for 1 singular (ex staff) person, everybody else had issues with noises and other stuff after it 😭

anyways heres vac lite https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#virtual-audio-cable

[viral mason]

I have a friend who got it to work but I couldn't

[west rampart]

yea would prob need 2 virtual cables

[viral mason]

putting his okada voice through a daw like fl studio

[west rampart]

idk why they all only have like one that come with it

[viral mason]

?

[low shard]

on wokada deiteris fork, you can **optionally **use more advanced settings for benefits:

• Advanced Settings -> Force FP32 mode: on (THIS IS OFF BY DEFAULT! Turning this on improves stability. Increases VRAM usage by 200 MB)
• Advanced Settings -> Disable JIT compilation: off for faster loading speed of the program, on for slightly better performance (10-15 ms) for Nvidia only)
• Reduce the delay on Windows via: https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#reduce-more-delay

@west rampart do you need any other help ?

[west rampart]

should be good for now just trying to get it to sound non robotic really, still a bit off

thanks for letting me know about fp32 thought that was better off

[low shard]

try playing with the pitch and different models

yw

[west rampart]

yea that made it slightly better but pre off

[viral mason]

don't forget to put effort when you're speaking because speaking like a robot will make it sound like a robot

[west rampart]

bahha true yea

I do talk super monotone which is normally awesome but with this it doesnt sound the best 🥀

[viral mason]

yea with some characters you gotta be really expressive

[bronze plume]

how do i run the MMVCServerSIO on linux ( im on mint )

[low shard]

what's your pc gpu? and what you looking to do?

[bronze plume]

well im trying too be trans because my voice doesnt sound like a actual female and my cpu is a Intel N200 (4) @ 3.700GHz

my gpu is a Intel Alder Lake-N [UHD Graphics

yes i know its bad

lmao

[low shard]

Ohh, I'm guessing Wokada Deiteris Fork then, the realtime voice changer, just asked as some confuse the programs

[bronze plume]

im on linux btw

[low shard]

yep it's bad, integrated graphics, won't even be recognized

[bronze plume]

oh

[low shard]

you sure you don't have any other dedicated gpu?

[bronze plume]

no i can show you my specs

can i get temp image perms?

so i can send the image

[low shard]

unfortunately AI requires especially a good GPU, since this is Open Source and runs on your hardware

!give-media-perms 1h @bronze plume

[low shard]

yeah that's cooked simply, even 4gb of ram, it won't run anything locally

Your pc is too Weak for Wokada locally, You got 3 options:

• Buy a better pc
• Run it locally (on ur pc) using the CPU mode of the wokada fork which has better performance https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/ (but this isn't suggested as it could be unstable, especially on your PC)
• Use **cloud **(remote good pc):

About Cloud, there are different services:

• Google Colabs (4 hours daily of free T4 gpu, easy to use, require only a google account) :
  • W-Okada's Deiteris' Fork Voice Changer Google Colab (currently works only on google colab PAID tier)
• Kaggles (30 hours weekly of better GPUs, T4x2 & P100, harder to use, requires an account and a phone number):
  • W-Okada's Deiteris' Fork Voice Changer Kaggle (the best and only working one currently for free)

[bronze plume]

well im fine if it is unstable lol

[low shard]

with unstable I mean it could have over 4k of ping, I'm not sure but I really really really doubt it will be good enough to be any usable

[bronze plume]

oh

[low shard]

Im 100% sure for any game, if you really want you could try to use it on CPU on discord, but I have a feeling it will just be wasted time, but if you want give it a shot, I can't guarantee you it will be good

[bronze plume]

i have the cpu version

[low shard]

did you download the vac lite too?

[bronze plume]

what

idk what that is

is it on linux?

[low shard]

are you reading https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#virtual-audio-cable ?

what guide are you using ?

[bronze plume]

i just installed the cpu version

i didnt read anything

[low shard]

could you please link what did you download ?

[bronze plume]

https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/

[low shard]

alright, did you do the https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#linux step? and used portaudio?

[bronze plume]

no

i just clicked that link

and downloaded the cpu

thats all i did

i dont have nvidia or amd

[low shard]

this one? https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#download-amd-intel-and-cpu-on-windows this should be only for windows

btw are you really sure you want to do this? since it will be just wasted time, wokada deiteris fork kaggle cloud will give you 30 hours weekly of better gpu in remote for free

you don't even meet the RAM requirement of 6gb

[bronze plume]

yk what nevermind im getting really confused

its fine

[low shard]

I would really suggest you to forget what you downloaded, and just read the kaggle cloud version: https://docs.aihub.gg/rvc-voice-changer/cloud/deiteris-w-okada-fork-kaggle/

it's on cloud (remote good pc), and got 30 hours weekly for free

[flint vapor]

he/she went away...

[brittle wing]

How I get model maker role?

[low shard]

https://docs.aihub.gg/extra/model-maker-role/

[potent bone]

how do i put background noise the page got deleted from ai docs

[low shard]

You mean the realism page? Don't follow it, every single user reported issues with the post processing effects
Is there a specific reason why you need them?

[potent bone]

nah

[low shard]

Are you trying to do e girl trolling/catfishsing? Or did you need them for roleplay? Or idk

[potent bone]

i prank people i know

[low shard]

So you're doing e girl trolling/catfishing to people you know ?

[potent bone]

i just troll them

[bleak smelt]

What site (no colabs or manual training) would you guys recommend is the best for voice (singing) training?

[low shard]

Yeah, I asked if you were doing e girl trolling/catfishing

[low shard]

It would be better you first check your PC GPU first to see if it's good enough

[bleak smelt]

Hm? I'm trying to find a site that will do it instead of my GPU

[low shard]

Yeah but it would be better to First check if you GPU is good enough, because local got no limits

You sure you don't care about that and just want a site?

