✨│ai-help

[eager wadi]

is he processing everything

is he trying to find sound clips

[proven tapir]

voice changer takes too long to work

like

after me speaking

it wont work

for

3 seconds

[zenith coral]

Im having the same issue

[eager wadi]

which voice changer are you guys using

dont know much about it but im seeing this get recommended a lot in the past so hopefully it helps? https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#reduce-more-delay

[zenith coral]

What does that do

[maiden wren]

decrease chunk

[proven tapir]

how

[zenith coral]

@maiden wren I thought decreasing chunk makes it laggy no? my ChunkSec is 2400 0.05 seconds and my exatrameSec is 131040 and I still have like a 3 second delay how should I fix it

[maiden wren]

2400 chunk is crazy

[maiden wren]

whati s your gpu

[zenith coral]

I got a 4090

[maiden wren]

are you using cpu or gpu for deiteris w okada

check it

[zenith coral]

Should be gpu

Where can I check

settings?

[maiden wren]

no lemme open it

wait

[zenith coral]

Ok waiting

@maiden wren Lemme know what I should change ima go get some water brb

[maiden wren]

the setting is under chunk and extra

it will say gpu:

and if it says your cpu then change it to gpu

[maiden wren]

make sure to confirm

and after selecting your gpu decrease chunk

extra should be 2 seconds

[zenith coral]

Wait

Can u show me what urs looks like

[knotty moth]

the beep boop thingy sounds just brainrot

[zenith coral]

I got the newest version I could that was compatible with my gpu

[maiden wren]

im on really bad pc right now cus my rtx pc is being used by family but if i can send screenshot here it is

it still can handle the voice changer absolutely fine

no robotic voice

even if pc is from 2016

[open pelican]

1 Day Left To Be Finished Maeko Ai Voice

[maiden wren]

is your gpu selected

[west cave]

I assume the 1070 would be able to handle a lower chunk which is better

would it not?

[zenith coral]

Ok so mine looks NOTHING like this thats the issue LOL idk if I got the wrong version

[maiden wren]

ohh

[maiden wren]

it can handle big and low chunk

[west cave]

ok

[maiden wren]

no problem at all

it struggles at chunk 100 and lower

[west cave]

because around 300 is what most modern gpus (10 series or later) should be able to run minimum.

[maiden wren]

no, it handles it pretty well

[zenith coral]

@maiden wren WHAT DO I DO CHATT

[maiden wren]

i also use windows lite for context

[maiden wren]

go to guides and you will see voice changer follow the guide and download it

#1159513888199540817

delete that old version

[west cave]

send a screenshot

[maiden wren]

the higher the chunk the better it runs

at max with perfect non robotic voice i can run with 1 second delay

or less

although models arent meant to be perfect

its close

[west cave]

chunk is basically the length the audio is split into, thus a higher chunk will always be longer no matter how good your pc is. higher chunks may sound slightly more realistic in certain situations (but also sometimes less realistic), but will go from a quarter second delay to multiple seconds.

just for anyone that wanted to know if you already knew

[maiden wren]

if knew

i knew

i just change chunk depending on how model works

this is the settings i mostly use it works really well

but i change depending on voice

alr gotta play adofai rn

[west cave]

Unless you're not recording live, that chunk setting will give you so much delay that the voice is pretty much unusable in normal situations

[maiden wren]

im recording live and its pretty well tbh

[west cave]

the delay will be more than a second

[maiden wren]

nahhh

its like a second

[west cave]

why not lower it to like 300 or even lower?

it shouldn't affect quality too much unless you go below the "perf" number in the top right

[maiden wren]

nah, it would sound less realistic for the current voice im using

[grave vale]

how well does this run with like nvidia cards

[west cave]

left*

[west cave]

good, it's made for them

[grave vale]

so like itll run good with a 3060 12gb

[maiden wren]

im using perfect settings for the current pc

and runs amazing.

with a 1 second delay

at live idk but my pc handles it well

while gaming on top

[west cave]

is the difference in quality really that much at 300 where it will perform 3x faster? in some cases a lower chunk should even be better quality wise

show an image

[maiden wren]

nope lower the chunk the more robotic it is although speed isnt issue for me lol

and 300 chunk wont really work with my current voice im using

[west cave]

yes, that is more than enough. just make sure chunk is set just high enough so the perf is green but barely any higher

[maiden wren]

it works for specific ones

[west cave]

I see, but I'd change your model if a low chunk isn't working but it depends what you're doing

if there are no alternatives, then it may be "good enough"

[maiden wren]

am basically chillin

[zenith coral]

mannn im ngl I found one off yt some video lol

[west cave]

Those guides are outdated.

[maiden wren]

lol

[zenith coral]

BROO HELP ME OUT HERE MAN IM NEW TOO THIS OKAY LOL

[maiden wren]

broo #1159513888199540817

get it from there

it has download link

alr gotta play adofai

[west cave]

basically just install the deiteris fork as shown in the guide, then download an rvc model. use #1175430844685484042 or google "rvc models"

[maiden wren]

while i check on here

[zenith coral]

@maiden wren Lowkey ima do all that tmmr @west cave Ty both for helping if I have more questions tmmr ill ask

[maiden wren]

ok

[grave vale]

like how high tho

[west cave]

there will be text that says "perf" in the top left. make it slightly above that

above so there's some wiggle room

[solar ivy]

Hello! I am currently trying to find me a good ai art generator that has unlimited generations daily and can generate images the size of a phone screen. Does anyone have any recommendations? I’d prefer app versions that don’t use a coin system if possible, but if not web versions that don’t use one are fine too!

[latent kettle]

If you have a good GPU you can use flux, sdxl, hidream on your system locally and unlimited

[solar ivy]

Are they hard to set up? Do I just need to google it and it be pretty simple, as I’m not too tech savvy,

[latent kettle]

Its not that hard, you just need a good PC with at least 16gigs of RAM. 8gbs of VRAM or GPU memory (12 or 16 is good [more is better]) and some storage to download the stuff.

[latent kettle]

-realtime

[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

[latent kettle]

You should use wokada deiteris fork guide

[proven tapir]

bro what is that version

mine is different

[latent kettle]

I sent you the guide

[hushed thicket]

do u guys know why it starts off quiet then gets louder the more i talk? also if i stop talking it does the same thing

[flint vapor]

Voice Changer?

[hushed thicket]

yeah

this one

[flint vapor]

I recommend you to use client and not server

[proven tapir]

yoo

where can i download that version

[knotty moth]

nah just let him try both client and server

[hushed thicket]

both have the same issue

[knotty moth]

also you try another model

[flint vapor]

Oh yes of course, i thought he only tried server

[proven tapir]

can you give the link for that version pls

[flint vapor]

Which GPU do you have?

