✨│ai-help

[lost vector]

do you have any idea btw?

[viral mason]

not sure, I only use nvidia

[minor saffron]

hii im using a nvidia 4070 mobile version with windows 11 trying to use w-okada to change my voice and i get NotAllowedError: Permission denied by system
no stack error on every browser i use except firefox

but on firefox when i press start nothing happens

[cold turret]

how come almost every time i talk it cuts out

[low shard]

!give-media-perms 1h @lost vector

[dull iron]

Oops, We got some errors!

• [Error] Giving <@&1313370885650124821> to @lost vector for 1h

[low shard]

gave u temp perms

wait wtf

!give-media-perms 1h @lost vector

[dull iron]

Oops, We got some errors!

• [Error] Giving <@&1313370885650124821> to @lost vector for 1h

[low shard]

Hello!
Please Elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • E girl trolling
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[low shard]

It's hard for us to understand your issue
Please Elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[minor saffron]

how do i send a screenshot..

[low shard]

!give-media-perms 1h @minor saffron

[low shard]

now you can, please elaborate everything

[pastel orbit]

https://canary.discord.com/channels/1159260121998827560/1314241091779825684

guys this one its not stable why

sometimes you hear normal then voice you hear it sooooooooo low sometimes you never hear anything at all

no delay .. all other models work fine

idk whats wrong is it my settings or is it something else

[viral mason]

Use snipping tool

[placid grotto]

anybody have a reccomended model for egirl trolling?

[low shard]

E girl trolling is catfishing, which is illegal. You have been warned both verbally and via sapphire and continuing to ask will lead to further actions possibly like a ban

[viral mason]

People like you are the reason ai is hated

[placid grotto]

but you said e girl trolling as a subcategory

[low shard]

it's written in the server rules and help guidelines that e girl trollers aren't allowed, also i didn't talk to you before?

[placid grotto]

ok sorry then

[dire topaz]

Hey new update from vonovox just came out can I update the existing file or do I need to reinstall the whole program?

[low shard]

extract the new one to the same folder and replace the files

[dire topaz]

Saw that as well but which ones do i replace only the setup and start ones or everything?

[shrewd vortex]

IndexError: list index out of range
PermissionError [Win32] the process cannot access the file because it is being used

then it just closes the console and opens nothing

[royal jacinth]

does anyone now how to fix this when using zluda applio on the version of zluda they provide and the second error is when I use latest zluda

[unique burrow]

my audio isnt owrkingh

working

[sterile tiger]

i click start_http and nuthin happens

and if i click on the native client its just a blank screen

[tranquil stone]

literally any model that isnt vanilla beatrice doesnt work?

[minor saffron]

no image perms

lol

[shrewd vortex]

audio input causes it to refresh Error
unhandledrejection
no error stack
NotSupportedError: AudioContext.createMediaStreamSource: Connecting AudioNodes from AudioContexts with different sample-rate is currently not supported.
98/t/Un/this.setup@

[quaint flame]

yo i need help so i use the voice changer and it used my real voice when i can hear the vocie changer talking but my friedns hear my voice how do i fix?

[minor saffron]

set ur input to ur voice changer

on the app ur talking to ur friends with

can i get image perms to show my issue

😔

[shrewd vortex]

everytime i press on input or output it just gives an error

[short torrent]

Vonovox just closes whenever I press start how could I fix this

[viral mason]

Try getting Nick in here

He usually does that stuff

Pretty sure only mods can give out image perms

[knotty moth]

I think "e-girl trolling" shouldn't be mentioned if it's not allowed

[verbal pier]

i need some help with my ai voice

can someone help me

rq

itll be super quick

hey sky high can u help me rq

im having this issue where my anime voice is sounding WAY too AI like and i need it to sound like an actual girl

[knotty moth]

bruh ur too rushed

[verbal pier]

can i get some help?

it just sounds too ai like

[verbal pier]

pretty please

with a cherry on top

[knotty moth]

[verbal pier]

do u not know? 😭

[knotty moth]

@quaint flame please ask here or post a thread in #1192011222023950368

@quaint flame @verbal pier
regarding the voice changer, please elaborate:

• your GPU & pc spec and operating system used
  - what purpose you want do for? like e-girl/boy trolling or not?
• what kind of voice model are you going to use?
• tutorial link you're using and referring to

[low shard]

could you please elaborate more?:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[low shard]

Hello :)
could you please elaborate more?:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[low shard]

you seem to be using an outdated version of original wokada, and beatrice models are lightweight but lower quality, please elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what tutorial link are you using
• a screenshot of the program

[low shard]

start_http is apart of the old original wokada, don't use it,
could you please elaborate more?:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[low shard]

wdym?

[low shard]

@simple ore you got any ideas? I think you used Applio Zluda before, ofc if ur free

[low shard]

are you free now? because the perms are temporary

[simple ore]

you need to follow the instructions. if you're using 25.5.1+ driver, then this

[shrewd vortex]

Error
unhandledrejection
no error stack
NotSupportedError: AudioContext.createMediaStreamSource: Connecting AudioNodes from AudioContexts with different sample-rate is currently not supported.

what do i do

[royal jacinth]

Using a 5700xt gpu

[short torrent]

GTX 1070 Ti Windows, I wanna use character voice models in games, where do I send a screenshots??
I used the installation guide at https://docs.aihub.gg/realtime-voice-changer/local/vonovox/

[low shard]

vonovox and wokada deiteris fork b2332 are 2 different programs

!give-media-perms 1h @short torrent

[short torrent]

Oh wrong lick

link.

[shrewd vortex]

nick my lord and savior help me

[low shard]

are u sure
input: microphone
output: line 1
monitor: ur headphones optionally?

[shrewd vortex]

yes i only get that when i choose any input

but everything else works fine

[noble meadow]

The voice doesent work and i doesent pick up anything

[low shard]

Hello :)
could you please elaborate more?:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • E girl trolling
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[low shard]

have you tried other browsers?

[shrewd vortex]

yup does the same w chrome

[noble meadow]

Im trying to e girl troll and the program im using wont pick up my voice

[shrewd vortex]

ill try reinstalling since that fixed the first error w the console

[short torrent]

I believe I just had the sample rate of my headphones wrong, srry

[low shard]

E girl trolling is catfishing, which is illegal. You have been warned both verbally and via sapphire and continuing to ask will lead to further actions possibly like a ban

Also if you came from a youtube video tutorial, they are all old, you can just delete everything since they use an over 2 year old program

[shrewd vortex]

out of every model why egirl.. peter griffin/stewie/cartman

[low shard]

that's pretty weird, but let me know :)

[low shard]

oh lol 😭

[knotty moth]

it is okay to use peter griffin, etc voice and let ppl know that you're doing so

but trolling with egirl voice is not allowed

[shrewd vortex]

seems to work now i cant tell if something bad happened or could just be my settings but some model works better on the original wokada

[low shard]

could u pls show a screenshot of ur wokada deiteris fork settings?

[minor saffron]

now i am

[low shard]

!give-media-perms 3h @minor saffron

[minor saffron]

like on firefox

its on start

but it never starts

its stuck on this

this happens on other briowsers like chrome

[violet slate]

when i talk people can hear my real voice then the ai one right after how do i make it so they only hear the ai @vernal bone

[knotty moth]

please elaborate:
- what purpose you want do for?

• what kind of voice model are you going to use?
• your GPU & pc spec and operating system used
• tutorial link you're using and referring to

[violet slate]

windows 10 just for trolling 1660 ti 😭 https://www.youtube.com/watch?v=SxdnGxicJOg&t=618s and this is the link

[knotty moth]

trolling what? using like e-girl model? or some well-known character/figure voice and you let ppl know that you're using that voice?

[violet slate]

im using MEA asmr model

[knotty moth]

and you let ppl know that you're using that voice, also receive consent with the voice owner?

[violet slate]

ok i will do

[minor saffron]

ahhh idk how to fix

[knotty moth]

sorry but that doesnt quite sound convincing

[violet slate]

so u want me to let people know if they ask what im using/model if so i will do

[knotty moth]

check some error message in the terminal window

also fyi deleting the model in the first slot might not work so you'd have to do it manually in the model_dir folder

[minor saffron]

nothing pops up

[knotty moth]

u mean the error message?

u could also try restart the program and try again

[minor saffron]

yep nothing shows up

just connected sid

i also tried restarting and doing a fresh install

same error

and on firefox it doesntg give error

but it just doesnt work?