[bleak smelt]

Yeah I'm sure, I know how to train via colabs, but I'd rather just upload dataset and have it done for me since I'm pretty busy atm

[low shard]

Try https://weights.com/models , it does exactly what you want

I thought at first you confused with other programs, since it wouldn't be the first time people confuse RVC with Realtime Voice Changer

[low shard]

This is a General AI Server, AI has many fields

Elaborate:

• your PC GPU
• your operating system
• what you want to do
• what tutorial link are you using
• a screenshot of the program

[chilly nacelle]

why am i server muted

[viscid moss]

What models are you looking for?

[craggy idol]

hey everyone, im trying to find a tutorial on how to create a ai model, but i cant, can someone help me plzz?

[viscid moss]

-rvc

[patent trellis]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[viscid moss]

@craggy idol

[craggy idol]

thank youu!!

[woeful basin]

can someone help im using the voice changer and something is wrong, i dont know how to describe it but its kinda quiet
example (im using a miku voice changer fyi)

"hello, so close, almost working, something is off" is what i said too

pls help

[proven mortar]

Commands

Our website contains lots of resources to get you going in our bot!
Try using slash commands (/) for a faster experience!

pls achievements
-# View your progress towards achievements to flex and earn rewards.

pls advancements
-# Includes prestige, omega, upgrades, level rewards, and vote rewards.

pls adventure
-# Exchange adventure tickets to go on exciting adventures for unique rewards!

pls badges
-# View and manage your badges.

pls balance
-# See someone's balance. Pocket, bank, net worth, and more.

pls beg
-# Beg for coins to help bolster your pocket balance.

Page 1 of 9

[proven mortar]

@woeful basin

You have an unread alert!

Use </alert:1011560370864930852> to read it!

[woeful basin]

bro

[woeful basin]

can someone help

[west cave]

is it just me or is there not a single ai that can say "an error occured" correctly without extreme emphasis from the original voice

it always comes out as gibberish

[long obsidian]

hi in comfyui im getting this clip missing: ['text_projection.weight'] which really slows my generation does anyone know how to fix it?

[west cave]

https://tenor.com/view/mods-discord-mod-moderator-moderation-clash-of-clans-gif-24080525

epic embed fail

[cobalt turtle]

hello, can someone help me please, I am using the w-okada voice changer. it is running but, when I am in a discord call the voice of the voice changer doesn't go though. Why and can someone please help me

[onyx badge]

whats index used for ?

[west cave]

are you using vb virtual audio cable

[cobalt turtle]

what is that?

[west cave]

index changes the sound of the voice to more closely align with the accent of the speaker

[onyx badge]

so at 0 its disabled

and at 1 its most like the accent ?

[west cave]

the voice changer does not modify your INPUT mic, it only OUTPUTS a changed sound. you're going to need a virtual audio cable to use that. google how to use either vb virtual audio cable (easiest) or voicemeeter banana (harder) with a voice changer

[west cave]

yes

[cobalt turtle]

ahhhh thank you thank you

[woeful basin]

can someone help why does it sound like it has voice cracks constantly, it also has always been sounding weird
voice is a miku voice

oh yeah this is the settings

ive been spending all day trying to make this work 😭

[hushed thicket]

the voice changer stops picking up audio after I stop talking for a while how do I fix this?

[maiden wren]

yoo wsg

[hallow kite]

Maybe turn off noise suppression and echo

[hallow kite]

Try turning off noise suppression and lowering the extra to about 2 seconds, there's a chance the model just isn't good

maybe change the f0 det., i'm no expert but that's fixed issues for me

[woeful basin]

what do i make the f0 det

@hallow kite i did what you said and i made the f0 det rmvpe_onnx

its kinda fixed but its picking up backround noises i cant even hear

and still sounds different to what the examples should from where i got it

-# model im using if u wanna know or smth https://www.weights.com/models/clm72ye2a16c0cctcphtywhxi

[pastel oak]

The examples on pages are usually with RVC inferencing and not realtime inferencing, thats why they will always sound different.

The audio you sent sounds very weird. Can you check a few things please:

• what number is the "perf" on the graph when you speak?
• use audio: SERVER mode with
  S.r. 48000
  Input: [windows wasapi] microphone
  Output: [windows wasapi] line 1
  Monitor (leave this empty for now to see if this fixes the echoing I hear in your audio)

[pastel oak]

It is possible your gpu goes idle. Are you sure it stops picking up audio or does it "start up" with voices cutting out?

Which version of the voice changer do you have + whats your gpu

[woeful basin]

i might have changed something here is my settings as of rn
the perf is like 150-200 and il do the server audio, what does it do though

soemtimes the perf is like 96

[pastel oak]

Wasapi uses better audio processing on your pc. Client uses mme by default which is old

Perf is your gpus performance. With amd theres a bug so i hope everytime you change chunk or extra you also change gpu to cpu then back to gpu to "reset"

The lower the better

[woeful basin]

yeah i saw i had to go to cpu and amd i do that each time

i got a white ⚠️ thingy near it

[pastel oak]

Can you use a different virtual cable than voicemod? That usually causes doubles and echoes since it routes your audio weirdly and goes through it multiple times

[woeful basin]

i dont have any others

voicemod is only there because i use voicemod

[pastel oak]

Download vac from the guide

-rt

[patent trellis]

💻 Local Realtime RVC

Guides for Programs that use RVC Models in Realtime for Calls/Games

• Wokada Deiteris Fork

Most suggested WebUI with the best general support for many platforms. GUIDE

• Vonovox

A Realtime Voice Changer with similar performance to Wokada Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. and without cloud options GUIDE

• Original Wokada

Not suggested nor maintained, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated

[pastel oak]

1st link

[woeful basin]

Wokada Deiteris Fork
Most suggested WebUI with the best general support for many platforms. GUIDE
this one?