[proven tapir]

1050 ti

[hushed thicket]

other models onyl slightly do that

[flint vapor]

Wait i will give you 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

[proven tapir]

alrighttt

[flint vapor]

Follow the first link for deiteris

[flint vapor]

https://huggingface.co/Shadicti/deiteris-Fork/blob/main/voice-changer-windows-nvidia-b2332.zip

If you have 1050ti i don't recommend you so much to play while the voice changer is opened

It can causes some issues

Like slow your performance a lot

[proven tapir]

alright broo

tyyy

[flint vapor]

You welcome 😄

[proven tapir]

yoo kakao

how do i run it

[paper lotus]

yo

i got a r5 3400g

nd like 16 gigs of ram

and 5 terabytes of nvme ssd

what can i run rn?

theres a month left till my gpu arrives

what fork do i run for now

it would be better if the fork is hosted online rather than locally on my device

someone answer it rq

[paper lotus]

@steel forge

ur cooked gng 💔

[flint vapor]

Did you downloaded?

Which is your gpu

Oh

You need to use voice changer on cloud

I will send you now a guide

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

[flint vapor]

Do you want run a voice cahnger?

https://docs.aihub.gg/rvc-voice-changer/cloud/deiteris-w-okada-fork-kaggle/ you can run only this until your gpu will arrive, which gpu will arrive to you?

This is on cloud

You have Integrated GPU

[paper lotus]

rtx 4070

[paper lotus]

no i meant i download it

but it runs on chrome locally

or smth similar to that

i used to do that before

[flint vapor]

If you have on integrated gpu i don't recommend you to install it, i don't know how with amd will work , but i searched that the r5 3400g have only a integrated gpu, usually the tool don't detect it

https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/ if you want read same the guide for local is this

[paper lotus]

i used to use one similar before

using this same gpu

and cpu

it worked great

not that much delay

[flint vapor]

You can try install the amd version

I will give you

[paper lotus]

i remember it opening up a chrome tab

[flint vapor]

https://github.com/deiteris/voice-changer/releases/download/b2332/voice-changer-windows-amd64-dml.zip

[paper lotus]

and i switch everything

there

do u know what im like talkin about

?

[flint vapor]

Yes

[paper lotus]

yeah what is that

[flint vapor]

I think is this the voice changer you talking about

[paper lotus]

oh ightt wait

oh yeah right

how do i fix this

@flint vapor

[flint vapor]

Do you have some antivirus installed?

[paper lotus]

yeah windows

[flint vapor]

Did you download this exact version?

[paper lotus]

but its turned off

[paper lotus]

yeah

[flint vapor]

So before you said it was work? you can try uninstall and reinstall maybe sometime the installation can be corrupted

[flint vapor]

Try to delete the stored_setting.json and pretrain folder and try again

There aren't yt videos , but the guide is very good

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

[paper lotus]

ightt wait

[flint vapor]

Oh i'm sorry at the moment i can't, which is the problem you have with steps?

Follow all step by step don't worry look difficult but is easy

I can in a bit i can't now, if you want you can send me a screenshot

I will give you permissions wait

[paper lotus]

also where can i find stored settings json?

[flint vapor]

Oh i think it doesn't let me give perms, you can send me a screenshot in dm

[flint vapor]

Is in the folder

Of mmvc

[paper lotus]

no like

inside interval?

internal*

@flint vapor

here

it said this

[maiden wren]

bro can you try disabling firewall

[paper lotus]

i did bro

it didnt work

[maiden wren]

maybe you didnt disable everything in firewall

i used specific app to disable firewall fully and it worked

[paper lotus]

what did u use

[maiden wren]

dms

yes

[simple ore]

it tries to download the model files

and fails

[maiden wren]

disabling firewall works

cus it happend to me i turned firewall off and it worked

yo

[maiden wren]

yoo

how you guys doin

[proven hill]

https://tenor.com/view/ineedit-needit-spongebob-squarepants-need-it-gif-4883495

[maiden wren]

bro needs help

help him guys

[proven hill]

yea help him NOW

[latent cypress]

https://tenor.com/view/chinese-man-angry-dishu-gif-27128084

[proven hill]

whats the issue

follow the guide

[flint vapor]

I'm back sorry if i make you all wait

About what?

Where? RVC, Okada?

Where do you want use it?

[maiden wren]

there is bro

[flint vapor]

On voice changer or where?

Did you follow all step by step? are you trying to use voice changer?

Sorry for what? What are your intents? With girl voice changer

Tell me before your intents with girl voice changer, do you want e troll? what are your main intents?

Yes but for which intent mainly? So i can help you

Ok, but if you use girl voice changer, there should be a reason, testing, e troll people, what you want do?

[proven hill]

are you a catfish?

[flint vapor]

Or many other cases

[maiden wren]

noice lol

[open pelican]

Guys, I Just To Faster But Here This

[maiden wren]

yo whats up

[latent cypress]

buy a premium subscription, mine took 6 days to finished (i started from 1000 queue)

[maiden wren]

6 days is crazy lol

[west rampart]

does it matter if you run okada on a hard drive opposed to an ssd?

[maiden wren]

nah doesnt matter cus it ran well for me

[paper lotus]

@flint vapor

help

i turned off my firewall

and

ran as admin

didnt work

[simple ore]

dont run as admin

never run anything as admin

[simple ore]

you can try manually downloading the files using URLs from https://github.com/deiteris/voice-changer/blob/master-custom/server/downloader/WeightDownloader.py and see what error you get

[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]

@flint vapor yeah no the sound is still there. I'll record it later maybe you can help better that way

[flint vapor]

Is your wifi speed good?

[flint vapor]

Sure thank you

[alpine onyx]

@flint vapor

[viral mason]

this is fucking scary

there's a creature

inside your machine

[alpine onyx]

yeah happened after I ran the gpu clock batch in the folder

that's on idle btw

[viral mason]

might be possesed

[alpine onyx]

Nah this has never been the case

only after I ran the GPU batch

[alpine onyx]

@flint vapor

[open pelican]

For My Maeko Voice Ai

[latent cypress]

https://tenor.com/view/males-haduh-gif-3346317461364735873

[viral mason]

don't buy anything on weights, just use applio

weights is dog shit

[open pelican]

PLEASE!!!!!! 😭

[viral mason]

for making ai voices it is

[viral mason]

learn

how

to

use

applio

on

kaggle

can u read?