😔

it worked completely fine a few days ago

[knotty moth]

try on the browser that might work on you

perhaps edge otherwise?

[minor saffron]

same error on edge

chrome

opera gx

brave

im running out of options lol

[knotty moth]

try getting fresh install of the program, letting it download the required stuffs

I'd recommend to ensure you should use the recommended version https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork

[minor saffron]

i tried installing default one and nvidia one

fresh install

both dont work

😔

[knotty moth]

mb I think you should try not to run as admin, and I'd recommend putting the program in a directory path like D:\ without any spaces or strange unicode characters, and check the permissions on the directory

[minor saffron]

ok ill try

[minor saffron]

it did not work

😔

[knotty moth]

I mean have you tried on a separate storage drive without permission issues?

otherwise:

• try scanning the entire system for possible malwares
• try fresh install windows as last resort
• you may have been using a work laptop/whatever with such restricted access

[minor saffron]

it works mostly fine ion firefox

but nothing happens when i press the start button

nothing pops up in console either

[knotty moth]

so firefox works on you

but did the "permission denied" show up when u pressed start?

[knotty moth]

^ ?

[minor saffron]

nope

[minor saffron]

its on start

bvut it doesnt start

and the console doesnt show anything

[knotty moth]

did the console seem to not have issues or not responding

[minor saffron]

nope

console isnt freezing

just doesnt show anything

this is last message on console

[knotty moth]

I mean did the "permission denied" issue no longer show up?

[knotty moth]

so it does mean it may work properly but you havent tried turning up the volume till you could hear?

[minor saffron]

no because im on firefox

the perf isnt showing and updating so i dont think so

it says 0 ping too

[knotty moth]

bruh

you've answered no to everything

[minor saffron]

lol

idk what to do

[knotty moth]

so the point is that it shows 0 ping

not just everything because ur on firefox

[knotty moth]

then check the input device should be a real mic and output to a virtual cable or headphones if u want to hear

[minor saffron]

worked fine on firefox before :D

[knotty moth]

it worked fine on firefox before but now it doesnt work on the same firefox?

because of the 0 ping crap?

[smoky arrow]

Hello anyone use realtime voice changer i had some issues where i cant hear the voice can anyone help me

@knotty mothhi

can you help me

No embedder my problem is this

i already upload both index and pth

[smoky arrow]

@viral masonhi can you help me with the realtime voice changer

[crisp lichen]

Chatgpt said there were ai here which made my lyrics adjust to the music i like listening to

So is there any?

[crisp lichen]

Also im on phone

[charred crater]

Yo guys, what linear regression means and what are weights and biases? I asked chatgpt but still didn’t understand

[brazen raven]

is there a way that you can make okado even better?

im alredy using high quality

[patent wind]

i wish it would sound good when your singing 😔 😔

[tropic pulsar]

Will there be an Okada for Mac M2?

[knotty moth]

never https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#download-mac-silicon

[austere cipher]

what do i set the input and output in okada and discord

[viral mason]

That like at 2 am for me man, I just woke up

[lethal iron]

How do i fix this and get my comfy working on amd

What tutorial do you guys use overall to install comfyui on amd?

[lethal iron]

rx 6750xt
windows 10
I was trying to install comfyui-zluda on my pc and i was trying to open it and this error came
https://www.youtube.com/watch?v=X4V3ppyb3zs and https://www.youtube.com/watch?v=gfcOt1-3zYk were the tutorials used

[minor saffron]

no

beacuse it isnt starting

😔

the start button does absolutely nothing for me anymore

console no show nayhthing

max volume max monitor

uhh

i double checked the mic and audio output

[spice iris]

Someone please tell me what to do if i wanna do voice conversion on acapella thats overloaded with layers, with what software is it possible to somehow select one layer? thank you

[flint vapor]

It can be very complicated do voice conversion on acapella that have many layers, the acapella you have just one voice or many ?

[strange grail]

whats the diffrence between sio and rest

other than speed

[honest remnant]

where do i donwload

[viral mason]

download what

[honest remnant]

the vc

[honest remnant]

uj

i donwloaded the vc

and now ium not getting anything coming out of the mic

and the volume shows as if its at 0?

how to fixc this

[viral mason]

which one did u download

[honest remnant]

this one https://huggingface.co/wok000/vcclient000/blob/main/MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a.zip

[viral mason]

btw do u have nvidia, intel, or amd

[honest remnant]

nividia

[viral mason]

alr cool u prob should download one of these instead

-rt

[vernal bone]

💻 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

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Deiteris Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[honest remnant]

i used the first one

from the guide

the official one from wok

[viral mason]

if u don't need a lot of models at once and prefer really high quality I'd switch to vonovox tbh

deiteris is good as a second choice but it's a littttle old

[honest remnant]

ia that the one im usingh

[viral mason]

if you're using the first one that's not vonovox

[honest remnant]

im using this one

[viral mason]

oh that's an old version of the voice changer

[honest remnant]

dym sending me a new one and maybe it will work

[viral mason]

u should download one of these ones as they're the most up to date, I recommend vonovox

the downloads are in the guides

[viral mason]

https://github.com/dr87/Vonovox/releases/tag/v1.6.7

[honest remnant]

do you hav ea tut?

do i put start

or setup..?

[viral mason]

u do setup then start

[brittle wing]

-colab

[vernal bone]

📒 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

[honest remnant]

@viral mason it finishewd

now what

[inner reef]

yeah how do I import these models?

[viral mason]

did u open it?

[honest remnant]

yeah but it works but i cant hear it thru discord

[viral mason]

what mic setting do u have set rn?

[honest remnant]

cable output from vb

but i also downladed the other vb

the

vac

[viral mason]

set ur setting like this to see if it works

[honest remnant]

now what?

[viral mason]

does it work on dc?

[honest remnant]

no

wait

these are my settings on the new vc

where the middle one

or is my output the vb

OMG

I FIXED IT

[sudden oak]

ai

help

[viral mason]

https://cdn.discordapp.com/attachments/122108999146405888/1317174256664055838/melon.gif

explain what u need help with, or uh nobody knows what u need

[knotty moth]

https://tenor.com/view/nobody-cares-gif-18196907282428815626

[sage lagoon]

can somone send me link for okada

nvidia

[viral mason]

-rt

[vernal bone]

💻 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

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Deiteris Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[viral mason]

It's in the guide just scroll down some

[deft condor]

where is this? im encountering the same problem @honest remnant

[sage lagoon]

what does this mean?

[empty charm]

how do you make your own models?

[spice iris]

thanks for the reply, it has a loud voice in mono and layers in stereo, anything i can do about it?

[dry sentinel]

Can someone explain if I start my model why you hear everything twice when someone speaks

[honest river]

Hey!!
By any chance you free today?
We are trying it but its not working and we could use your help 🫠🫠☺️

[coarse falcon]

Give me rvc Collab @vernal bone

-best colab

[honest river]

Hey guys, the voice is cracking and has lots of latency, any solution for it?

In w okada

[warm stone]

a

[hallow thistle]

!howtoask

[vernal bone]

❓ How to Ask for Help

✅ Before You Ask!

1. Check Docs & Guides: Your answer may already be in the AI Hub Docs or the https://discord.com/channels/1159260121998827560/1159513888199540817 channel.
2. Search the https://discord.com/channels/1159260121998827560/1192011222023950368 : Look for existing posts that solve your issue. Do not invade someone else's post.

📝 How to ask?

Tell your:

• Full GPU Name: (e.g., NVIDIA RTX 3060)
• Operating System: (e.g., Windows 11)
• Detailed Description: What were you trying to do and what went wrong?
• Tutorial Used: Link to the guide you were following.
• Screenshot: A picture of the full error message is very helpful.

🚫 Prohibited Topics (We Will NOT Help With These)

To maintain a lega, safe & ethical community, we will NOT provide help for:

• (E girl, as an example) catfishing/trolling, scamming, impersonation.
• NSFW/Porn.
• Any illegal activities.
  Requests for these topics will be ignored and may result in moderation action.

<:matsuripray:1159685390156967936> Community Expectations

• Be Polite & Patient: Our helpers are volunteers. You may ping the Helpers role once.
• English Only: Please keep all conversations in English.

[hallow thistle]

More like to "train a voice model". Are you looking for an option to train an RVC voice model?