[pastel oak]

Yes that is the first link and the voice changer you have ;D

[woeful basin]

ok il get it il be back

uh what do i open

setup, setup64 or setup64a?

or something else

@pastel oak

[pastel oak]

It says in the guide but izs setup64

[woeful basin]

oh im blind sorry

@pastel oak (sorry for anther ping) i got it and i pressed start on the voice changer and this showed up

[pastel oak]

Did you select a voice model?

[woeful basin]

pretty sure miku was still equipped

does it unselect when u go to server

[pastel oak]

Shouldnt, but input output could be resetted

Screenshot before clicking start

If it happens again, send the console window error

[woeful basin]

wha do i do to fix this screen

[pastel oak]

Close everything and restart if initialize doesnt work

[woeful basin]

i cant hear anything @pastel oak but i did press start

ajnd i dont think theres a record option on nserver

[pastel oak]

Did you open up sound devices to change back your defaults

After installing vac

[woeful basin]

yes

[open pelican]

Help

[pastel oak]

Send screenshot of voice changer

[pastel oak]

Select line 1 on discord input and then use the self hearing test

If it picks up and if it sounds normal etc

[woeful basin]

i heard something but its wayy to quiet

[open pelican]

@pastel oak, Help Me

I Wanna To Make A Ai Maeko Voice In Weights

[woeful basin]

nvm

nvm again its still quiet

[open pelican]

I Wanna To Make Ai Maeko Voice

[woeful basin]

dont ask me

[pastel oak]

Weights has everything you need on their website

https://www.weights.com

[woeful basin]

do i change these if its quiet and if so what do i chnage

[pastel oak]

Output is the voice changer volume

[woeful basin]

so do i make that higher

[pastel oak]

Yes

[open pelican]

I Deleted To My Account @pastel oak

[pastel oak]

Make a new account

[woeful basin]

i did the discord test and it sounds pretty bad how do i show u

[open pelican]

Okay, Where Is Sign Up @pastel oak

[hushed thicket]

I have the modified version and a 4070

[pastel oak]

Record with shadowplay or audacity

Though make sure you turn off echo cancellation and noise suppression on discord and test again

[woeful basin]

i have obs recorder

i dont have those

[pastel oak]

On the website I linked, click on "lets get started" or whatever it says in your language, then it asks you to sign up

[open pelican]

Link To Website

[pastel oak]

Anything that records works

https://www.weights.com/

[pastel oak]

Send a screenshot of your interface

[pastel oak]

You have a bad virtual cable known for issues. If you dont have any issues you can keep it but most users have issues

F0 det rmvpe

Audio: server
S.r. 48000
Input: [windows wasapi] yeti mic
Output: [windows wasapi] virtual cable
Monitor [windows wasapi] speakers

The rest seems fine. Can you go outside the mmvcserversio folder and run "force gpu clocks.bat" and see if the issue persists?
You can turn it off at the end of your session by running the other bat file

[open pelican]

@pastel oak, My Account And New Account Is Got Banned

[pastel oak]

What country are you from

[open pelican]

English

[pastel oak]

Aha

How old are you

[open pelican]

I'm 18

[pastel oak]

Which country do you live in

[open pelican]

U.s.a

[golden walrus]

Question, is it okay to increase the output volume by 4db

[pastel oak]

Cant help you pinpoint the issue if you keep lying

[golden walrus]

Does it result in cracking and robotic voice?

[open pelican]

Look To Problem

[pastel oak]

Its a matter of testing yourself. Nothing is wrong or right, some models are naturally quieter and need increasing

[maiden wren]

Instead of cable in 16ch choose cable input vb-audio virtual cable

[golden walrus]

i mean, i test it already, and i have some problem with hearing so i have no idea if it is okay

[open pelican]

I Want To Try Realtime voice changer client Voice Ai

[pastel oak]

Try different email

[maiden wren]

isnt that funky friday

[pastel oak]

Weights has nothing to do with using realtime voice changer. So do you want to train a voice or use voice changer

[open pelican]

I Deleted Account Too Recently To Sign Up Again

[maiden wren]

create new email

[pastel oak]

You should be fine shouldnt crack

[golden walrus]

If i were you, i would go to #1159513888199540817 and read first

[golden walrus]

roger that, sir

[maiden wren]

its so easy to set up are people that new to pc

[woeful basin]

i dont know if u can even hear anything

[golden walrus]

No, just they haven't used it correctly for long time

[maiden wren]

i got app and every settings ready to best in like 5 min

[pastel oak]

Why dont you simply select line 1 as your input on obs and talk

Lol

[maiden wren]

fixed errors easily myself too its easy if you understand how pc works.

[woeful basin]

huh

[pastel oak]

In audio mixer

Make a new entry for microphone

Use line 1 as input

Hit record and talk

[woeful basin]

how do i do that

[pastel oak]

You are the one using obs not me 😭

You got some stuff in there

[woeful basin]

and im really bad at using obs i only know how to screen record

[maiden wren]

bruh

[pastel oak]

Just use audio client and use the record tool

On voice changer

[woeful basin]

also isnt moniter suppose to be the one playing back ur voice

[maiden wren]

choose your speaker

[pastel oak]

You can use it to play back your voice in your speakers yeah, ig just do it that way and screen record your audio

[knotty moth]

I haven't read your convos allat

but have you tried some other models, besides fixing your settings?

