and uses your PC power to power other users, as it stays in background for free credits
#✨│ai-help
low shard
crisp slate
oh alright
my laptop dont got a dedicated gpu
i cant run rvc models
is rvc v2 gonna work on w okada's voice changer
low shard
my laptop dont got a dedicated gpu
either buy a better laptop with dedicated gpu or use cloud services with limited free gpu time
you can't do much about it
AI needs powerful resources
It's not really simple 1 click all free
crisp slate
either buy a better laptop with dedicated gpu or use cloud services with limited...
does 16 gb ram help?
what about beatrice v2 how do i download models for it
it works on cpu too and is faster right
subtle laurel
i downloaded, right after i deleted it, my mic started going static why is it happening?
viral mason
It's ass
viral mason
what about beatrice v2 how do i download models for it
Nobody makes Beatrice models anymore, is ur PC able to run wokada deiteris
crisp slate
Nobody makes Beatrice models anymore, is ur PC able to run wokada deiteris
i didnt test
i dont got a dedicated gpu
viral mason
O
viral mason
If you have the money you could get a 306p possibly
That would run any of the 3 voice changers just fine for normal use
crisp slate
If you have the money you could get a 306p possibly
i got a laptop can it be inserted now?
viral mason
Oh.. a laptop
Yeah laptops are not good for running any kind of ai
Not locally at least, you'd have to use Kaggle or something
crisp slate
Not locally at least, you'd have to use Kaggle or something
oh
crisp slate
Yeah laptops are not good for running any kind of ai
is there something i can do
im not an expert
can laptops have gpu added in them
or do i upgrade to 16 gb ram then it will work
viral mason
Either try out the online alternative on kaggle or try and save up money for a desktop PC with an Nvidia GPU around 20-30 series
That's really your only two options since I'm not sure if laptops can have powerful gpus like that added
crisp slate
That's really your only two options since I'm not sure if laptops can have power...
oh alright
crisp slate
Either try out the online alternative on kaggle or try and save up money for a d...
will kaggle allow me to use this real time voice chnager on discord calls?
viral mason
I have no idea as I have always used the local version and don't even know how to use the version online
I would ask Nick he's pretty smart and knows more about it
low shard
does 16 gb ram help?
doesn't matter
low shard
what about beatrice v2 how do i download models for it
no one trains beatrice v2 models
it's worse than rvc v2
crisp slate
it's worse than rvc v2
oh alright
crisp slate
will kaggle allow me to use this real time voice chnager on discord calls?
@low shard
viral mason
Oh hey look, I summoned him
subtle laurel
Oh hey look, I summoned him
how do i get my mic back to how it was? it used to have no static at all, after i fix it ill re-download again.
viral mason
how do i get my mic back to how it was? it used to have no static at all, after ...
what gpu do u have?
you downloaded an outdated voice changer since earlier u said u got it off youtube
so I wanna get u a better one
subtle laurel
what gpu do u have?
intel
viral mason
No idea, Intel isn't made to run so it struggles horribly with it only know how to properly set up wokada and no idea how to fix any weird issues with it
Maybe try reinstalling it
limber finch
Hello i am running wan2.2 animate and having issue when click run it process a little then get stuck is there anyone who can help me with this runnig it locally using comfyui
main plinth
Hello, hello, good afternoon, how are you? It so happens that I've been away from Colab for quite some time, and I decided to come back, but I don't think I'm up to date with the updates. When I go to my last saved settings page, I get this error.
/bin/bash: line 1: ./MMVCServerSIO: No such file or directory
WARNING:pyngrok.process.ngrok:t=2025-11-20T18:34:27+0000 lvl=warn msg="Stopping forwarder" name=http-18888-4eba4a04-6a04-49ac-9a4c-68c715b272ef acceptErr="failed to accept connection: Listener closed"
Does anyone know how I can fix this?
subtle laurel
No idea, Intel isn't made to run so it struggles horribly with it only know how ...
well, i deleted everything, i mean everything related to the app thingy, and my mic kept doing the static
pastel oak
I set up tg wokada fork for the first time but using audio effects causes the cmd to crash and thus losing connection. Theres no error message either. I cant find anything about it
low shard
will kaggle allow me to use this real time voice chnager on discord calls?
yes, with a limited free gpu time, and you need to verify your account and phone number
low shard
I set up tg wokada fork for the first time but using audio effects causes the cm...
mm that's weird
what's your pc gpu, os and the version you used?
pastel oak
yo nick how you been
low shard
I didn't play much with voice effects on that program but like I don't remember having any crashes
low shard
yo nick how you been
good, wbu?
pastel oak
I am using the latest version b2397 from 3 weeks ago
RTX 3060Ti, Windows
It works without any effects, its only when I use effects that it crashes after cliking start. I tested on server mode with windows wasapi, as well as client mode which uses mme
pastel oak
good, wbu?
good good ty
low shard
I am using the latest version b2397 from 3 weeks ago
RTX 3060Ti, Windows
It wor...
ohh that version used to be pre-release, now it has been changed to a normal release
I just tried it with an echo effect on the output and I haven't gotten any issues
what specific effect did you try?
pastel oak
what specific effect did you try?