[hallow thistle]

Did you follow any tutorial from YouTube before? What do you use W-Okada for? Trolling or catfishing? And what is yoru PC GPU?

[plain wave]

Guy's could you give me a guide of using AI in way that won't make me depend on AI for everything?

[honest river]

My friends who are facing the problem, will take over now

[steel sigil]

hello iam krisa's friend

so my problem is that the models that we are uploading is not working properly and latency is very much my gpu is amd rx 550 with i5 8 gb ram and a nvme ssd to run the applications ,windows 10 @hallow thistle

[dry knoll]

can somone pls give me good settings for vonvox voice changer

[hallow thistle]

What do y'all use W-Okada DirectML for? Trolling or catfishing? Just wanna make sure you won't use it in an inappropriate situation,

[hallow thistle]

What do you use Vonovox for? Trolling or catfishing? And what is your PC GPU?

[dry knoll]

for playing games with friends doing roleplays

i have a rtx 4070ti

[dry knoll]

do u have some good settings?

[hallow thistle]

I'm sorry about that, but you admitted to use the program for "catfishing", so no settings for you.

[steel sigil]

for playing games with friends doing roleplays

[dry knoll]

broo its to troll my friends

does that count

[hallow thistle]

That sounds like an excuse. "Trolling with your friends" is fine, as long you won't use "E-girl" voice model, but saying that after admitting to "catfishing" is straight up wrong.

[dry knoll]

i mean isnt trolling my friends catfishing or nah

[hallow thistle]

Also, editing your message won't help your situation either. Mods can see your logs anyways.

[dry knoll]

bruh im sorry

[hallow thistle]

Roleplay?

[spice talon]

hello

[steel sigil]

yes like anime characters roleplay u be the demon lord i be the queen of the humans type while building castles in minecraft

[hallow thistle]

Hi. How can we help you?

[spice talon]

hi i downmloaded this voice mod called vonovox

why does my voice sound so weird when i use the voice changer

[hallow thistle]

Did you follow this guide? This specific W-Okada version is better than the one from "YouTube tutorials". https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#download-amd-intel-and-cpu-on-windows

[hallow thistle]

What do you use Vonovox for? Trolling or catfishing? And what is your PC GPU?

[spice talon]

um i use it for youtube videos

and my gpu is rtx 40 something i forgot

[hallow thistle]

To check your GPU name, open Task Manager.

[spice talon]

oh ok

it says 4080

[lethal iron]

Which tutorial do you guys use for AMD on comfyui?

[hallow thistle]

Vonovox is quite less documented than "fork W-Okada" so let try these settings in Vonovox:
Pitch smoothing factor: 10.0
Block size: around the minimum possible value; if the audio is unstable, try increase the block size up until it stable
Extra: 2.0
Crossfade duration: 0.08-0.1 for fast or 0.15 for high quality
https://docs.aihub.gg/realtime-voice-changer/local/vonovox/#settings

[viral mason]

I've never seen this before so idk

[hallow thistle]

What do you use W-Okada for? Trolling or catfishing? And what is your NVIDIA GPU?

[steel sigil]

@hallow thistle thanks for your help i will try it right now i will get back to you after i try it

[hallow thistle]

Let me know if you look for settings.

[steel sigil]

yes please that would be very helpful

[spice talon]

whgat about pitch, index rate, and format

[hallow thistle]

AMD Radeon RX 550 is not quite the fast for AI, so let try these settings:
Chunk: "perf number" from the top left screen + 60 ms; if you wanna use it with a game or when the audio is unstable, try increase the chunk number until it stable and perf number green
Extra: I'm not sure about this, but it can be 2.0 s
https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#finding-my-own-settings-for-chunk

[hallow thistle]

You can set pitch up to +12 if you wanna use a female voice model and your actual voice is low; the other two not quite needed.

[steel sigil]

thanks for the help i will try these settings and will get back to you thanks for your guidance 🙂

[spice talon]

why does it keep cutting out near the end?

[viral mason]

could be bad settings

[spice talon]

ok i fixed it thanks

[proud moat]

i need help when i try and switch over to the GPU on the GPU or CPU tab it

[Voice Changer] Loading index...
Try loading... model_dir\4\added_IVF3005_Flat_nprobe_1_v2.index
Pipeline has been deleted

im on rtx 5070

it doesnt work on my GPU only my cpu which isnt as fast as my GPU

[hallow thistle]

What do you use W-Okada for? Trolling or catfishing? And did you follow any tutorial from YouTube before?

[viral mason]

don't forget the only good reason to use it, playing as characters like Goku or Darth Vader ect

[proud moat]

Im just messing around with my friend and probs go troll as a egirl on roblox later

to get reactions

😭

bruh

[oak edge]

GANGGGGGGGG

[viral mason]

heloo

[oak edge]

i got money makin ideas

[hallow thistle]

Wow, so you admitted to use the program with an "E-girl" voice model huh. Sorry about that, but I won't be giving you the better W-Okada on this one.

[proud moat]

🙏

[viral mason]

lemme hear

[hallow thistle]

Regardless of your intention, using W-Okada with an E-girl voice model still counts as "catfishing", adding the statement with "Roblox" makes thing even worse.

[honest river]

👀

[proud moat]

why is there a egirl model in the server if it counts as catfishing

[oak edge]

https://tenor.com/view/hahahaha-gif-27342590

[viral mason]

maybe he likes them young like the ceo of roblox

[proud moat]

bruh what

[hallow thistle]

[proud moat]

im 17

[viral mason]

they should hire him

[hallow thistle]

Even worse.

[proud moat]

i js wanna have fun with the voice changer but I was asking for help to get it to work on my GPU

[lethal iron]

rx 6750xt
windows 10
I was trying to install comfyui-zluda on my pc and i was trying to open it and this error came
https://www.youtube.com/watch?v=X4V3ppyb3zs and https://www.youtube.com/watch?v=gfcOt1-3zYk were the tutorials used

[misty star]

Hey I dont have the best pc But i downloaded a voice model from weights. how can I use it to have real time voice changer? any reccomendations? Ive been searching server and I am lost lol

[low shard]

Hello Teto
Please Elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
  So I can suggest you a realtime voice changing program

[misty star]

Nividia Geforce GTX 1660 super

windows 10

Intel core i7

[low shard]

are you looking to use it to roleplay in discord vc or games?

[misty star]

16G of ram

[low shard]

bc games are more intensive ofc

[misty star]

roblox and discord

yeah

[low shard]

it's fine then

[low shard]

btw windows 10 support ending in october 2025

-realtime

[vernal bone]

💻 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

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Deiteris Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[low shard]

You can use either Vonovox or Wokada Deiteris Fork

[misty star]

Oh lol, im not sure if I upgraded or not

[low shard]

vonovox might be suggested more for nvidia cards

but if you want u can check both their pros&cons

[misty star]

Okay thanks ill read up on them

the file with the voicechanger i downloaded from a friend has a .index and a .pth will that work on there?

[low shard]

yup, that's a PyTorch model, not getting technical but it's an RVC (Speech To Speech) model specificaly in this context:

• pth files: contain the voice
• added index files: contain the accent
• metadata.json file: it's just some extra info about the model download link if you downloaded it off weights.com, it's not needed and won't impact the actual model at all

most models in #1175430844685484042 are RVC, which means they will work with both vonovox and wokada

[viral mason]

omg hi Teto

I made a model of you

[low shard]

https://tenor.com/view/teto-teto-pear-pear-cursed-sad-gif-2951834529774838348

[viral mason]

teto model real
https://discord.com/channels/1159260121998827560/1411557807077199914

[misty star]

tyyy

wait lol @viral mason This is the one my friend sent me. tysm for making this!

[viral mason]

np!

[misty star]

When I run MMVCSERVERSIO it just opens command prompt and does nothing. I tried doing what it said in the doc and doing the webpage thing but that didnt load either

[viral mason]

you downloaded the correct version right?

try a couple times

sometimes it bugs out

[misty star]

I downloaded the first one for navidia

Download NVIDIA on Windows
The latest version as of December 7th 2024 is: nvidia-b2332 (click here to download)
If you have a GTX 700 card or below, use AMD/Intel version instead.

[viral mason]

I believe that's the one I use

you could also try Vonovox

[tough plover]

hello, i've been trying to use W-OKADA for a few times, but i've had this problem that external models don't give any sound. like the beatrice model that is a default model works, but when i put it to other models, suddenly there's nothing coming out. any suggestions?