[woeful basin]

ok it sounds good but idk if this is normal

[pastel oak]

Put chunk to 250 btw

It sounds ok, sounds like how youd expect a japanese model to sound like when you speak english

[woeful basin]

quiet

and still sounds weird

[knotty moth]

I've told him a few times to try another model

before comparing to the current miku model

[woeful basin]

i could try that lol

hold on

but uh why is it quiet

btw

[open pelican]

And Take It To Picture New Maeko

[woeful basin]

i got a different one how to i make it not quiet like this

@pastel oak @knotty moth sorry for ping idk who to ping

[open pelican]

Make To New Maeko Ai Voice In Weights.gg, Please @pastel oak

[pastel oak]

Shits loud af for me

[woeful basin]

huh

what

how

im on 42 volume so it cant be that right

[pastel oak]

What are these 💀

[pastel oak]

They are loud enough

[woeful basin]

wait whats ur volume

[pastel oak]

I dont use volume like u do

I sent you a rough waveform to show you its loud enough

[woeful basin]

then im confused why its quiet for me

[pastel oak]

Either way if you use it in a call with a friend it will be loud enough

Overall the issue you mentioned is fixed, voice models are not all perfect so some might sound more robotic than others like this one

You can move in. sens. to the right if you notice more background noise getting picked up

[knotty moth]

what are you doing with the fnf brainrot in this help channel?

[open pelican]

That's Not Brainrot, Okay?

[open pelican]

Is A Voice Maeko And Maeko Profile @pastel oak

[pastel oak]

https://www.weights.com/models

You can train it with the audio you sent

[open pelican]

You Try @pastel oak, But I Don't Know, There Is Not Maeko

[pastel oak]

There is no Maeko voice. You said you want to train a voice. I sent you a screenshot how to train the voice

[open pelican]

Make To Maeko Ai Voice, (3 Maeko Voice Line And Picture Maeko), Please @pastel oak

[craggy bough]

theres no way ur british or american

[knotty moth]

this ain't a prompt to anyone for making the voice model

[pastel oak]

south east asian, itd be so much easier if he explained it in his language at this point

[open pelican]

Is Maeko Ai Voice

[craggy bough]

🥀

[knotty moth]

go make the Maeko voice model by your own using weights.com

instead of keep prompting it to anyone

not even chatGPT or any sort of agent AIs could do it for you lol

[open pelican]

I Deleted To Maeko Voice, I So Sorry @knotty moth

Done

Look @knotty moth

[oak edge]

I'll tryyy thankss

[sick plover]

Can someone tell me if there are any robotic voices on the server?

[low shard]

You can search rvc ai voice models at:

• https://discord.com/channels/1159260121998827560/1175430844685484042
• In https://discord.com/channels/1159260121998827560/1163592055830880266 , Do /find with @earnest musk
• https://weights.com/ (login required)
• https://huggingface.co/models (but watch out cus in hugging face there arent only rvc ai voice models)
• Suggested Models for Realtime Voice Changing (Wokada)
• https://voice-models.com/
• https://thevoicemodels.com/ (for Turkish Models, login required with discord and level 2 on their server)

if there isnt one, you can:

• make it yourself with our docs guides
• Ask a free request in https://discord.com/channels/1159260121998827560/1159289738314919936
• Be aware that we don't allow any paid comms

I think by just searching robot you should find them

[pastel oak]

@hushed thicket did your issue get fixed?

[hushed thicket]

i followed everything but i think its thje model voice

[shadow flax]

Helloo,
i was considering running a LLM locally and try some fine tuning because i found it interesting, what could i run with 16GB of vram? (optimally no quant because i heard it takes longer per token)

[hushed thicket]

cuz i tried other models and it didnt take forever to start up

[latent kettle]

Have you tried deepseek r1 ?

[shadow flax]

nope '-'

havent tried anythin locally yet

js comfy for image

[latent kettle]

Go ahead

[shadow flax]

is it good?

btw

i saw there exist tierlists for local models, any chance you could refer me to something like that which is reputable

i also heard llama 3 8b is good

[low shard]

#📰│dev-updates message

[latent kettle]

All the models with high parameters, are more resource demanding

[open pelican]

Guys

Look

Can You Guys Make Maeko Ai Voice, Please

[flint vapor]

Hi, this is not a channel where you can request model, please ask it in #1159289738314919936

[open pelican]

Yea, But Make To Maeko Ai Voice

[flint vapor]

You have to request it in that channel, here is only for help

and troubleshooting

[open pelican]

Here You Go https://discord.com/channels/1159260121998827560/1396136746294444133

[proven hill]

stop asking here

[open pelican]

Click To New Maeko

[proven hill]

what

[open pelican]

Maeko Ai Voice

[proven hill]

do it

[open pelican]

Okay, But Where?

[proven hill]

idk, applio

[open pelican]

Applio, What Is Applio

[proven hill]

program

[open pelican]

What Is For Ai Voice

[proven hill]

yes

[open pelican]

Can You Make A Maeko Ai Voice

[proven hill]

why

give me one good reason

[open pelican]

Because, I Love Maeko Ai Voice

[proven hill]

are you 12

[open pelican]

I'm 18

[proven hill]

why do you speak like that

[open pelican]

Is My Girlfriend New Maeko

[proven hill]

shes... not real man...

[shadow flax]

what parameters does r1 have?
i heard low parameter models tend to give worse results/hallucinate

[open pelican]

Look @proven hill

[proven hill]

so you DO HAVE the ai voice

[open pelican]

Yes

[proven hill]

then what are you even asking

[open pelican]

Is A Ai Maeko Voice Create By Me

[proven hill]

okay but what are you asking, you want this person voice but you already have it, im confused

[open pelican]

Do You Want Maeko Ai Voice?

[proven hill]

no

[open pelican]

Please 🥺

[proven hill]

please what

[open pelican]

Make To Maeko Ai Voice By Me, Please

[proven hill]

the... what?