[open pelican]

I wanna To Make A Maeko Ai Voice

[latent kettle]

Yep, it's better than wasting money on subscriptions

[open pelican]

But How Dare You, You Monster

leaving

[viral mason]

what

😭

[open pelican]

I Never Only A Friend Request DM

[viral mason]

what riddle is this

[latent kettle]

We told you to learn applio. That's up to you. Nvm

[viral mason]

I don't think he can read

I've been trying since yesterday

[latent kettle]

Google Translation and chatgpt is there to help

[flint vapor]

These should be normal, the sound you send looks like Coil Whine, let me check if is possible to return it back, with other methods different than the drivers one we tried

try do this on cmd as administrator: nvidia-smi -q -d CLOCK

[latent kettle]

It's very easy nowadays, for example if I want to know what is Mel, or kl or anything about it I just say chatgpt to explain it in baby terms.

[viral mason]

all I remember is orange line going up is bad, and down is good

ez

[flint vapor]

Bro do you even know english? You just asking here in ai help 1000 times at day for your maeko model, just leave it until someone don't give you, or if you want learn how to make models just ask

[latent kettle]

Yep. But I went to the depth, why it happens etc

[viral mason]

I just need to know the basics and then I'm good

[flint vapor]

And i will be here for help you

[latent kettle]

Yep, but I was having curiosity 🎶

[viral mason]

fair enough

[viral mason]

I'll be honest if he tried to ask anything outside of the 3 sentences he can (maybe) say properly it wouldn't be understandable

[open pelican]

No Maeko Ai Voice? 🥺

[flint vapor]

No one is forced to do for you, i don't know if you are trolling or what

If you want learn how to make model

Just say it

[open pelican]

How Dare You To Delete My Maeko Ai Voice, You Monster 😡

[flint vapor]

Lol you definitely trolling, you are not fun anymore

[viral mason]

yeah this guy needs medication, and english lessions

[open pelican]

There Is Not Maeko Ai

Is A Black Hat

[viral mason]

yes

[flint vapor]

So? are you a black hat hacker?

[viral mason]

don't ask him anything too complicated

[flint vapor]

I really think he is trolling or idk, he just spam from days always this maeko model

[viral mason]

yea idk, he was doing the same thing in the weights server where I found him

[flint vapor]

If he want maeko model he can learn, but is impossible to speak with him

[viral mason]

it's like poking at a dying insect that's stuck on its back

[viral mason]

tru

imagine if a timeout could last for 5 years

[alpine onyx]

what is this command supposed to do ?

[flint vapor]

It should always tell you mhz of your GPU don't worry 😄

[flint vapor]

It should say you if your gpu is clocked

Oh wait let me check

[flint vapor]

Oh wait now that i checking this in fact if you are in idle are a bit high, when you was in safe mode when you opened ddu the gpu was always do weird noises?

[alpine onyx]

No the gpu only started doing weird noises after the Voice changer stuff and the gpu_clock batch file

I've never had this issue before

[flint vapor]

nvidia-smi -rgc
nvidia-smi -rac
nvidia-smi --reset-gpu-clocks
nvidia-smi --reset-memory-clocks if you try to do these again? there are 2 new commands, is always for reset any evenutally clock that the gpu have

always in cmd as admin

[flint vapor]

Then see if you have same values with cpu z of mhz

[open pelican]

No

[open pelican]

Maeko Is My Girlfriend

[alpine onyx]

done, still the same

[flint vapor]

Ok

[viral mason]

I think he's like 5

max

[flint vapor]

Try do nvidia-smi -q -d PERFORMANCE

Is always for see some values

[flint vapor]

Oh wait

[simple ore]

why are you messing with nvidia-smi?

[flint vapor]

Oh is for see values of his gpu

Basically

He run gpu clocks

.bat

[simple ore]

just reboot

[flint vapor]

And then his gpu started to weird noises

[flint vapor]

He did, he have problem from yesterday

[flint vapor]

@simple ore

His noise

[viral mason]

the creature in his pc

[flint vapor]

He also cleared his drivers with ddu

[viral mason]

sounds like smth from godzilla

[flint vapor]

And reinstalled them

[flint vapor]

He also clicked on reset gpu clocks

But nothing works

I really don't know if the gpu had already some issue and the bat causes some problems, i was never saw that before

[alpine onyx]

it didn't

There were no prior issues

[flint vapor]

Oh yes i know, maybe there was something and then it caused this, sorry if i make you always write things in terminal is only for see the values

[alpine onyx]

No it's fine I really awnt to get this fixed too so I'll type anything xd

[flint vapor]

It should be weird that the gpu is in p0 state, it should be max performance, but are you in idle now right?

[analog obsidian]

its just some hardware failure

[alpine onyx]

Got my browser open that's about it

[alpine onyx]

can't be. It ran perfectly before

and the moment I ran that batch file and did voice changer stuff

the sound immediately came

[flint vapor]

I have p3 with browser opened, but if you cleaned drivers idk, i can't go more through, i'm only Helper for AI, but i don't want risk more for cause more issues with pc, there are various things to try, like reset bios, remove cmos for seconds, but i not suggest you to do it, i hope that someone more competent that me will help you

[flint vapor]

if you try reboot after you did that commands before?

[flint vapor]

How much temperature and usage there is also on your gpu on task manager?

[alpine onyx]

50C

30%

[flint vapor]

What is using mostly your gpu?

Like this

[alpine onyx]

ehhh taskmanager doesn't really say that

[flint vapor]

It should say you in processes tab

scroll on right for see gpu

[alpine onyx]

ah didn't have it enabled

[flint vapor]

Oh try to enable it 😄

[alpine onyx]

well right now it's valorant. But finally enough after opening valorant the noise has stopped lol

but discord and browser are the second highest

[flint vapor]

It does mostly while gaming?

[alpine onyx]

no it does while not gaming xd

[flint vapor]

Oh so is like the opposite? So weird

[alpine onyx]

yep

[flint vapor]

Do you play valorant with low graphic settings?

[open pelican]

Guys

I Going To Play Geometry Dash

[flint vapor]

Ok

[flint vapor]

I think the gpu was already had a failure and probably the bat make you discover the issue, it happened that okada sometime make some know about some failure

[stable steppe]

Hy

[latent kettle]

Hello

So firstly, tell me your pc specs

[stable steppe]

1 TB

[latent kettle]

I mean, the GPU, CPU, and RAM

[simple ore]

nVidia gpu with 16GB VRAM + 48GB RAM if you want to have a good time

the more the better

[stable steppe]

Why?

Just give me tool 🥲
Image to video

[latent kettle]

Just to know is your pc is capable to run it

[simple ore]

"I want to move 50 tons of gravel from the quarry to my home. I only have a push cart. Just give me a shovel"

[latent kettle]

The tool run locally on your PC hardware

[stable steppe]

Ohh

What's the requirements?