[misty star]

Okay, ill redownload and if that dosnt work ill prob try vonovox

[neon chasm]

i have CalledProcessError in colab how i can fix that

[viral mason]

what collab is it??

[neon chasm]

https://colab.research.google.com/drive/1DySwjZTHkCz3sKWQQ68y0eTMXbLh2t9i?usp=sharing#scrollTo=FBpJW3YB5zu1 this collab

[viral mason]

that's RVC disconnected

it's been discontinued and hasn't been fixed in weeks

probably months

you should switch to applio, and I'd say it's better to switch to Kaggle instead of Google collab

it gives 30 hours for free and u can keep deleting your account and remaking it after u use those hours up

[neon chasm]

okaay so thanks i am trying now

[viral mason]

if u can't train locally here's the links u need for it
https://www.kaggle.com
https://dashboard.ngrok.com

[misty star]

Every time I run it this is what I get

Wait nvm it's doing stuff now

Tf

[viral mason]

told u it's weird lol

[misty star]

Now what do I do lol

Oh shi it's doing more

Okay I'm good now

It opened this for me

[neon chasm]

@viral mason I've learnt how to use RVC, but I'm not quite sure how to do this 😄

[misty star]

How do I make it so I hear myself to test this, on bottom I put monitor as headphones I thought that would work. It says I'm talking when I click start I just don't hear myself

Wait nvm

My cpu is at 100% using this lol

[viral mason]

idk what would cause that, I don't use the index feature although idk if you're using it

[misty star]

Yeah when putting in the voice model I put in the index and the other file, should I not have done that?

[viral mason]

I meant the index slider

I keep it at 0 normally

[knotty moth]

you are using cpu, so your gpu usage is literally 0%

[viral mason]

ohh yea I just noticed that

change it to your nvidia gpu

[misty star]

How do I do that.

Yeah slider is at 0

[knotty moth]

then go read the guide carefully

[misty star]

Okay I'll go back and re read

[viral mason]

click where it says cpu and change it to your gpu

make sure the app isn't started

like uh

start isn't on

I'm bad at words

change rmvpe onnx to regular rmvpe, and use these settings as sort of a starter kinda thing

[misty star]

Oh I'm dumb I see that ty

Oh it was on in rip

It's still high in memory but it's much better now lol

[viral mason]

is the delay bad or nah

btw if your voice is deep like if you're a guy u should up the pitch, I have mine set at 10 for Teto's voice specifically

[misty star]

Nah it's not horrible, won't be playing any fps lol

10? I'll try that ig it sounds decent at 1

[viral mason]

it depends on your voice tbh

[honest river]

I am having the same problem. 🫠

[steel sigil]

hello can anyone help me out? in my voice changer the default models are running smoothly but when i upload any other model they are lagging with much high latency.. why is it happening? and if anyone know any good working tested male and female model (english) plz let me know.. Thankyou

i have to restart the external model after every use for it to even work

[elfin sable]

Yo, for the Okada one, why does it never stop talking? Like, when I say something, the voice changer repeats it 3–4 times and makes silly noises

is there any fix for this?

[rotund vigil]

I created a voice model by recording my voice in the place where it sounds best with Trenchboard. My data was 38 minutes. However, when I record my voice live with Wokado, my voice becomes very high-pitched or out of tune. What's the problem?

[viral mason]

what is a trenchboard?

btw u most likely didn't clean the audio you used

use this
https://colab.research.google.com/github/Eddycrack864/UVR5-NO-UI/blob/main/UVR5_NO_UI.ipynb?authuser=4#scrollTo=gmjUWmz8iecd

[rotund vigil]

trenshboard

[elfin sable]

help

any solution

[rotund vigil]

What is this? I had already recorded my voice cleanly. There was no problem with the sound.

[viral mason]

you can use that to clean the audio you used, like remove mic noise or reverb

[rotund vigil]

Could it be due to another reason? I think the sound is clear.

[viral mason]

if there's any background noise it could be why

do you know what settings you used to train it?

[rotund vigil]

ye I used Applio. I ran it on my computer. I saved the model in the most convenient location using the board.

I added the entire 38 minutes to assets. It automatically cut the files.

[viral mason]

I meant these kind of settings

[rotund vigil]

Yes, I did the most appropriate ones.

[viral mason]

that doesn't really help

[rotund vigil]

Is that why I'm doing Turkish voice-overs? Maybe because you're doing English voice-overs.

[rough canyon]

y is my voice so delayed

it takes like 30s just for it to come

[rotund vigil]

Most of the settings in the screenshot you posted are the same as mine. I just have normalized mode off and silence set to 0.

[viral mason]

you should have normalization on post, it makes the model come out better

idk about silent training files tho

[rotund vigil]

Could it be because I'm making a Turkish voice model? Maybe Conventec is suitable for English versions. I've heard it used for split pronunciation, but I'm not sure.

[viral mason]

I'm not entirely sure

[edgy oak]

can someone help me set up my ai voice changer

i recently got a very good stable internet and i wanna use the web version of it

[viral mason]

do u have nvidia?

[edgy oak]

no amd

[viral mason]

use the first one in the list, it has a guide

-rt

[vernal bone]

💻 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

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Deiteris Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[edgy oak]

this one?

but my pc is not that great and i wanna use the browser one

[viral mason]

yea, that's the browser one

download should be there once you scroll down

[edgy oak]

this?

[viral mason]

yup!

[edgy oak]

okay now it works

but it feels so bad

what are the best settings?

[rapid smelt]

how do i get that ui

i have a different one

[viral mason]

It's the settings

[viral mason]

Try these

[agile lynx]

use gpu

not cpu

[deft rain]

Yo

This the right place for help with something?

[agile lynx]

yep

[deft rain]

I got a weird question to ask

If u mind

[agile lynx]

tell me

[deft rain]

I have isolated vocals from a song using huggingface models, so with the isolated vocals is it possible to remove the background vocals (ad-libs) and seperate them from the lead vocal?

[silk dragon]

I know no one cares but help me 😭https://discord.com/channels/1159260121998827560/1412147824090681488

[agile lynx]

i think you can but idk how

[deft rain]

The song has alot of ad-libs

[deft rain]

Damn,

[empty knoll]

Guide to using the voice changer for the AMD 9000 series?

[agile lynx]

i think you should use uvr mdx net karaoke

uvr 5 is a program

mdx net karaoke its a model

to remove

[deft rain]

Thanks, Imma look into it now

[latent cypress]

yes, use melroformer karaoke by becruily, mdx karaoke is outdated

https://huggingface.co/becruily/mel-band-roformer-karaoke/tree/main

[toxic ermine]

how can i make ai cover? what's the updated web rvc

also is ilaria rvc still active?

[agile lynx]

oh, didnt knew

[latent cypress]

right now we have 3 reliable karaoke models for extracting lead vocals

karaoke becruily, gabox karaoke v1, and fuzed (a mix of karaoke becruily and gabox), which is only available on the mvsep website

[next knoll]

should i use w okada or deiteris if i have an amd gpu?

[empty knoll]

Is it better to use it with a 9800x3d CPU or a 9070 xt?

[agile lynx]

gpu.

[agile lynx]

deiteris.

[empty knoll]

Is there a special version for AMD's 9000 series?

[agile lynx]

only this

[empty knoll]

Ugh, those don't work well for AMD. Let's hope that someday it will be compatible with Rocm

[agile lynx]

use cloud

its the only option

[misty star]

Is there a way to fix my audio randomly sounding like I'm whispering? My voice good but randomly it will bug out

I think it might be lag related but idk

It will randomly make it like bug out mid sentence

[coarse breach]

Does anyone know if it's possible to convert an RVC model to onnx/json that can be used with Piper?

[next knoll]

how do i delete a model

[viral mason]

if you're using Deiteris you really can't in any simple way, but you can just put another model into the spot of the model you want to get rid of

[next knoll]

i have index and idk how to get rid of the index

[viral mason]

wdym

[next knoll]

i put in an index file and i dont know how to get it out

[viral mason]

I don't get it

[next knoll]

this index file is still there even though i replaced the model file

[viral mason]

you can't place an index file in the part where you put the pth file

it will give you an error

what voice changer are you using, Deiteris or Vonovox

[next knoll]

Deiteris

i didnt put the index file into the pth file

i put another pth file into the pth file i want to replace but the index file is still there

[viral mason]

just replace the index too

[next knoll]

with what

the model i am trying to replace it with doesnt have one

[viral mason]

oh

if a model doesn't come with an index that means it's most likely really old

and you shouldn't use it

tho keeping that index wouldn't hurt it at all

[oblique bay]

bit of a odd question, but it is possible to use rtc on kaggle while on Oculus quest 3s?