[viscid topaz]

hell nah

[eager wadi]

man wtf

[open pelican]

Maeko Ai Voice By Me

[proven hill]

okay good boy

but i still dont understand your question

[eager wadi]

so do you want to make one or use one that already exists and make something with it

[open pelican]

Just Make Maeko Ai Voice By You

[pastel oak]

I told him to speak in his own language itd make things easier

[proven hill]

why would I do it

[pastel oak]

Are you indonesian

[proven hill]

he alr said hes from us

[pastel oak]

He also said hes from "english"

[low shard]

@open pelican
Please stop begging, people aren't obbligated to make your voice

[proven hill]

LMAO

[pastel oak]

Hes from southeast asia tho i just want him to speak his native language so its easier to help

But bro keeps lying to me 😔

[proven hill]

i think he wants someone to remake the voice model in HQ

me think

[shadow flax]

idk why but voicechanger just works so mid for me

like its very obvious

[proven hill]

considered that the settings are no good? or like the voicechanger is not vonovox/deiteris? or that the model... simply... sucks?

[proven hill]

ofc, rvc cant laugh or cough or scream correctly

[flint vapor]

I told you before to put it in request channel, make it on your own, no one is forced to make the model

[sick plover]

if the package says, for example, "eng", can I speak other languages and it will be able to do it normally?

[proven hill]

not really, also wdym for package? the pth?

[sick plover]

I didn't quite understand the answer

[proven hill]

wdym for package?

[sick plover]

I want to use it for the first time and I wonder if the voice pack says only English, can I speak other languages and use it?

[shadow flax]

i tried deiteris, VM showcase sounded VERY good, pitch and stuff was also correct so idk

[proven hill]

yes but it will probably stick with an english/american accent you cant remove

[proven hill]

can you give me context?

[shadow flax]

whats that mean :v

[proven hill]

what are you asking

[shadow flax]

wdym by context

[proven hill]

https://dictionary.cambridge.org/dictionary/english/context

[shadow flax]

just monotone and robotic voice (meaning shotty tone changes/shifts)

when talking normally

[proven hill]

tell me your gpu, wich voice changer youre using and what model are you using

[shadow flax]

5070ti,

idk i just installed some from here that sounded good in the preview, im not home rn im workin so i cant really check which they were

[shadow flax]

wokada deitiris fork

[low shard]

This is a General AI Server, AI has many fields

Elaborate:

• your PC GPU
• your operating system
• what you want to do
• what tutorial link are you using
• a screenshot of the program

@shadow flax you have to elaborate, this is a general AI server, and there's tons of possible factors on why the program doesn't work, AI is complex and intensive

[shadow flax]

i went through the installiation with you so i know its setup correctly 😔

[low shard]

I dont remember you at all, i just checked #✨│ai-help message
you got an rtx 5070ti, may i check the settings you're using?

[chilly furnace]

whered yhu get tht from

?

[latent kettle]

Yep you are right

But to run high parameter models you need a good gpu

[shadow flax]

uhh if youre still Here in like about 4 hours sure, im at work rn, also explains my delayed awnsers

[analog obsidian]

are u using a egirl model? these type of models are always trained using very monotone speech/asmr
just use something better, anything that is not trained with asmr should suffice for realtime

rvc outputs are flat by nature bc it's not a true voice cloning ai

its just trying to follow the pitch changes

[young halo]

Hoe much time does the GPU for applio take to restore?

[latent kettle]

What do you mean

[young halo]

I tried to resume a training and it said that i don't have enough GPU or something like that

[latent kettle]

Show me the console screen

[shadow flax]

i actually tried one aswell yes, that was pretty horrendous

if youd want to you could refer me to a good one, then i can judge if the issue were the models

later obv

[young halo]

Gimme a sec

[low shard]

alr, not sure but lets see

[shadow flax]

sure

[frank imp]

anyone help me i dont know how to get a voice changer and use it pls

[flint vapor]

Ok, first of all, which is your OS, and GPU?

[latent kettle]

I see

The colab only allows you to use GPU for 4 hours

And now it's over

[frank imp]

how do i see that

[flint vapor]

OS is for example Windows 11, Windows 10, Linux, etc...

GPU you can see it going on task manager, search on windows bar for task manager

Then click on the line above and select performance

[frank imp]

ty i have windows 10 and AMD Ryzen 5 4600H (idk if this is needed but with radeon graphics)

[flint vapor]

That is CPU, you see that also as your gpu?

[frank imp]

yep

[flint vapor]

Mmhh, you can use only by cloud

I will send you a guide right now

[frank imp]

k

[flint vapor]

https://docs.aihub.gg/rvc-voice-changer/cloud/deiteris-w-okada-fork-kaggle/

If you have some problem let us know 😄

During the setup

[frank imp]

ty

[low shard]

you sure you don't got a gpu?

You can check your pc gpu on Windows via:

ctrl+shift+esc (task manager) -> Performance tab -> GPU

[alpine onyx]

@low shard Ever since I did the force_gpu_clocks yesterday, not matter how often I run the reset_gpu_clocks again something has been incredibly wrong with my PC because of it. My PC makes insane scratching and vibration noises now.

[flint vapor]

The noise came from your gpu? or pc fan?

[flint vapor]

Run reset gpu clocks

[alpine onyx]

I did multiple times, it doesn't do anything

it just says allright but the noise is still there

whenver I'm not in idle

pretty sure the force_gpu_clocks broke something

[flint vapor]

nvidia-smi --reset-gpu-clocks
nvidia-smi --reset-memory-clocks

Try this in terminal

[alpine onyx]

oki

[flint vapor]

Or clean your drivers with DDU, i will put you some guide, and reinstall them again

[flint vapor]

https://www.drivereasy.com/knowledge/how-to-reinstall-gpu-drivers-with-ddu/

Don't be scared of these steps

It happened to me too, ,my gpu was showing just black screen

Then i fixed all with this

[alpine onyx]

I don't want to break it further

[young halo]

And how much do i have to wait for another 4 hours?