[simple ore]

"I want to make chiken like KFC does. All I have an easy bake oven. Just give me a recipe."

you see how silly it sounds?

[latent kettle]

Bro's rosting skills day by day 📈📈

[simple ore]

#✨│ai-help message

[latent kettle]

Noobies told you. At least 32 gigs of RAM a 16gb vram gpu. More is better

[simple ore]

32 is barely enough, most of the time you'd have to close every other application

[latent kettle]

Yep, I was going to suggest hin Framepack GUI

[simple ore]

framepack (with proper offload and quantization in SD.Next) can run on 6GB VRAM

but you really need 48GB RAM

it barely squeeks by

[latent kettle]

But it takes ages 😭

[simple ore]

I did not say it gonna be a good time

[latent kettle]

That's why. I think it's not recommended

[simple ore]

on 4070tis it does ~1sec video per minute

depends on resolution though

[latent kettle]

On 4060 it takes 7 minutes for 512x512 only for 1 sec

[simple ore]

with original framepack?

[latent kettle]

Yep

GitHub - lllyasviel/FramePack: Lets make video diffusion practical! https://github.com/lllyasviel/FramePack

This thing

[lethal iron]

who knows how to get comfyui working on amd gpu but rn its running on its cpu rather than gpu

[simple ore]

yeah, the original one is not the best

SD.Next folks did a lot more improvements to the workflow and offloading and quantization

[latent kettle]

Which one you recommend to use

[simple ore]

SD.Next dev branch

[latent kettle]

Lemme see

Is it in comfyui ?

Or is it a custom model ?

[unkempt timber]

Does anyone know how to get a good and free voice changer for Android and doesn't require pc or any kind of computer

[stable steppe]

I have 16GB

[simple ore]

https://github.com/vladmandic/sdnext.git

[simple ore]

16GB what? VRAM? RAM?

[simple ore]

realtime voice changer?

[unkempt timber]

Yea

[simple ore]

not without using some online service

[latent kettle]

https://github.com/comfyanonymous/ComfyUI#directml-amd-cards-on-windows

[unkempt timber]

On like a browser?

[lethal iron]

but i have directml but it shows that its running on cpu

[simple ore]

yeah, run the actual voice changer on Kaggle, use browser on your phone

[latent kettle]

That's what it means. It took gpu as CPU maybe.

[unkempt timber]

Ok

[simple ore]

I dont even know it would work on android though

how do you even route audio there?

[lethal iron]

so i dont need directml or what do you mean

[simple ore]

depends on amd gpu, there's ComfyUI-Zluda

https://github.com/patientx/ComfyUI-Zluda

[lethal iron]

i have rx 6750xt

[simple ore]

perfectly fine

directml is slow

[latent kettle]

If it doesn't work get comyui zluda

[lethal iron]

so i need comfyui zluda?

[latent kettle]

Yepp

[simple ore]

hip sdk + libraries for the gpu + comfy zluda

read the page

[lethal iron]

I will try in a week since im going away for a week and its too late to do it right now

[latent kettle]

No it not gonna work due to virtual cable support on Android

[lethal iron]

I love you dr jr

[latent kettle]

Okay server is opened.

[latent kettle]

Credits goes to noobies and Dr. Jr

[unkempt timber]

I don't know how

[latent kettle]

For windows you can use virtual cable but unfortunately, it's not available for android

[unkempt timber]

Oh ok

What about iOS?

[simple ore]

likely even worse 🙂

[unkempt timber]

Dang

Then what about a voice changer for videos

[latent kettle]

There are few chances for Android in future, but IOS naah

[unkempt timber]

Oh ok

[latent kettle]

What type of videos ?

[unkempt timber]

Like making my voice sound like characters (won't post tho probs just for friends)

[latent kettle]

Yep you can do it without any problems. But you can't do it real time.

[unkempt timber]

Oh ok is there any apps or sites I could do it for free?

[latent kettle]

Do you have a good pc or a gpu

Like a 2070 or something similar

[unkempt timber]

I don't have a computer

[flint vapor]

I don't think you can use it on phone, you need same vac, also if you can run on cloud

I don't think exist something similiar to vac on phone i not sure

[unkempt timber]

I would say 11 Labs but I don't know how to use it

[latent kettle]

That's a TTS

[unkempt timber]

What

[latent kettle]

Text to speech

[flint vapor]

That is not a voice changer

[unkempt timber]

Oh I thought they were a feature for voice changing

[latent kettle]

You can use applio on kaggle or colab

[flint vapor]

Do you want voice changer in real time?

[unkempt timber]

Kinda

[latent kettle]

Yes, but can't, czz Don't have a pc

[flint vapor]

You need only pc because it requires vac virtual audio cable

[flint vapor]

Yes true, he can only do voice conversion on colab or kaggle if he want

[latent kettle]

Bro I already explained

[unkempt timber]

What about a computer laptop from my school would that work?

[flint vapor]

If you use it on kaggle or colab yes

[unkempt timber]

Ok

[latent kettle]

Can you use discord on school pc ? Or can you talk to someone?

[unkempt timber]

Zoom?

[flint vapor]

What?

[lethal iron]

do you install it as normal comfy as in git clone

[unkempt timber]

Zoom is an app and site where I could call

[latent kettle]

https://github.com/patientx/ComfyUI-Zluda?tab=readme-ov-file#setup-windows-only

[flint vapor]

I know what is, but what it mean now zoom?

Dr Jr was asking you how can you use school pc

[lethal iron]

theres a step by step tutorial you mean

[unkempt timber]

Oh

[latent kettle]

Yep

[lethal iron]

oh thank you again

[lethal iron]

you were a big help🙏

[unkempt timber]

Brb

[latent kettle]

Tq for appreciation

[latent kettle]

I mean, how can you use voice changer in school? I mean for what purpose. Like you can't talk with friends on school pc or even can't play games so what is the purpose of doing it ?

[unkempt timber]

Oh I can call my friends but only on zoom though

Or Google meet I think

[maiden wren]

yo wsp

[latent kettle]

Yep you can, it's good if you want to try, but if you got caught doing this ? Also you need a phone number to verify kaggle account. And you have to install voice changer every time when you want to use it

[unkempt timber]

Oh ok

And I can use my number

[latent kettle]

Yes you can

[unkempt timber]

And I'll probably only use the voice changer either during break or when I'm at home

[latent kettle]

How can you access your school pc at your home? Or you can borrow laptops from your school?

[unkempt timber]

They give us laptops for us to do our work at school and at home

[tepid moss]

Since we are talking about compute a quick question, I would like to know if GPU manufacturer makes a difference in terms of optimization?

[latent kettle]

Got it. Good.