[viral mason]

I'm not entirely sure

Could you explain more of what you're trying to do?

I use the voice changer while running around in VR so I kinda have an idea on local stuff in VR but I do know kaggle runs fine

[cobalt abyss]

Couldn't find a clear answer on this: If you upload a model to here, does Weights copy that upload? And if it does, how does it know your Discord account and your Weights account are one and the same?

Why I ask is because when I download a model from Weights it comes with a metadata.json file which contains the same information I'll see given in the description of a model here, so I was wondering if it's even necessary to upload manually to Weights.

[viral mason]

it only uploads if it's in the voice model section

[deft rain]

How do u use this thing? There is no option to upload and stuff, I jus saw a 72GB file.

I sound like a noobie🤦

[cobalt abyss]

I see. So does it create a profile for you on Weights automatically? I made a profile on Weights today but I'll delete it if my guess is correct.

[ancient sierra]

Does anyone know why i cannot use my mic while while having the voice on? it just is not working at all i cannot talk on discord it says error and i cannot do anything about it.

[viral mason]

I'm not entirely sure

I think so?

I think I signed in with my discord

[brittle bobcat]

I've been using deiteris wokada with kraggle for weeks now and all of a sudden its giving me an error when i try to make the ngrok server. Does it occasionally go down sometimes or what

[cobalt abyss]

Alright, sounds like that's probably it. Thanks!

[viral mason]

If you need any more help ask here or dm me, I may or may not know ^^

[deft condor]

my voice changer doesnt work on discord, can someone help me detect the problem? i put the correct input device

[snow hazel]

Hello is theyre any webites that help with robot voices

Or is good at extracting them

[knotty moth]

havent tried to see if it works, but if you find it doesnt work yet, stay tuned

[knotty moth]

the best model so far is melroformer karaoke aufr33 & becruily version

still it may be hit or miss esp for adlibs & harmonies

[hearty mango]

hello, please help me with this problem when I start talking in voice changer it pops up in the console "waiting generate pipeline" "pipeline is not initialized"I realized that the program does not support voice models in pth, downloaded others, the virtual cable is all there, it still does not work on any version

[latent cypress]

i don’t get what you mean by 72gb, but you can use the uvr gui software (local) or uvr huggingface space by not eddy (cloud)

[south verge]

its there anyway to suppress the like glitchy ness?

[elfin sable]

yo for the Okada one, why does it never stop talking? Like, when I say something, the voice changer repeats it 3–4 times and makes silly noises

[deft rain]

I mean I saw three files, there was two that was couple Kbs and one for 72GB, btw can u guide me into doing it please.

[weary minnow]

when i play roblox the voice changer cuts out but it works fine on discord help pls

[viral mason]

watcha using that on roblox for hm?

[knotty moth]

please elaborate:

• your purpose to use the voice changer
• what kind of voice model you're going to use
• GPU & PC spec and the operating system
• tutorial link you're using and referring to

[opaque dock]

hey I wanted to ask, if some of the options on F0 Det. say N/A does that mean the issue is with the model or voice changer

[pure wharf]

yo i did everthing and made mt input as virtual cable but nth works

[low shard]

Hello
Please Elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what tutorial link are you using
• a screenshot of the program

[low shard]

Welcome here, it's hard for us to elaborate what you mean by "everything" since this is a general ai server,
Please Elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • E girl trolling
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[pure wharf]

i have a rtx 3050 and windows 11 and i wanna turn my voice into a egirl to troll on games

[low shard]

E girl trolling is catfishing, which is illegal. You have been warned both verbally and via sapphire and continuing to ask will lead to further actions possibly like a ban
Also if you followed a youtube video tutorial, they are all old, delete everything and forget about catfishing

[opaque dock]

Processor: AMD Ryzen 5 5650G with Radeon Graphics
Cores: 6 cores
Threads: 12 threads
Clock Speed: 3.9 GHz
Graphics: Integrated Radeon Graphics (part of the Ryzen 5 5650G)
Motherboard: ASUS PRIME A520M-K
RAM: 16 GB

tutorial link I used :
https://youtu.be/SxdnGxicJOg?si=qEC6NCTQMiqpiI6x

sadly I cannot provide a screenshot since I dont have image perms

[low shard]

That tutorial is outdated

delete everything

are you trying to do e girl trolling like in the video?

[opaque dock]

yep, mostly in valorant only

also mind if I ask what do you mean by delete everything? like- delete the voice changer?

[low shard]

E girl trolling is catfishing, which is illegal. You have been warned both verbally and via sapphire and continuing to ask will lead to further actions possibly like a ban

[opaque dock]

oh, my bad then, l'll stop

[low shard]

all the programs used in the video are extremely outdated (vb audio cable can cause issues on windows, that original wokada is outdated by 2 years), you can just delete them

[pure wharf]

Not actually egirl trolling it’s a skit

[oak edge]

damn

[low shard]

We don't allow any e girl trolling, it's written right in the server rules and help guidelines, in fact all of those youtubers are all banned, and you have admitted to wanting to do it

[opaque dock]

well, I do wanna use it to voice my OC, if that's allowed

[oak edge]

what's that

[pure wharf]

Ok got it what if it’s for another purpose that’s legal

[opaque dock]

OC stands for Original Character

[low shard]

You have used a catfishing tutorial and explicitly admitted to wanting to e girl troll 5 seconds ago, we don't want to help people for such things

[opaque dock]

understood

welp, I give up

[low shard]

It wont matter if you try to switch the topic 5 seconds later, you have admitted to wanting to do e girl trolling, trying to change it now even asking me for a legal reason you don't even know, trying to bypass moderation won't work

[low shard]

those damned youtube tuts 😭

[oak edge]

i kinda feel bad fr you man, the amount of peeps you are sayin it again and again

[knotty moth]

regardless if you draw your own OC, but do you have consent with the voice owner? otherwise it'd be nothing but egirl trolling which is not allowed

[low shard]

Nah dw about feeling bad lol, it's just i don't understand why it's so much normalized only because of youtube vids tbh

[opaque dock]

No? but I'd ask for consent, if not provided then l'll just move on and call it a day

[oak edge]

i think most are kids

or just immature peeps

Y_Y heck when I was a kid I used to watch these trolling people in fortnite as Donald Trump, Penny wise

kinda vids

[low shard]

it's not like dudes trying to act girls for money

[oak edge]

[low shard]

there were people who deadass joined with statuses like "looking for older men to spoil me"

[viral mason]

https://tenor.com/view/teto-teto-pear-pear-cursed-sad-gif-2951834529774838348

[low shard]

or blackmailing shi

ofc they all got moderated

[oak edge]

holy

[opaque dock]

[oak edge]

good riddance

[viral mason]

I hope all the people joining just to get the voice changer for egrill shit fail in life (original removed to not upset snowflakes in the future)
-# anyone who got mad at this is, it's a joke, do not message me to be rude or harass me and to the guy who told me to kill myself, FUCK YOU

[opaque dock]

Nick's words are final so uh, yeah-

[oak edge]

[knotty moth]

nvm what nick said about catfishing intention is right, and you cant make excuses anymore

[viral mason]

https://cdn.discordapp.com/attachments/1374941364709490711/1412079142790369350/image0.gif

[opaque dock]

okay?

fair enough I guess^^

ty

[knotty moth]

same on you, no more excuse over your egirl trolling intention

[viral mason]

yes please go away and figure it out on your own yucky egrill troller ledo

[harsh grove]

Iss there anyway android version of voice changer that is real time for games? If possible that can convert it into engrish to japanese right away

🫤🫤🫤🫤

[simple ore]

it is like asking 'can I move 50 tons of boulders using a push cart'

[viral mason]

that's a good idea

[harsh grove]

xD

[tidal flame]

Hello I'd like some help with setting up the voice changer I have tried following some guides but didn't seem to work

[maiden sapphire]

Guys, why do I hear myself?