[latent kettle]

Probably 24hrs

[flint vapor]

Don't worry, is for clean drivers, or you can just reinstall them normally

But ddu is recommended for clean drivers if they broken

[alpine onyx]

the GPU still shows everything clearly it just makes an insane type of noise now

like it's just breaking apart

whenever I'm not in idle

[flint vapor]

Sometime it can be for drivers too, have you tried first to do /sfc scannow ?

Open your cmd as administrator

It will scan for issues in PC

[alpine onyx]

doesn't that take like 8 hours

[flint vapor]

No no don't worry

Max 5 minutes

[flint vapor]

Then we will try this method, i swear it will not broken nothing, and is usually recommended to clean drivers sometime

I know it seems strange, but broken drivers can cause every type of troubles in GPU

[alpine onyx]

It's stuck at 77%

[flint vapor]

Don't worry it will fix in a bit, sometime is a bit slow is ok

[uneven bear]

Can someone help me with real time voicechanger

[flint vapor]

Hi!

Of course

First of all, which GPU And OS you have?

[uneven bear]

Wanna call maybe, but i9 and Rtx 4090 Ti

So Nivida and win

[flint vapor]

I can't call at the moment

[flint vapor]

Perfect! i will send you a guide right now

Download the one for nvidia, not 5000 SERIES

-wokada

[patent trellis]

💻 Local Realtime RVC

Guides for Programs that use RVC Models in Realtime for Calls/Games

• Wokada Deiteris Fork

Most suggested WebUI with the best general support for many platforms. GUIDE

• Vonovox

A Realtime Voice Changer with similar performance to Wokada Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. and without cloud options GUIDE

• Original Wokada

Not suggested nor maintained, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated

[flint vapor]

Follow the first link for deiteris

[flint vapor]

You are on Windows?

[uneven bear]

yes

[flint vapor]

Perfect, follow the guide ! and let me know if you have some error ! 😄

[alpine onyx]

@flint vapor still at 77%

[flint vapor]

Close it , try run again or we will do with DDU

[shadow flax]

can someone explain to me when to use quantization and what the "results" are?

[alpine onyx]

Windows Resource Protectio ndid not find any integrity violations

[flint vapor]

Nice , I recommend you to use ddu method now

[sand bison]

My GPU is an RX 580. I know it's not appropriate for training AI. I just want to create the AI voice model since I do the inference locally by Applio and I haven't followed any tutorials for a long time.

[low shard]

Train (make) RVC Models on cloud:

1. Prepare the Dataset
2. Setup RVC:
   Choose a cloud way to use RVC,

• Google Colabs (max 4 hours of daily T4 16gb gpu not granted for free, not much hours for training, but easy to use, there's a paid tier):
  • Applio (UI)
  • Mainline (U, END OF LIFE UPDATE)
• Kaggles (a bit harder to use and needs phone number but gives 30 hours weekly of better gpus, either T4x2 16gb each or P100 16gb, only free):
  • Applio (UI)
  • Mainline (UI, END OF LIFE UPDATE)
• Lightning.ai (Kinda hard, needs login, no issue with web uis or anything, but only free 15 credits monthly, Free Studios run 24/7 but require restart every 4 hours. There's a paid tier):
  • Applio (UI)
  • Mainline (UI, No guide as of right now)

3. Be sure to know about the tensorboard

Google Colab = Easier but risk of getting disconnected
Kaggle = Harder but way more gpu time
If you are looking for the easiest way and for free, is using https://weights.com/ which ofc uses RVC

RVC Inference (use models) on pre-recorded audio on Cloud

You can use either:

• Weights.com: Easiest Possible Ever Automatic
• Ilaria RVC Zero: Fastest free on cloud
• Applio UI Colab: RVC Fork with some extra features like TTS
• RVC AI Cover Maker UI: Automatically Separates the Vocals and Instrumentals, converts the voice and mixes them back

[alpine onyx]

would you happen to have a link for it

[flint vapor]

Yes of course i send you now

[flint vapor]

https://www.reddit.com/r/pcmasterrace/comments/1ccssm0/complete_ddu_guide/

[flint vapor]

https://youtu.be/hEeDnSAKl8s?si=CF_mNgkD6a2qOzq0 Video

Before run ddu, install the newest drivers from nvidia website

This is very important

You might lose cookies on browser, you can login on social and website again

[flint vapor]

Let me know for every question, and how is it going

[alpine onyx]

The reddit thread says NOT to install the newest driver

[flint vapor]

Yes, it might be some problem but usually it not have problems

You can download

Your current drivers

do on cmd

nvidia-smi

And see which version you are now 😄

[open pelican]

I Waiting On 2 Days

[flint vapor]

It said not install but you have to download it

Install means like you applying the new driver on pc

You have to download, and then install it after you execute ddu

[viral mason]

that guy needs to know how to use applio Kakao, his english is bad tho so I couldn't help him

[flint vapor]

Which guy?

[alpine onyx]

I'm already on the most recent driver

[viral mason]

Neon David, the one waiting on weights for a billion years for him model that will never be made due to the wait

[flint vapor]

Ok perfect, install it from the site

But not execute it

Then do all the steps with ddu

[flint vapor]

He want in local?

[viral mason]

not sure

[flint vapor]

Neon do you have a GPU? And which is your operative system?

[alpine onyx]

ok restarting the Pc and I'll see if it helped

[flint vapor]

Ok nice, did you do all?