[simple ore]

you follow the install instructions

https://github.com/patientx/ComfyUI-Zluda/issues/188

[latent kettle]

Yes they do. They release newer optimization in Ray tracing dlss 4 and stuff to optimize the games

[tepid moss]

I was talking about in Ai applications. Lets use t2v for example

[latent kettle]

Text2videos ?

[tepid moss]

Yes

[simple ore]

on windows - nVidia is the king, AMD takes an effort, but hopefully the'll release their ROCm for windows sooner than later

Intel is their own thing

[tepid moss]

We can expand to all the image based AI stuff, Image2Vid or normal image gen

[simple ore]

on Linux both AMD and Nvidia are fine, nVidia drivers take more effort to set up

[tepid moss]

Uhuh, continue...

[simple ore]

other than that they are both serviceable

t2v, i2v, etc

[tepid moss]

Ok, ok good. Looks like I'll won't have too many problems with these.

[simple ore]

nvidia is skimping on VRAM, so AMD is better in this regard

more vram is better for any video models

[tepid moss]

Now I want to talk about something more specific. What about apple users? Products like the new MacPro have their own unique silicon. What would be in this case?

[simple ore]

you not gonna do any serious AI on it

[latent kettle]

You should discuss this stuff in #🧬│ai-chat

[tepid moss]

I've seen clusters of macpros and minis as servers for compute, not for AI but definetly for performance applications.

[tepid moss]

Lets go there

[tepid moss]

Just the open source model level stuff

[cedar sky]

hi guys, does anyone know how to run this model and create an api for it? I've spend 4 hours to trying to figure out
https://huggingface.co/mradermacher/WebSailor-3B-GGUF

[latent kettle]

@simple ore is there any nunchaku thing for video generation?

[cedar sky]

but if I want to be this on cloud and use as api to make a requests how I can do this?

I've tried to deploy throw huggingface and it's now working, the request said is isncorrect

[simple ore]

you can use llama.cpp + this model in the cloud

google colab

replace gemma with your model here https://medium.com/@AleNunezArroyo/run-google-gemma-llama-cpp-gguf-inference-in-google-colab-ecb1c03bc440

[latent kettle]

Is there @simple ore

[simple ore]

it is quantization

SDNQ in SD.Next

[latent kettle]

What I supposed to do to use it

[cedar sky]

thanks, but codelab would be dead if I'm not using it, I wanted to use huggingface cuz they have autoscaling and sleep mode if model is not using

[simple ore]

https://www.youtube.com/watch?v=wOix7DlZ4EY ?

generally you run llama server in the cloud and you access it using the standard API

[cedar sky]

I think this is what I've did, lemme try again and who result. Thanks!

[cedar sky]

yeah, it's to use a model from local machine, but I want to run this on cloud and then communicate with it throw api

[pearl lion]

Hey guys I wanted a way to parse pdf documents and extract the title and the headers.
So given a particular pdf, I need to extract the pdf in such a way that:
{
"title": "This is the title",
"outline": [
{ "level": "H1", "text": "Introduction", "page": 1 },
{ "level": "H2", "text": "Main content", "page": 2 },
{ "level": "H3", "text": "Conclusion", "page": 3 }
]
}
It unfortunately needs to be complately offline and the model size(not including dependencies) must not cross 200mb.
Would really appreciate any help!!!

[simple ore]

same thing

get a VPS with GPU acceleration

install stuff to run llama.cpp server with you model and provide API acess

and a tunnel so you can use to access it

https://github.com/jojje/runpod-gguf-server

a docker container will do

[cedar sky]

thanks, will try now 🙂

[twilit forge]

why is my applio doing this

can any1 help

[edgy rapids]

can some one help plz

where do l get the rea fle

cuz mlne ls not worqlng

[viral mason]

what happened

are you still alive

[twilit forge]

does anyone know how to fix

[simple ore]

download the compiled version

https://huggingface.co/IAHispano/Applio/blob/main/Compiled/Windows/ApplioV3.2.9.zip

[twilit forge]

can i still locally train of this

i cant even download python its fucked

[twilit forge]

how tf

i downloaded it

@simple ore

[simple ore]

do not run as admin

just unzip and start run-applio.bat as a normal user\

[west cave]

should I use client or server mode

I think client sounds worse

[frozen ledge]

-colab

[patent trellis]

📒 Google Colab Notebooks

Google Colab is a Cloud (Remote Good PC) Service. While the Free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.

• **Applio**

by IA Hispano
Google Colab

• **RVC Mainline**

by Hina
Google Colab

• **UVR5 NO UI**

by Eddy
Google Colab

• **UVR5 UI**

by Eddy
Google Colab

• **Wokada Deiteris Fork**

by Deiteris & Hina
Google Colab

• **Hina's Modified Original Wokada**

Google Colab

• **RVC-AI-Cover-Maker-WebUI**

by Shiro & Eddy
Google Colab

• **FaceFusion UI**

by Nick088
Google Colab

• **FaceFusion NO UI**

by Nick088
Google Colab

• **Music Source Separation Training (Inference)**

by Jarredou & Makidanye
Google Colab

[fleet cedar]

u can get reported in discord for showing u got vencord or msth lol

[viral mason]

don't be stinky

[knotty moth]

https://tenor.com/view/choso-discord-gif-13300109348650345658

[viral mason]

true

can't use em here at all

[quiet flint]

Anyone know how I can fine tune the helpful assistant tone out of Phi4-MINI (yes, mini version, not the 14b one), without lobotomizing it’s function calling capabilities. How can I make Phi4-Mini better at roleplay/not sound robotic.

[paper lotus]

yup

[paper lotus]

how would i do that

which one do i get

[knotty moth]

depend on your gpu

• cuda: Nvidia RTX
• dml: AMD/intel for windows
• rocm: AMD for linux

[paper lotus]

no i meant like

he said get all the files manually

how do i do that

[maiden wren]

dont need all files

[paper lotus]

.

[paper lotus]

i need it tho

cause its not downloading it

[knotty moth]

then he didnt mean that

[maiden wren]

okay bro then try

[paper lotus]

can u go online

i need u to respond

[knotty moth]

what

[knotty moth]

in case of the failed pretrain downloads, try again or use manual download as noobies said

before retrying, delete the pretrained folder first

[merry crypt]

yall i need help with realtime voice changer

[open pelican]

I For Got Faster

[maiden wren]

yoo .

[timber silo]

How can I go about getting better at ai?

wdym

[maiden wren]

just pracitce bru learn stuff off of internet

[knotty moth]

this is a channel for support on AI applications

not for random chit chat

[maiden wren]

💔

[knotty moth]

so if you have any question about it, just ask

[maiden wren]

ik bro im not a helper💔

im a chit chatter

[knotty moth]

!howtoask

[patent trellis]

How To Troubleshoot

__**GIVE CONTEXT.**__ 📝

• Don't simply mention your issue, like "my rvc is not working".
• Describe the step you are on, what you're trying to do, the RVC you're using, a screenshot, etc.
• The more context, the better.