[viral mason]

That's called schizophrenia I think

[weary minnow]

1. to sound like a brwawl stars character
2. the mortis one on the weights website
3. CPU - Ryzen 5 5500
   GPU - XFX SWFT 210 RX 6600
4. none

[snow hazel]

Is there a good website that’s good with extracting robotic voices?

[formal jackal]

looks like the real time voice changer doesnt support internet congestion provider BBR2 yet.

[hoary zinc]

hello! today when I went to start up wokada on Kaggle, I got the error below

"/kaggle/working/MMVCServerSIO
Traceback (most recent call last):
File "client.py", line 10, in <module>
File "PyInstaller/loader/pyimod02_importers.py", line 384, in exec_module
File "main.py", line 21, in <module>
File "PyInstaller/loader/pyimod02_importers.py", line 384, in exec_module
File "downloader/SampleDownloader.py", line 11, in <module>
File "PyInstaller/loader/pyimod02_importers.py", line 384, in exec_module
File "voice_changer/RVC/RVCModelSlotGenerator.py", line 4, in <module>
File "PyInstaller/loader/pyimod02_importers.py", line 384, in exec_module
File "torch/init.py", line 367, in <module>
ImportError: /kaggle/working/MMVCServerSIO/_internal/torch/lib/libtorch_cpu.so: cannot read file data

Press Enter to continue..."

it works on another kaggle account just fine, but for some reason it's suddenly not working

i've been trying to get it working for the past hour or two, and i'm not quite sure what needs to be done

[snow hazel]

Is there a good website that’s good with extracting robotic voices? Can someone help me with this?

[chilly cedar]

Is it still possible to make AI chromatics with the voice models here?

[torpid jacinth]

#dork

[viral mason]

https://cdn.discordapp.com/attachments/1045168452278702162/1381048873564508200/togif-23.gif

[snow hazel]

Is there a good website that’s good with extracting robotic voices? I need help with this

[faint phoenix]

Hii

Does someone knows why the exported audio doesn´t create? (on applio)

the output info says that the audio was made successfully but when i try to hear the preview or download nothing happens

[clear mortar]

hey can anyone help me when I try to inport my files I cant find my AI voices I put in the model folder

or when I try to upload a zip does not work

[true swallow]

How do i fix the voice sounding so robotic / lagy?

[charred drum]

reloading the page didn't work what do i do

[delicate wedge]

When in vrchat on steamvr, the ping on my rvc goes up quickly.

How do I fix this?

[simple ore]

why dont you use the Application's UI to do that

[agile lynx]

попробуй нажать на перезагрузку сообщения, а не перезагрзку сайта

[charred drum]

failed

[jovial dock]

does chunk make it sound better

in mmvc

[jovial dock]

better graphics card or just do a max fps in the game

maybe lower the chunk

lor incrase it

[next geyser]

discord cant hear line 1

discord doesnt recognize the voice changer and doesnt work

the voice changer isnt feeding audio into line 1

[low shard]

what's your Operating system?

Also I see you're using some e girl models, are you trying to do e girl trolling?

[low shard]

it's not how that works, and it depends on what program you're talking about

Please Elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • E girl trolling
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[verbal widget]

Which collab can I make a model from?

[low shard]

You sure you wan to start with cloud? if you got a good pc you could use it locally

[balmy dew]

não estou achando a voz da hatsuni miku

@vernal bone não estou achando a voz da hatsuni miku

[low shard]

Hello, this server is English only, it looks like you can't find a voice model
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)
• https://voice-models.com/ (just scrapes the huggingface link from the voice-models channel)
• 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 in the server

[earnest musk]

:wave: @low shard, How can I help?

Available Commands:
• @weights find <query> or /find <query> - Search for RVC Voice Models

[austere dust]

on the admin panel it says wait web server and doesnt fully load the ai voice changer what should i do

[low shard]

There are many AI Voice Changers
Please Elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • E girl trolling
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[long pier]

How do I set up the voice changer on a Dell pc

[lament wagon]

Any new Huggingface version of Ilaria RVC because I don't think the version I'm using works for me

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

I'm looking for a real human to answer for me

[agile lynx]

not fixed yet

#📰│dev-updates

[misty star]

Hey I can't hear the voice changer and I'm not sure why. Can I get some help?

(this is for Mav # her)

@

[unborn elm]

what are the minimum requirements to successfuly train an rvc model

[low shard]

#📰│dev-updates

[low shard]

There are many Voice Changers
Please Elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • E girl trolling
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[low shard]

are u using it in games?

[low shard]

an nvidia card with atleast 8gb vram would be the minimum for a decent model

you could train on shittier gpus, but with lower batch sizes and not really suggested

[unborn elm]

so how much should i do with a nvidia 4060 and 8gb vram

whats the maximum

[potent bone]

which one should i download https://huggingface.co/IAHispano/Applio/tree/main/Compiled/Windows

[misty star]

Yes but it's not making sound

Please @long pier this is for her

@low shard

[agile lynx]

3.4.0

its the latest

[hoary zinc]

hello! today when I went to start up wokada on Kaggle, I got the error below

"/kaggle/working/MMVCServerSIO
Traceback (most recent call last):
File "client.py", line 10, in <module>
File "PyInstaller/loader/pyimod02_importers.py", line 384, in exec_module
File "main.py", line 21, in <module>
File "PyInstaller/loader/pyimod02_importers.py", line 384, in exec_module
File "downloader/SampleDownloader.py", line 11, in <module>
File "PyInstaller/loader/pyimod02_importers.py", line 384, in exec_module
File "voice_changer/RVC/RVCModelSlotGenerator.py", line 4, in <module>
File "PyInstaller/loader/pyimod02_importers.py", line 384, in exec_module
File "torch/init.py", line 367, in <module>
ImportError: /kaggle/working/MMVCServerSIO/_internal/torch/lib/libtorch_cpu.so: cannot read file data

Press Enter to continue..."

it works on another kaggle account just fine, but for some reason it's suddenly not working on this one

i've been trying to get it working for the past day, and i'm not quite sure what needs to be done

[cobalt abyss]

Had an issue in Applio I couldn't find an answer on: When I click "Generate Index" everything works perfectly fine except for the fact that there is no separate "added" and "trained" index file. There's only one file that's created in the logs folder which is simply an index file with the name of the model and nothing else.

There were no errors in the terminal and no errors in the web gui so everything worked properly. What am I missing?

[simple ore]

there's only one index

[unborn elm]

Anyone know how to train faster? Going at about 5 epoch every 15 minutes

Using applio

[simple ore]

which version?

[unborn elm]

3.31

I believe

[simple ore]

how big is the data set? what gpu? what batch size?

[simple ore]

switch to 3.4.0

[unborn elm]

1h 55m data set nvidia 4060 4 batch

[unborn elm]

What are the improvements

[simple ore]

training actually works 🙂

[unborn elm]

so training doesnt work on 3.31

[simple ore]

2h is too much for finetune, batch 4 is too small for 2h

[cobalt abyss]

Oh, I was following the written guide in the AI Hub docs and it explicitly said to pick the "added" version along with an image of said file. Plus I've seen you guys tell people they put the wrong index file before when it hasn't been the "added" version.

Things have changed since then?

[unborn elm]

what settings whould i do

[simple ore]

training was broken for a quite long time as we found a few days ago

[simple ore]

pick best 30 minutes from the set, use batch 4 to 8

[unborn elm]

are u sure 30 minutes is enough

i want a high quality model

[simple ore]

yes, it is perfectly fine now

[simple ore]

'trained' index was empty, it was created for some stupid reason in original rvc and never populated, actual index was called 'added' (with data added), in applio we removed this nonsense, there's just one populated index file

[unborn elm]

and what is the batch difference

which is better for best sounding

[simple ore]

it is up to experimenting

[unborn elm]

i want this to sound realistic compared to other models of this person

[simple ore]

depends on the dataset

[cobalt abyss]

Cool, thank you.

