okada voice changer

[slender trench]

hello , i can't hear myself or check if the voice changer is working , im 100% sure my mic is working but when i hit start nothing happens

[pseudo tinsel]

oh, uh

its better u use the deitris version instead

it runs faster and is easier to work with

[slender trench]

cool

can i have the link

and im nvidia tho

[pseudo tinsel]

-realtime

[low atlas]

💻 Local Realtime RVC

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

• Wokada Deiteris Fork

Most suggested. GUIDE

• Original Wokada

ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are 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Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated

[slender trench]

tysm

do u think it works rly well?

and is it real time voice changer

[pseudo tinsel]

I used both, they are both Real Time, with the major difference being the Deiteris version is alot faster

im sure there are ways to get og wokada run faster, but deiteris runs good out of the box

[slender trench]

good thing

sorry cause im asking so many questions

but

how do i install this 😭

ik im so dumb sorry.

[pseudo tinsel]

can I ask what gpu u have specifically

[slender trench]

yes

[pseudo tinsel]

like ... the name ?

[pseudo tinsel]

hmm

alright so, It isnt one I know

but looking at the specs overal it doesnt look like its fancy enough for "realtime" delay

[slender trench]

im good with 3s delay

5s

[pseudo tinsel]

delay would probably be ... 0,5-1s, without the audio delays

[slender trench]

no worries

[pseudo tinsel]

im not sure which version to download, because

Im not sure if that gpu is considered above or below, or even supported

but ig that one is the best bet

[slender trench]

gotchu

[slender trench]

thats

the voice changer right?

not nvidia driver 💀

[pseudo tinsel]

its on the "deiteris" wokada voice changer guide

you got different versions depending on ur specs

[slender trench]

alright

then i click download right?

[pseudo tinsel]

there is also a CPU version, however it is painfully slow and delay is probably higher than 5s

[pseudo tinsel]

thats how it works ...

[slender trench]

if there is a version that uses ram more

that would be good

got bulk of them

[pseudo tinsel]

ig the CPU uses effectively exclusively ram only

the gpu-version does too ig, but I dont see a reason why not using your much faster gpu ram as well

[slender trench]

gotchu

ill let u know if it works

@pseudo tinsel btw

ive installed it , extracted it

many files showed up

MMVcserverISO
force gpu clocks
reset gpu clocks

[compact wave]

Click into MMVcserverISO file

Then open the MMVcserverISO application

[slender trench]

gotchu , its loading

downloading sum files

[compact wave]

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

You should follow this guide

Really helps

[slender trench]

uh

..

[compact wave]

Restart it

And see if the there's still the same error

[slender trench]

alright

[compact wave]

If that dosent work, delete the folder and reopen the zip again

[slender trench]

worked

[compact wave]

Nice

[slender trench]

@compact wave is it a website?

[compact wave]

Yup

[slender trench]

i got redirected to a website

cool

af

[compact wave]

It's a localhost

Now you can download models and mess around

[slender trench]

yay

it comes with zero models right?

[compact wave]

Yeah

There should be a model channel in this discord

#🔍│find-models

You just download any of these

[slender trench]

good!!

are they all safe to install?

[compact wave]

Yeah

[slender trench]

alrighty

tysmmm

[compact wave]

Np

Have fun!

[slender trench]

i installed that thing to test it out..

@compact wave ..

[compact wave]

What did you press on?

@slender trench

[slender trench]

i fixed it

now im struggling with discord

[compact wave]

What's wrong

The rvc should have your mic as input and line 1 as output

[slender trench]

it said in the guide input vitural audio
output my headphones

[compact wave]

Then in discord your input should be line 1

[slender trench]

hm

[compact wave]

I think it's the other way around

[slender trench]

in discord

rvc options are like u said

line one output

[compact wave]

Oh yeah in discord the imput should be virtual cable

[slender trench]

input my mic

[slender trench]

uncut?

[compact wave]

Input

[slender trench]

ah

output speakers?

[compact wave]

The output in discord can be you headset

Or whatever you use to listen to people talk

[slender trench]

ah

[compact wave]

Yeee

[slender trench]

i hear nothing 😭

[compact wave]

Are you windows?

[slender trench]

yes

[pseudo tinsel]

click "lets check"

[slender trench]

yh i did

wait

it works

but the delay is 7s

☠️

[compact wave]

Oh show us your setting

[pseudo tinsel]

https://tenor.com/view/renge-shrug-non-non-biyori-idk-duck-face-gif-9724581

[compact wave]

Your chunk might be wrong

And extra

[pseudo tinsel]

he got some quadro nvidia

2GB vram

[pseudo tinsel]

to be expected ig

[slender trench]

[compact wave]

Yeah your chunk is a bit high

Set it to 80

O wait nvm

[slender trench]

done

[pseudo tinsel]

thats too fast

[slender trench]

hm

[compact wave]

Change cpu to your gpu

[pseudo tinsel]

start by pitch 0, formant 0, extra 2.7, and chuck 1000ms

[slender trench]

got like

90 error

[slender trench]

okk

[pseudo tinsel]

go to the cmd

[compact wave]

Restart the application

[slender trench]

done

same error

[slender trench]

done

[compact wave]

What does the cmd say

[pseudo tinsel]

is it, illegal memory access ?

[slender trench]

wym

like overclock ?

[pseudo tinsel]

and this happend when you switched to your gpu ?

[slender trench]

yes

[compact wave]

Prob gpu dosent work rip

[slender trench]

☠️ it works

[pseudo tinsel]

on the gpu ?

[slender trench]

works on the cpu
Gpu gives me that error

[compact wave]

Gpu prob not compatible

[slender trench]

😔

[pseudo tinsel]

yeah ... needs gtx 700-series or higher

[slender trench]

bruh

[pseudo tinsel]

I wasnt sure with his gpu because it wasnt of the gtx/rtx series

[slender trench]

😭

[compact wave]

Yeah idk what spectrum nvidia quad falls under

[compact wave]

You just have to stick with cpu

[slender trench]

yeah

im good with 3s delay

[pseudo tinsel]

ur lucky if u get 3s ngl

[compact wave]

Ok at least it's working

[slender trench]

a

17s delay

holy shi

[compact wave]

Lol

[slender trench]

💀

17s

[compact wave]

Yikes

[pseudo tinsel]

convert to onnx in advanced, and max out the chunk and put extra on 0,5s ig

[slender trench]

fcpe_onyx?

[slender trench]

ah

[slender trench]

YO

3s delay

GOOD SHI

w this options

[pseudo tinsel]

ehh

[slender trench]

i take it back

it broke again..

[pseudo tinsel]

what did the cmd say

[slender trench]

the delay broke

went above 8s again

[pseudo tinsel]

that means it needs more delay

or, simply it cannot process 1s of audio into 1s of ai-voice and forms a backlog

[slender trench]

ah

u did everything possible to low the delay?

[pseudo tinsel]

as far as I know, yes.

[slender trench]

yeah im cooked

tysm for ur time

ill be like a bot replying after 8s 😭

[pseudo tinsel]

if you dont mind me asking, what were you planning on using it for anyway

[slender trench]

exploring new things

first time ive ever seen this thing

[compact wave]

Or until you get a pc upgrade lol

[arctic valve]

wrong, RTX 2060 or 3050 laptop are minimum recommended

1650 may also work but with limited performance in gaming

or you could consider the cloud solution using kaggle