__**BE POLITE.**__ <:matsuripray:1159685390156967936>

• Don't be desperate. You can ping a Helper, but if they ignore, they aren't available/don't know the answer.
• It's okay if you're frustrated, but don't take it into this server.
• Don't DM without prior consent.

__**BE PRODUCTIVE.**__ 🤝

• Don't ask for every little instruction. Put your own effort & test things by yourself.
• Don't ask to ask.
• Check if your answer is a Google search away/on our guides website.

If you've changed your mind about your message, feel free to delete/edit it.

[maiden wren]

💔

ikt already bro

[pastel oak]

Just ask, someone will help when they see it

[pastel oak]

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

Help section

[open pelican]

I Did It 🥰

[mystic tangle]

should i use vonovox or deiteris

[craggy bough]

good job

now you can stop spamming the chat asking for it

[open pelican]

Okay, I So Sorry

[knotty moth]

• the former is only available for Nvidia GPUs but can also work on RTX 5000 series
• the latter is recommended for starters and supports Nvidia, AMD and Intel gpus

[formal wind]

Samples dont actually work in kaggle atm, anyone else have this issue?

[simple ore]

what are you trying to do?

[formal wind]

use inference to make a sample

[oak edge]

I need some help after seperating vocals the vocals only song has this slight robotic hint to it

are there anyways to make it natural?

[simple ore]

turn off auto-tune

[formal wind]

tried that

its been doing it all day

[simple ore]

this

disable

[formal wind]

I did

It still doesnt work

[simple ore]

checkout 3.2.9 branch then

you've pulled the latest code with a bug

[formal wind]

how tf do I use that version

here?

[simple ore]

the code shoudl've been !git clone --depth 1 $uioawhd $new_name --branch 3.2.9

[formal wind]

bet

[simple ore]

I'm not sure how to look at the code files on Kaggle

otherwise there's a small manual code fix

[open pelican]

Play Here

[flint vapor]

Now just stop

This is ai help channel, we are not interested in the model you did, you spamming this from a week almost

[knotty moth]

https://tenor.com/view/mr-bean-listening-mr-bean-nds-nds-fingers-in-ears-gif-1138282966845560934

[open pelican]

Hey, Can You Stop It!

[flint vapor]

No you have to stop, it's from a week almost that you keep spamming this model

[open pelican]

👊

[flint vapor]

We understand you did maeko model, but this is an ai help channel, not for listen models

[open pelican]

Ok, Delete To Maeko Ai Voice

[formal wind]

yeah still doesnt work..

[flint vapor]

I can't delete your model, is just that you don't have to spam here

[knotty moth]

ok deleted

[simple ore]

okay, close the notebook, and re-do... I've fixed the main branch

[formal wind]

Alrighty

is the code supposed to be different btw

cause it isnt

[simple ore]

depends on how you're doing it

[formal wind]

wdym

[simple ore]

are you copying some notebook from kaggle, or are you importing it from Applio's github?

[formal wind]

github

[formal wind]

ill try again

[simple ore]

it should be pulling 3.2.9 branch

and there should be no such error there

[formal wind]

Doesn't even train now lol

Imma head to bed and hope it's fixed tomorrow or sumin, gn

[simple ore]

what

[simple ore]

show me something

[knotty moth]

post a thread in https://discord.com/channels/1159260121998827560/1192011222023950368 and explain your problem on the training

[tribal tangle]

Is there any AIs and yt tutorial for this.
I basically want movie characters to speak in their own voice with emotions like sad, happy, excited.. maybe option to Laugh, sigh etc to make it more realistic

[keen crescent]

• rvc

[patent trellis]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[hallow thistle]

You mean like to make an AI vocal of a character to speak from a given audio? There's RVC. If you mean by TTS, Applio has TTS feature built in but you might wanna download a voice model for this one.

[tribal tangle]

Yes but I got no idea what rvc or appilio is. is there a video tutorial maybe

thanks for the reply

[odd shale]

I don't think there are any tutorials for that.

[odd shale]

Nope, we strongly suggest to not follow any yt tutorials since these tend to get outdated quickly.

But no, with applio you can only replicate a character's voice, and there are no controls for emotions. (neither you can't laugh)

[tribal tangle]

How do I learn then

[hallow thistle]

The correct name for the program is Applio. Yes, Applio RVC can convert audio to sound like a character or a person. The emotions depend on the audio track you provided there; there is no manual emotion controller as of now.

[hallow thistle]

There's a guide.

[tribal tangle]

ok

[odd shale]

You can start by reading the docs.

-rvc

[patent trellis]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[tribal tangle]

ok thanks

[hallow thistle]

[odd shale]

You're welcome.

[shut stratus]

what do i use to t=now train a voice model, i have just over 10 minutes of uninterupted audio

[hallow thistle]

Applio RVC can train a voice model.

!howtoask

[patent trellis]

How To Troubleshoot

__**GIVE CONTEXT.**__ 📝

• Don't simply mention your issue, like "my rvc is not working".
• Describe the step you are on, what you're trying to do, the RVC you're using, a screenshot, etc.
• The more context, the better.

__**BE POLITE.**__ <:matsuripray:1159685390156967936>

• Don't be desperate. You can ping a Helper, but if they ignore, they aren't available/don't know the answer.
• It's okay if you're frustrated, but don't take it into this server.
• Don't DM without prior consent.

__**BE PRODUCTIVE.**__ 🤝

• Don't ask for every little instruction. Put your own effort & test things by yourself.
• Don't ask to ask.
• Check if your answer is a Google search away/on our guides website.

If you've changed your mind about your message, feel free to delete/edit it.

[hallow thistle]

You mean W-Okada the realtime voice changer, right? Which W-Okada version are you using? Are you using VB-Cable?

VB-Cable gives random issues for Windows users. What is your PC GPU? The W-Okada version you're using is the original and old one. There's a better one to try.

[wise fiber]

hey guys how do i use colab for the RVC and the voicechnager thingy

for w okada n tuff

[agile ingot]

guys what is the best voice to troll my guy friend?

[hallow thistle]

The "brr skibidi dop dop dop yes yes" voice model.

[agile ingot]

no like a girl voive

voice

[hallow thistle]

If you're looking for an E-girl voice model, can't do much about that. I mostly use anime girl voice models.