[simple ore]

big variety in the set requires higher batch for smoothing

[unborn elm]

how many epochs should i do for what im looking for

like barely noticable ai

you would have to listen closely

[simple ore]

watch tensorboard, test epochs when the loss converges

as long as you save all weights every x epochs you can pick the best model from the bunch

[unborn elm]

@simple ore what are plugins for

[simple ore]

there uvr and elevenlabs, but you're better off using standalone

[unborn elm]

ok

[tidal parrot]

Guys, Weights.com is doing so great to me with AI covers. However I need help. Is it possible to transform a choir song in a solo? <@&1159293204038955078>

[unborn elm]

@simple ore what do i do in tensorboard

[simple ore]

go to scalars tab

these are the charts of interest

[unborn elm]

what does this stuff mean

[unborn elm]

yea im there

is this just to see the quality of it

[simple ore]

this is to see the training progress

[unborn elm]

ok

[simple ore]

orange chart here converging around 31.5

[unborn elm]

when its done i press stop training and then generate index right

[simple ore]

you can generate index right after you started training

[unborn elm]

ok

[unborn elm]

when its done what do i do

or will everything i need automatically go in the logs folde

[simple ore]

'save every weights' saves the voice model every x epoch

[unborn elm]

ok

tysm

[vast fog]

Hey, does anyone have a fork/downloaded repo of the following - https://github.com/microsoft/VibeVoice ? I am currently unable to access is as it has probably been made private.

[simple ore]

it was not good

[unborn elm]

@simple ore what does the important stuff in scalars mean

im trying to check in on it before i go to sleep

[simple ore]

grad_avg_50 charts should not be all over place

fm chart should not go up too much

kl/mel/total g should be slowly going down

[unborn elm]

is it good for grad avg 50 to staircase down and stabilize while still going everywhere

[simple ore]

grad (gradients) are updates the model makes to its components, they should not vary too much.. no going from 200 to 1000 to 5000

[unborn elm]

my fm is at 6.5

[simple ore]

the chart direction matter, the value not so much

[unborn elm]

its going up

[simple ore]

show the screenshot

[unborn elm]

wait which fm

[simple ore]

if you're using 3.4.0 version it should not go up too much

3rd

[simple ore]

you are not looking at scalars

you're looking at Time Series

[unborn elm]

i selefted scalars

[simple ore]

click Scalars

[unborn elm]

it is a darker shade

does that mean its pressed

[simple ore]

underline means it is selected

[unborn elm]

oh

well its still rising

[simple ore]

expand loss_avg_50

[unborn elm]

i did

[simple ore]

and show screenshot of all

[simple ore]

well, it does not rise too much

click Audio

then

there are going to be samples from each saved epoch

[unborn elm]

ok

[simple ore]

you can keep running the training for a bit more

[unborn elm]

at what epoch will it start to sound like the person i chose

[simple ore]

but you can also go to inference tab, click refresh and try inference using some of the safed epochs after 10k steps

usually you can hear resemblance just after 5-10 epochs

[unborn elm]

it sounds very robotic

[simple ore]

after that there are small improvements

[unborn elm]

150 epoch isnt much different from 50

[simple ore]

you're using version 3.4.0, right?

[unborn elm]

yes

im coming up on 200 epoch

[vast fog]

Hey, so the Vibe Voice model was not good or the official code repository did not contain enough details to be considered good. Sorry, I am a bit confused.

[simple ore]

the outputs for default speakers were really weird

but you can find a few works if you just search github for VibeVoice

[vast fog]

Okay thanks, but I saw a comfyui work that was using the model to generate some great results, however I wanted to use it for an academic work, just wanted to look at the design and everything, are you aware of the intricate details, like which language model they used, or the design of the CFM, which vocoder, etc.?

[hallow thistle]

!howtoask

[vernal bone]

❓ How to Ask for Help

✅ Before You Ask!

1. Check Docs & Guides: Your answer may already be in the AI Hub Docs or the https://discord.com/channels/1159260121998827560/1159513888199540817 channel.
2. Search the https://discord.com/channels/1159260121998827560/1192011222023950368 : Look for existing posts that solve your issue. Do not invade someone else's post.

📝 How to ask?

Tell your:

• Full GPU Name: (e.g., NVIDIA RTX 3060)
• Operating System: (e.g., Windows 11)
• Detailed Description: What were you trying to do and what went wrong?
• Tutorial Used: Link to the guide you were following.
• Screenshot: A picture of the full error message is very helpful.

🚫 Prohibited Topics (We Will NOT Help With These)

To maintain a lega, safe & ethical community, we will NOT provide help for:

• (E girl, as an example) catfishing/trolling, scamming, impersonation.
• NSFW/Porn.
• Any illegal activities.
  Requests for these topics will be ignored and may result in moderation action.

<:matsuripray:1159685390156967936> Community Expectations

• Be Polite & Patient: Our helpers are volunteers. You may ping the Helpers role once.
• English Only: Please keep all conversations in English.

[hallow thistle]

Read "help guildelines" before you start asking anything here.

Otherwise, I can't help with that.

You really got an answer huh? Now answer my questions for real:
What program are you looking for? RVC or W-Okada realtime voice changer? What do you use them for? Trolling or catfishing? And what is your PC GPU?

Intel Core i5 14400F is a CPU, not GPU. The GPU is something like Intel Arc, NVIDIA GeForce RTX 4090.

To check your GPU, open Task Manager, go to Performance tab and check if there GPU 0 or GPU 1.

[simple ore]

you can find a fork here https://github.com/toddllm/vibevoice

to me the results were terrible, some random music, noise, sound effects... with 7b model it started to sing instead

[hallow thistle]

The program name for "realtime" one is W-Okada. https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#download-nvidia-on-windows

Let me know if you're looking for "best" settings in W-Okada.

[vast fog]

Thanks a lot, really appreciate it, I am mostly interested in their network architecture, will test their model out at a later time or if they make further updates, thanks once again! 🙂

[pallid herald]

Oh dear God weights AI has restrictions now 😭🙏

[hallow thistle]

You concerned? I'm not sure why one would make 20 AI covers a day there, but I already got Weights Premium.

[pallid herald]

But that was your money bro 💀

[pallid herald]

I ain't buying premium

[hallow thistle]

The thing is, I got it for free but with a reason.

[fading skiff]

is this for VCClient to?
bc i got that issue where the voice changer isnt working (i can hear my voice normally)
its prolly an easy fix but idk how

[low shard]

you should be fine

[low shard]

3.4.0 is the latest as of now, but are you sure you got an nvidia gpu?

[potent bone]

yes

3060 ti

[low shard]

what game? play in lowest graphics 720/1080 30/60 fps cap, your gpu is pretty old
f0: rmvpe without onnx
output: line 1
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)
• Advanced Settings -> Crossfade Lenght: Controls how smoothly the AI stitches different processed parts "chunks" of your voice back together. 0.1 for fastest voice, 0.15 for improved quality but increases delay by ~50 ms
• Reduce the delay on Windows via the Wasapi / Asio Guide

[low shard]

you'll be fine for inferencing pre-recorded audios and training models then :)

[low shard]

Hello! This is a general AI server, we help for any type of AI, included AI Realtime Voice Changers, but we need more info to know how to help you

Please Elaborate:

• your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
• your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
• what you want to do? There isn't a program that does everything, there's a program for each thing:
  • AI Covers
  • Train RVC Models
  • TTS
  • Roleplay in VC
  • Roleplay in Games
  • etc
• what tutorial link are you using
• a screenshot of the program

[low shard]

Like every other Cloud site, because they pay for the server and the expensive GPUs you use remotely tbf

[pallid herald]

I literally can't make more then 3 people sing one song cuz I have to buy "PREMIUM >_>" Like c'mon bro I want the old weights back I hate the new weights make it the best ai cover 😭🙏

[low shard]

Everyone loves free things, me too, but someone has to pay

If you got your own good pc, you can use Replay or Applio to use RVC freely with your own hardware

[pallid herald]

But I'm broke 💔

[pallid herald]

I'm broke ass fuck 💔🙏

[pallid herald]

I tried looking for some more ai cover websites but none of them helped and fuck kits AI for doing the same thing 💀

[low shard]

AI is actually pretty expensive, RVC could run on consumer grade PCs that cost around 2k with a good GPU, or companies prob use AI specialized (faster, made for AI) GPUs like A100 which are either rented or can cost 100k

You can also get free premium things by doing streaks if i'm not wrong :D

[pallid herald]

Idk sounds pretty sus to me....

[pallid herald]

My username is RockyBfdia

[low shard]

https://docs.aihub.gg/rvc/cloud/weights.com/#streaks you basically do for example a strek (using their service) for 5 days and get a free premium item

[pallid herald]

And I have a 1 day streak

[pallid herald]

Wait so how many days does that take

[low shard]

That's because all services gotta pay for their GPUs which can go up to 100k and the electricity bill monthly
tho tbh, id not really compare weights most expensive plan of 20 euros with the 60 euros of kits ai

[pallid herald]

Like I'm saying how long would that take cuz I have a 1 day streak going

Wait why is my name tag green?