[alpine onyx]

yes

[flint vapor]

You did it in safe mode right?

When you reboot pc then you have to install the drivers

[alpine onyx]

yep also ran

nvidia-smi --persistence-mode=0

in case it was persistant in keeping the clock changes

[flint vapor]

Yes nice 😄

[alpine onyx]

ok after testing it's still the same

[flint vapor]

You still have the noise?

[flint vapor]

https://github.com/w-okada/voice-changer/issues you can open an issue here they will reply you, they are the one who developed the fork

[flint vapor]

@low shard Do you know what can causes this noise? After he did gpu clocks, his gpu do a weird noise, he even cleaned his drivers with DDU

[open pelican]

Yes

[flint vapor]

Which?

[spice glacier]

hello guys I am asking for what's better AI only study program or AI/data what would you reccomend and why

[open pelican]

GPU

[flint vapor]

Do you want to basic study for school or for programming?

[flint vapor]

Yes which

[spice glacier]

university

[open pelican]

👍

[flint vapor]

There are many, I recommend you Claude or ChatGPT, but in general i recommend you perplexity, it have all the ai in one service

[open pelican]

Oh, Great ☺️

[flint vapor]

What?

[spice glacier]

I just finished integrated prep school and going into engineering cycle

[flint vapor]

I asked you which gpu you have

[flint vapor]

Claude is very good, but also perplexity, it have many ai in just one service

[spice glacier]

claude is a chat bot , I am talking about program

[alpine onyx]

Yes. It's a phsical noise btw. Not digitally. But from the actual PC

[spice glacier]

sorry I am new to this , am I missing something do they offer programs,

[flint vapor]

So you don't want chatbot? You want an AI that program for you?

On claude you can program too

Or also perplexity

[timber merlin]

hello, currently im having an issue where my voice isnt being picked up by the program, it messed up after i switched from one mic to another?

[spice glacier]

(I meant study programms to get a diploma ) AI or AI/data 😅

[flint vapor]

For study still Perplexity is good, it have many AI integrated, they are chatbot, but there is lab mode where you can create schemes with AI for make you study

[flint vapor]

Which version of voice changer you have?

[flint vapor]

Is just the gpu that do noises?

[flint vapor]

And which gpu you have?

[spice glacier]

I understand what you are saying thank you , but what why I want is to go study in a university and some offer AI and some offer AI/data

[timber merlin]

amd 9070 XT

it used to work fine but as soon as i switched it from my vr mic to my normal mic i stopped hearing the voice

[spice glacier]

how do I know which to choose ( I think that AI alone is better , but I might be wrong due to lack of expertise)

[flint vapor]

Do you have set also as default microphone your default mic on your browser settings?

[flint vapor]

Which one do you had already in mind for use?

[proven hill]

hi i need help with

[alpine onyx]

I have no clue but I assume it's the GPU cause only after I did the GPU_clock thingy yesterday did it start happening

[proven hill]

please egirl latona model no crackaled perfect indistinguashable from reality free

[flint vapor]

If you install GPU-Z and see your values of gpu in idle?

Sorry cpu z

But it will show you also gpu

[timber merlin]

No but i just switched the default option to my mic

oh

it works now

thanks

[spice glacier]

first my university , then Stockholm University ,Université de Lille ICG montpellier

a lot

[flint vapor]

Nice! you put your default in browser settings or it fixed randomly?

[flint vapor]

But which AI you had in mind to use? what this AI have to do? you want text with this AI? or what you want do?

[spice glacier]

I think u missed for I said , I don't want to chat with an AI , I want to study AI in a university

[alpine onyx]

Do you have a link maybe ? :3

[flint vapor]

So you want just a good university?

For study AI?

I really don't know for universities, This is server is for General AI but not for address universities

[proven hill]

wait kakkao youre an helper

[flint vapor]

Yes! They put me again today! 😄

[proven hill]

lol you requested it?

[flint vapor]

Oh yes i wanted try to do apply staff again and fortunately i have passed 😄

Also FanFierik!

He is QC now

[proven hill]

wait really wow

[flint vapor]

Yes! He is busy today but from tomorrow he will start chat, his request been accepted too 😄

[proven hill]

nice nice

[flint vapor]

2 Italians again as staff! As the old times hahah

[proven hill]

frfr

[spice glacier]

what do you do here , like what do you do with AI , do you code , do you study

can I get my hands dirty in AI before actually going to study it , I am trying to revive my memory in python

now

[flint vapor]

We can also use for study, code, but here in server is mainly for help, for voice cloning, images, tts, yes it is AI in general but is mostly this here, but we don't address for universities

[spice glacier]

and I am good at algebra/analysis , is it enough

so like you don't actually work with AI , like they do in companies

[flint vapor]

Today i have fever too so i'm doing helper with fever today😭

[flint vapor]

No this is a community server, for talk with other people about AI

Is not for money here

[spice glacier]

I just want to learn

[flint vapor]

Is like an hobby here in the server mostly

[spice glacier]

do you code ?

[flint vapor]

If you want learn do something like voice cloning, images, tts, videos you can ask

[spice glacier]

oh

ok

[flint vapor]

I'm not so much good for code AI, here is not mostly about coding, but is a general AI server

[spice glacier]

ok

ty

[proven hill]

noooo

[flint vapor]

If you have any request if you want learn how to voice clone, how to create images, tts, audios, you can ask

[flint vapor]

Yes sorry if i didn't saw message

[flint vapor]

https://www.cpuid.com/softwares/cpu-z.html

[low shard]

that's original wokada, not wokada deiteris fork

[low shard]

this is the first timem i ever hear this, yall sure that caused it? @analog obsidian you ever heard had this issue before?

you forgot to elaborate:

• what you want to do (and did)
• what tutorial link are you using
• a screenshot of the program

[analog obsidian]

sounds like hardware failure to me

[pastel oak]

Not to mention that is the first one, its just an extension - you are using an older version of the voice changer. We recommend you update it, I can give you a link.