[wise fiber]

@hallow thistlehey uh

you know how to use google colabs?

heard my laptop wasnt strong enough

because i dont have a gpu

[hallow thistle]

There are guides on how to use Applio RVC or W-Okada respectively, but you might wanna pay for Google Colab Pro to use either of these, because attempt to use free tier of Colab can get your session suddenly disconnected or your account terminated from using the service. Kaggle is another option, but you might wanna register with your phone number on that one.
https://docs.aihub.gg/rvc/cloud/applio-colab/
https://docs.aihub.gg/rvc-voice-changer/cloud/deiteris-w-okada-fork-colab/
https://docs.aihub.gg/rvc/cloud/applio-kaggle/
https://docs.aihub.gg/rvc-voice-changer/cloud/deiteris-w-okada-fork-kaggle/

[hallow thistle]

At least wait for my response, I was typing.

[wise fiber]

what do you mean by account terminated?

[wise fiber]

aplogies

[wise fiber]

like my gmail account just goes bye bye?

also uh,the kaggle thing,when i try to verify my phone,it doesnt let me verify it,it just says unable to verify,we need to verify your identity now stuff liek that

[maiden wren]

Only fun right bro

[maiden wren]

I meant only for fun right

[agile ingot]

Yes trolling my friend that Are a simp

[hallow thistle]

The "terminated" on Google Colab means you won't be able to use any service on Google Colab with your account, permanently until you appeal to reinstate it. This one won't impact any other of your Google account features. https://i.imgur.com/bbDvApa.png

[maiden wren]

Ok dont use for anyone else then only for fun

Server doesnt accept catfish

[agile ingot]

Not that saying she is my gf and trolling him

[maiden wren]

Ohh good then

[wise fiber]

oh ok,ty!

[maiden wren]

Shii

Mb

[maiden wren]

There are bunch yu can find persoanlly id go to weight.gg and search girl voice and 1st result that pops up works

But i normally dont use girl voice

I use ones that are meme and fun

[hallow thistle]

While the question "what is the best girl/female voice model to troll my friend" sounds suspicious to me, "to troll my friend for fun" is one of the only exceptions I would accept from anyone. Anything else bad is no.

[agile ingot]

When i downloaded it and pressed the file it Said contents when do i do?

[hallow thistle]

If Kaggle couldn't verify your phone number, it could be either your phone number was invalid, you set to a wrong phone dial region or your phone number was blocked in your country. While there's an option to verify your identity there with Persona, I'm not sure how this would make you to be able to use service as verifying phone number.

[hallow thistle]

Any elaboration on your current step?

[viral mason]

Use Kaggle don't pay for collab pro 😭

[wicked mica]

my thingy randomly sometimes takes a lot of time to convert input voice suddenly, why is that ? :<

[low shard]

This is a General AI Server, AI has many fields, so we can't know your issue

Please Elaborate:

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

[quiet flint]

#1396890571985129482

helpies pls

[wicked mica]

-Rtx 4050
-Windows 11
-recently its taking longer time for voice conversion even though i didn't change any setting or models

• One discord friend helped me set up

[low shard]

what version are you using? show a screenshot of the folder of the program

[low shard]

seems like wokada deiteris fork b2332, it should show cuda b2332 at the top of the program, right?

weird that it opens in a window program for u tho

it should open in the browser

[wicked mica]

it opens on the browser by default

but then i click the older version and the app turns on

[low shard]

• uncheck sup1, its useless
• set extra to 2.7, to not cause any cutoff issues
• uninstall vb audio cable, it causes crackling and random windows issues, get vac lite https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#virtual-audio-cable

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]

I wouldnt really suggest that btw

i would suggest to just use it on the browser

it may cause issues

[wicked mica]

okk thanks for the help !

[low shard]

You're welcome Frieza, let me know if its all fixed!

[quiet flint]

can you guys help pls: #1396890571985129482

[hidden crane]

wrong server sorry

[quiet flint]

?

[harsh violet]

when the page opens it keeps loading like forever any solutions ?

[low shard]

This is a General AI Server, AI has many fields, so we can't know your issue

Please Elaborate:

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

[harsh violet]

• 3060
• win 11 pro
• use the real time voice changer
• https://rentry.co/ForkVoiceChangerGuide
• I don't have permissions to

[cedar sky]

@simple ore as we yesterday discassed I've tried to deplay directly on huggingface with Text Generation mode, is this correct?

ah no, maube it's working

do not understand how to override right now keys which is described in docs, cuz all those docs created for alibaba cloud https://github.com/Alibaba-NLP/WebAgent?tab=readme-ov-file

[harsh violet]

@low shard

[low shard]

that's the old guide link, https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/ is the new one

also, what are you looking to do to understand what program u need?

• TTS
• AI Covers
• E Girl trolling/catfishing
• Roleplay with realtime voice changer
• etc

I'm asking because many people confuse RVC with other things lol

[quiet flint]

guys how can i run kimi-k2 locally on a chromebook cluster of ~2k chromebooks (dont ask about the number of chromebooks)

[frank slate]

@torpid blade

[flint vapor]

There is this but i don't know if is too much powerful https://huggingface.co/unsloth/Kimi-K2-Instruct-GGUF

[wintry hearth]

hihi im having issues with find-model chat

it telling it didn't respond when i ask it for a voice

looks like this

am i missing a step?

[low shard]

could you please try restarting your discord?

what did you search for exactly?

[wintry hearth]

loki

ill go goku'

it worked O.o

one sec

[low shard]

wtf I just noticed there's an issue only when searching for that 😭

[wintry hearth]

lol

[low shard]

it might be better you report this in discord.gg/weights , or dm BeatsBoy, since its their bot

[wintry hearth]

ok thank ill do that

[low shard]

you're welcome

[pearl spruce]

Can i ask u something in dm?

[low shard]

Hello, is there a specific reason why in dms? If it's something related to help, it would be better to do it here

[harsh violet]

I'm using the same version

[pearl spruce]

I mean maybe its a weird question so its why i take it too dms 😭

so can i add u and after u done answering u can remove me

[lament gale]

hi i cant hear myself when trying to use the real time voice changer

[low shard]

that guide had some old info tho, could you reply wqith what u want to do?

[low shard]

another staff informed me about it. We don't allow any type that could be possibly related to NSFW, don't dm anyone about it, we won't help

[viral mason]

nsfw ai usage is for losers

love how many idiots are in this server tbh, so easy to catch them

[formal wind]

what do you need to see

[simple ore]

what exactly is breaking

log from kaggle

[formal wind]

alr gimme a sec

[formal wind]

A whole chunk 😭

[simple ore]

what batch size did you use?