[low shard]

5 iirc, but You can check https://voyages.weights.com/rewards for more info :)

[low shard]

You got the AI Members role, levelling up by talking

[pallid herald]

Oh how unfortunate

May in 2025 using weights was a good month I couldn't remember the ai voices I did

[pallid herald]

Yup I remember those good times

[low shard]

Wdym those good times? it shows you're a new user here

[pallid herald]

I mean using weights dumb dumb

[low shard]

They changed because mostly because the free users were abusing too much of the few limits, meaning loss for them

As everyone loves free things, they were paying for the whole service to be up and running, I hope you understand what I'm talking about

[pallid herald]

Yes it was the best ai website unlike kits AI who had paying stuff before you did

[pallid herald]

I did BFDI ai covers with my song files I downloaded

[pallid herald]

I guess some people just loves to use AI covers so much

Oh someone is here

Um are you gonna say something

[fading skiff]

uuh got a mac book with m1 chip, wanna get voice changer in dc
https://www.youtube.com/watch?v=SxdnGxicJOg&t=328s&pp=0gcJCcYJAYcqIYzv this is the vid my friend send me
cant send pics here so i cant send a screenshot

idk much about gpu n stuff

[pallid herald]

😐

[low shard]

That was the Bot Sticky note about reading the guidelines

[low shard]

That video is outdated asf, delete everything
Are you trying to do e girl trolling like in the video?

[fading skiff]

nah jut wanted a good voice changer

[pallid herald]

:/

[low shard]

Huh? what's wrong?

[fading skiff]

and its hard to find a free one thats good

[low shard]

alright, btw be aware that macs aren't the best for AI, but you can still run wokada deiteris fork

-realtime

[vernal bone]

💻 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

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Deiteris Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[low shard]

1st link

[fading skiff]

alr ty

[hallow thistle]

What trolling? Are you thinking of using an "E-girl" voice model?

[viral mason]

You're on thin ice

Choose your words carefully

[knotty moth]

@mossy sonnet your message deleted and a warn has been sent to you

[viral mason]

they didn't meantion the word egrill in their message but they most likely meant that kind of trolling, there's too many people that join here for that

[edgy brook]

I want to help you.

[mossy sonnet]

When I said “program for trolling,” I wasn’t talking about harassment or anything harmful. I meant lighthearted, joke-style tools like voice changers or soundboards (for example, w-okada, Voicemod, or Clownfish) that people often use with friends in games or calls for fun.

I realize the word “trolling” can sound negative, but I only meant it in the harmless prank/joke sense — like making a funny voice or playing a silly sound effect. I didn’t mention or intend anything about e-girls or inappropriate use.

[barren acorn]

I have RTX 3060 TI 8GB MSI GAMING X what voice changer should I get?

[viral mason]

That actually made me happy, for once someone joined the server and didn't want to use the voice changer for something inappropriate

-# yes it is that common

[mossy sonnet]

i get your point

[viral mason]

But what card do you use btw
Nvidia
Intel
Amd

That'll depend on what you need to download

[mossy sonnet]

Nvidia

[viral mason]

Nice, so I'd say download Vonovox if you want quality over quantity (voice wise)

-Realtime

[vernal bone]

💻 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

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Deiteris Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[viral mason]

Just click the guide and go find the download link for it, if u can't find it I'll just put it here

[mossy sonnet]

is vonovox created by AIhub?

[viral mason]

No it's created by Dr87

He has his own server for the voice changer

[barren acorn]

Any idea

[mossy sonnet]

oh okay

[viral mason]

You would have to use the first voice changer here, Wokada deiteris

@barren acorn

[viral mason]

Btw if you need any help or get confused with stuff just dm me or @ me here^^

[low shard]

that's great, sorry for the inconvenience that @knotty moth assumed for that bad thing, btw be aware that AI Voice Changers are related to sounding like a specific voice model like goku for example, it won't work as a soundboard too, unless you use a separate soundboard program

[mossy sonnet]

i have soundpad

i use the voice changer to record my voice and then take it to soundpad for funny quotes

[knotty moth]

understandable, you could have explained it like that

[mossy sonnet]

next time will do😊

[viral mason]

you could use the voice changer with soundpad, normally something like voicemod works along with okdada or vonovox no issues

[knotty moth]

it's recommended to pre-record it and convert it non-realtime using the original RVC or Applio fork for better quality

[hallow thistle]

To check your NVIDIA GeForce ?TX GPU name, open Task Manager.

[mossy sonnet]

rtx 4070oc

[hallow thistle]

Cool GPU. That will definitely work with W-Okada.

[mossy sonnet]

and vonovox too, right?

[hallow thistle]

Yes.

[viral mason]

wait rtx is nvidia

I may have given someone the wrong directions

[mossy sonnet]

i have an amd rx7650 gpu too can i use vonovox or is it for nvidia users only

[viral mason]

for now, only nvidia users can use vonovox

[hallow thistle]

At least you could've ask them their full GPU name and not just simply GPU brand name.

[viral mason]

dr87 did say at one point he was trying to get a amd gpu so it works for both

[mossy sonnet]

okay thx

[viral mason]

true, I normally go off just if they say Nvidia or something else and if it's not Nvidia I just tell them to use deiteris

[hallow thistle]

Not everything should be Vonovox.

[hallow thistle]

As of now, Vonovox is only for NVIDIA GPU. Deiteris' fork W-Okada has both NVIDIA and AMD/Intel variants.

[mossy sonnet]

and what's the dif between the original okada and deiteris

[hallow thistle]

That's an expected question. It's mostly performance.

[viral mason]

it's kinda better tho, I've tested both voice changers and Vonovox makes models sound better

[hallow thistle]

Deiteris' fork W-Okada opens your browser, while original W-Okada opens its own separate app.

[mossy sonnet]

does that matter?

[viral mason]

I believe Deiteris also is more optimized preformance wise

[hallow thistle]

Whether if it's on browser or as a separate web app, it doesn't really that matter. Instead, Deiteris' fork W-Okada is better than original W-Okada, simply.

[viral mason]

https://tenor.com/view/brain-dog-brian-dog-cooked-wallahi-im-finished-cooked-dog-gif-1849480349705279416

mr beast image scammers, smh

[mossy sonnet]

yes better in like what? models sounds better there, or performance thing only?

[hallow thistle]

If I tell someone who has better GPU to test them both, would you still believe it?

[viral mason]

original wokada is outdated

deiteris is not

https://cdn.discordapp.com/attachments/1301431783677759499/1358209992360198194/attachment-66.gif

[mossy sonnet]

okay guys thanks for your help, have a good day.

[viral mason]

np ^^

[jaunty shale]

I've been using Kaggle mainline for creating voice models, but then recently I stumbled upon this
It's very rare that this pops up. Tried to open RVC URL and I have this.

File and Tensorboard URL works just fine

[viral mason]

Kaggle has mainline on it? I was told it didn't

[rancid pine]

Why does it keep prompting "Pretrained model sample rate does not match dataset audio sample rate" when it's both 48000?

[karmic cairn]

is anyone there who can help me in n8n credentials webhook whtsapp triggere settings??

[teal rain]

can u train a checkpoint for wan 2.2? what kind of dataset do you need?

[teal rain]

like i want to know what kind of dataset you need to do a full fine-tune on models like wan 2.2

[simple ore]

I imagine it would require a lot of videos of different things happening with a very descriptive explanations

and you wont be able to train it on a consumer hardware

[teal rain]

yh so like we would need videos of the certain style to train on with description and not images right?

because lora only needs images

what are some good dataset for full-finetuning

like video length /framerate/how many files (like is 20 files enough or do we need like hundreds?)

resolution?

[rancid pine]

Which pretrained model is best?

[teal rain]

video?

[rancid pine]

No, RVC

[viral mason]

If you're training a voice model use either the og or legacy core

[rancid pine]

Link to the OG?

[viral mason]

That one is used by default there is no link I know of

[rancid pine]

Ah, do I uncheck the "custom pretrained" box, then?

[viral mason]

There's two different versions of legacy core btw, the spin version and contentvec version