Need to know what GPU you have first

https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/

Download virtual audio cable, download the first nvidia link (nvidia-b2332), read opening on windows, audio setup, settings

[low shard]

dont follow yt tuts for realtime voice changer, they are all old, and this is a general ai server btw

[viral mason]

I feel most people still come here for okada

[flint vapor]

Oh true sorry, I clicked on deiteris fork, idk why it redirected me on mainline

[viral mason]

nick is a robot/j

[rough warren]

What happened to all my Weights Voyages images?. It is like my app got reset.

[viral mason]

weights is like doo doo

[nimble mortar]

is it an aux splitter or aux cord u have to have for the 0 delay

[jolly abyss]

Does this work w amd

[young halo]

how does my model sound so far? i'm not done training yet, this is at 340 epochs

[brittle wing]

i have train my voice model but my gpu is ended on colab. how can i resume the train or export the .pth file?

[viral mason]

use kaggle it's better

30 hours

[frozen zinc]

Is there a way to fix the perf keeps going up and down on b2300?

[open pelican]

@viral mason, Where Is New Maeko Ai Voice

[craggy bough]

stop asking about it

[open pelican]

New Maeko Ai Voice, Where Is It

[craggy bough]

you already made it on weights

dont ask me where it is, you should know

you made it

[open pelican]

You Know What, That's It

[viral mason]

Learn applio please

[craggy bough]

thats like asking "wheres my food"

how am i meant to know

you made it

you should know

#✨│ai-help message

[open pelican]

New Maeko Ai Voice Is Here, Link To A Screenshot

[viral mason]

What

[open pelican]

Is Now Here

[viral mason]

This thing must be like 5

[frozen zinc]

anyone pls answer my question

[open pelican]

Click To Screenshot To Be Finished New Maeko

[viral mason]

You are not real

[open pelican]

Please, Just Screenshot New Maeko

[viral mason]

???

[open pelican]

🥺

[viral mason]

I wish I knew what you were saying

This feels like talking to a dying insect

[open pelican]

Is Screenshot Here?

[tender latch]

Well

[open pelican]

I For Got Screenshot

@viral mason, Is Ai Voice New Maeko Here?

[viral mason]

Please just learn applio and make it yourself 💔

I can't talk to you because you sound like a broken record

[open pelican]

Where Is Ai Voice New Maeko Create

[viral mason]

♻️

[eager wadi]

im confused are you talking about the maeko who is a character from a friday night funkin mod

[candid vessel]

chat these aer my specs
Device name LAPTOP-8AE20BRV
Processor AMD Ryzen 7 5800H with Radeon Graphics 3.20 GHz
Installed RAM 16.0 GB (15.4 GB usable)
System type 64-bit operating system, x64-based processor
Pen and touch No pen or touch input is available for this display

should i use a broswer

or download

ive tried using the browser but that stuff always

like stops

every 40 mins, then restarts

[knotty moth]

he seems just trolling

[eager wadi]

id be tempted to help him with it so we can get this over with, but idk what character he even means

[knotty moth]

[knotty moth]

why do you even disclose your laptop serial number, like if we were the OEM customer service?

[knotty moth]

explain it with some error message in the terminal/console window

[candid vessel]

oh

i didnt even realize

[viral mason]

I think most of that stuff we don't need

Just like graphics card (gpu) and that's it

[knotty moth]

it seems an igpu laptop

[candid vessel]

My laptop doesn’t have a dedicated GPU, so I can only run the voice changer on Kaggle, but every time I leave the notebook for too long, the process restarts, and I lose everything — the model stops downloading or it fails.

[viral mason]

Aw dang

[candid vessel]

indeed

i fear i may not be able to use the downloaded version

[viral mason]

I'm so sorry for you and hope you can upgrade soon

[candid vessel]

well thank you for ur kindness

bless you

and for the help guys

might just gotta

upgrade

[knotty moth]

you need to keep the kaggle session active, or it could disconnect

[viral mason]

Yeaa

Only option tbh

[eager wadi]

man all this time this guy wanted an ai voice model of someone im not even convinced has any dialogue where she speaks english

[viral mason]

He sent me the audio

In the weights server

Just sounds

I mean he can barely form a functional sentence and definitely cannot read, he's like an NPC repeating the same thing over and over

[eager wadi]

yeah

i can only find like vocaloid type noises and cutscenes where the characters speech is literally just some sort of elevenlabs tts i think

[viral mason]

Lol

[open pelican]

Maeko (Ajthefunky)

[eager wadi]

yes that character has no actual distinct english voice

[viral mason]

Or any language outside of beep boop

[eager wadi]

yeah

[viral mason]

I used to make fnf models and know how to actually make them good

[eager wadi]

I am aware of the character yes, can you send me a sound file of her speaking

[open pelican]

That's Is Maeko

[eager wadi]

do you want her speaking, or do you want a visual animation

[open pelican]

Speaking

[eager wadi]

then i need a sound file of her speaking

not any weird beep boop noises

speaking

does something like that exist that you can point me towards

[viral mason]

He's loading new words

[eager wadi]

we'll get there in about a year

i was supposed to be asleep hours ago here we are

[viral mason]

Same

I should sleep

There's a bug in my room keeping me awake

I like watching it walk around

[eager wadi]

im up in an hour and a half so

no point sleeping now

[viral mason]

Damn

[eager wadi]

supposed to be drumming for the band at my church

im gonna need an energy drink

triple strength

no response from our maeko friend