[formal wind]

4

[simple ore]

all I see warnings, not an actual error

[formal wind]

It just started training now?

wtf is up with this code 😭

[simple ore]

it is nvidia on linux bs

[formal wind]

what

[low shard]

This is a General AI Server, AI has many fields, so we can't know your issue

Please Elaborate:

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

[formal wind]

does kaggle use nvidia?

[simple ore]

T4

what car do you have?
"honda"

[orchid mantle]

yhh what file do i use to open fumiama?

[simple ore]

open task manager, performance tab, click the GPU, make a screenshot or just read the name

any other?

well, you're trying to carry 50 tons of gravel in a 5 gallon plastic bucket

not gonna happen

you need an actuall GPU that can accelerate AI tasks

you can try DML version

but you would not be able to do anything else while it is running

https://github.com/deiteris/voice-changer/releases/download/b2332/voice-changer-windows-amd64-dml.zip

[low shard]

thats outdated asf

it uses an over year old version of original wokada

and vb audio cable creates issues on windows

delete the zip, folder and uninstall from windows app settings

[low shard]

fumiama?

could you please elaborate?

[orchid mantle]

https://github.com/fumiama/Retrieval-based-Voice-Conversion-WebUI/blob/main/README.md

[low shard]

is there a specific reason why you want to use that fork? why do you want to do?

[orchid mantle]

i've been looking for a good one to make an ai cover and saw people talking that it works the best

[low shard]

we don't suggest that rvc fork, it worked only for 1 person, i tried it also myself and it was broken, iirc it had some broken stuff and might still have them

could you please tell me your pc gpu and OS?

[analog obsidian]

fumiama's custom fairseq broke training, the models sound very choppy lol

[orchid mantle]

oh okay

[low shard]

if you tell me your pc gpu and OS, I can help you get a better one

[viral mason]

never even heard of this, where'd you find it?

[simple ore]

it is a known fork

[orchid mantle]

yhh i cant share screenshots

[low shard]

It's a fork made bc of drama about PRs

[orchid mantle]

but im on windows 10

[low shard]

!give-media-perms 1h @orchid mantle

[analog obsidian]

its a fork made by one of the members of the original rvc team

[viral mason]

ohh ok

[orchid mantle]

hope thats what you mean??

[viral mason]

amd user prob

[low shard]

nope thats the cpu, could u please click down here?

to check the full name

[simple ore]

3060 8gb

[low shard]

great

[orchid mantle]

mind you, i dont want to create a model, just use one already created

[low shard]

https://docs.aihub.gg/rvc/local/applio/

use applio

it's an rvc fork which is easier

[orchid mantle]

thank you <3 im a huge newbie so dont mind me if i sound very stupid 😵‍💫

[low shard]

there's also https://docs.aihub.gg/rvc/local/aicovermaker/ but this is meant ONLY for automatic ai covers, easier but depends if u want to do it manually with applio (which has also other stuff such as TTS)

[orchid mantle]

okay, ill try to figure it out, big thanks again

[low shard]

you're welcome, let me know

[orchid mantle]

uhh, idk whats wrong, the voice loaded okay but no youtube links work

[simple ore]

look at the log if there's any

if it is colab/kaggle, the error is there

[orchid mantle]

it only says it cant find a file

but im using a youtube link so idk how that would work

[simple ore]

it is likely looking for ffmpeg

to extract the audio out of this video

[orchid mantle]

i've tried multiple links and it gives the same error

[simple ore]

https://huggingface.co/IAHispano/Applio/tree/main/Resources

download ffmpeg.exe and ffprobe.exe and drop them into the folder with the app

[orchid mantle]

okay, ill try, thank you

[simple ore]

that's my guess, but I usually make good guesses 🙂

[orchid mantle]

:/ still

[simple ore]

sec

ah, you need "yt-dlp"

(ffmpeg too, after)

[orchid mantle]

where do i get it?

i added ffmpeg and ffprobe

[simple ore]

https://github.com/yt-dlp/yt-dlp

grab .exe from the releases

https://github.com/yt-dlp/yt-dlp/releases/download/2025.06.30/yt-dlp.exe

[orchid mantle]

omg it works

thank you!!

[zenith spade]

anyone know any rvc realtime voice changers that are trusted?

[viral mason]

do u have an nvidia gpu? there's two that are pretty cool and trusted

[zenith spade]

i have an nvidia 4050 laptop gpu

so uh ye

[viral mason]

should work for either then

think If I do this it'll work

-realtime

[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

[zenith spade]

let me try cramming a 4050 desktop gpu in here lol

i couldnt think of vonovox lol

i knew what i was thinking abt but couldnt think of the name

[viral mason]

follow either the first or second link, original is dead basically

btw vonovox is being updated still deiteris is not

both work well tho

[zenith spade]

im gonna try setting up vonovox for the 1st time and be dedicated this time lol

[viral mason]

you got this

it's pretty good although you can't upload the index file for models yet, that's next update

alsooo

from dr87

themself

[mortal sentinel]

alguem me passa o servidor brasileiro

[young halo]

what is the code for afk in colab?

[viral mason]

?

[zenith spade]

?

[viral mason]

how's vonovox going btw

[zenith spade]

still setting up

extracting tkinter and downloading more assets rn

[viral mason]

I shall keep waiting, hopefully it gives no issues

[zenith spade]

it just finished installing stuff

[zenith spade]

literally just as i sent this message

yo this had a ui update since i last used it

[young halo]

there's a code that makes your screen move, avoiding you to get disconnected for inactivity, i've used before, but i dont remember it

[viral mason]

when was the last time u used it? I only recently knew it existed

[zenith spade]

back around may/june i dont remember

it has model presets now

which i didnt see when i last used it

[viral mason]

this year?

[zenith spade]

yes

[viral mason]

ye that was added to the current 1.5.0 update I believe

[viral mason]

next update is the index update along with this stuff

[formal wind]

Then all we would need is an amd support

Us amd users need some love tol

Too

[viral mason]

true true

join the vonovox discord see if u can talk to dr87

[formal wind]

Send a link?

[viral mason]

dms since no discord links here

[formal wind]

Ya ya

[viral mason]

what are you going to use it for?

are you trying to become an e girl with it or do something like spongebob

[craggy bough]

too easy 🥀

[viral mason]

so easy to catch em

[formal wind]

What is with people wanting to be e girls mannn

[low shard]

Please Elaborate:

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

[low shard]

vonovox has its own discord server?

[viral mason]

yup

[upper socket]

@long forge I NEED HELP

[viral mason]

I hope you get set on fire

[eager wadi]

i swear these guys are getting less convincing by the day

[viral mason]

they have stopped trying

[eager wadi]

when i meet one that has the language button be functional, then ill be content

[viral mason]

finally they got kicked

[eager wadi]

is it too much to ask for a convincing scam so i can feel accomplished for spotting it