I tested on output with equalizer (crashed), i tested chorus (crashed) then tested a random effect for input to see if its just that, but crashed aswell
So im guessing every effect is likely to crash
pastel oak
ohh that version used to be pre-release, now it has been changed to a normal rel...
i can try with an earlier build but doubt itd change anything
pastel oak
ohh that version used to be pre-release, now it has been changed to a normal rel...
It works now. I ran a windows update and that seemed to have fixed it
I dont know the correlation between that maybe theres something in the code that fetches something but ya
low shard
i can try with an earlier build but doubt itd change anything
I tried with the latest one you said, I just didn't know it was now a normal release
low shard
It works now. I ran a windows update and that seemed to have fixed it
That's pretty weird, may I know the exact windows build and update so I can say it in the docs?
And btw are you aware of Vonovox?
gentle crown
how do i use the voice models on a mp3 or wav file
brazen bay
im using vonovox and it just never finishes "warming up voice conversion" i've already closed and reopened it a few times
pastel oak
And btw are you aware of Vonovox?
I havent been in the game since around the time I had to step down as a helper. I knew about it at the time but nothing else, why?
The windows update part is just speculation but it did conveniently start to work again after updating
https://support.microsoft.com/en-us/topic/october-14-2025-kb5066791-os-builds-19044-6456-and-19045-6456-657e5143-6c5d-4401-8efa-1641ca93c051 this was the last update i had before updating to the most recent from november 11
low shard
how do i use the voice models on a mp3 or wav file
What's your PC GPU and OS?
low shard
im using vonovox and it just never finishes "warming up voice conversion" i've a...
What's your PC GPU and OS?
gentle crown
What's your PC GPU and OS?
me or him ?
brazen bay
What's your PC GPU and OS?
4070 windows 11
low shard
I havent been in the game since around the time I had to step down as a helper. ...
Just talking about Vonovox because I thought you didn't know about it, since it does have some performance updates
Ehhh not sure if to add windows update part then 
low shard
me or him ?
Both of you, it's an important info
gentle crown
What's your PC GPU and OS?
AMD Ryzen 5 7600X 6-Core Processor
gentle crown
oh
low shard
4070 windows 11
Windows 11 or 10 right?
brazen bay
Windows 11 or 10 right?
yeah
low shard
yeah
Be sure your windows is up to date, check for GPU drivers
You have the Nvidia App?
brazen bay
yeah
brazen bay
It's an hotfix
alr, downloading it rn
low shard
low shard
@brazen bay
low shard
AMD Radeon RX 7900 XT
Get Applio, an RVC fork: https://docs.aihub.gg/rvc/local/applio/
gentle crown
ok and what does that do
low shard
ok and what does that do
You can use RVC (Retrieval-based-Voice-Conversion) models from #1175430844685484042 on audio files, like ai covers
gentle crown
oh
so whats the difference between that and a site that does it for me
will it sound better if it runs on my pc
gentle crown
yeah
low shard
so whats the difference between that and a site that does it for me
This is a local program, it runs on your PC power
It won't be as easy as using like weights.com, but it doesn't rely on a server and you can control it however you want
A site like weights.com is of course easier, but they use the same program at the end of the day
gentle crown
ok thanks
ill just use the website for now
i was just wondering
if the ai qaulity would be better running on my system insted of the website
brazen bay
if the ai qaulity would be better running on my system insted of the website
i think its more for speed than quality
pastel oak
Just talking about Vonovox because I thought you didn't know about it, since it ...
are you up to date with all of the variants? is that the current favorite still? i havent looked into the voice changer in a while, for example tg released after i left so thats why im interested atm
crisp slate
yes, with a limited free gpu time, and you need to verify your account and phone...
but i saw it was really complicated and whats up with the codes
i saw many people on youtube doing it with google collab can i use it instead i didnt see keggle tutorials
whats better
viral mason
i saw many people on youtube doing it with google collab can i use it instead i ...
Kaggle gives 30 hours Google will at max give 4
brazen bay
The Nvidia server is pretty useful
That fixed it, is there anything else I need to know about vonovox?
low shard
ill just use the website for now
What site are you talking about btw?
low shard
That fixed it, is there anything else I need to know about vonovox?
Yw and well, pretty much every setting is explained in the ai hub docs, be sure to play with the pitch and try an extra time like 4.5
brazen bay
Yw and well, pretty much every setting is explained in the ai hub docs, be sure ...
The docs didnt include anything about smart sine, is that important?
crisp slate
Kaggle gives 30 hours Google will at max give 4
is the performance same and the download method too
viral mason
I believe so yes
low shard
are you up to date with all of the variants? is that the current favorite still?...
There isn't much new tbh, but wokada tg develop fork is only QOL and UI, the developer doesn't work on performance related things
humble plinth
When I make L4 GPU or A100 CPU on Colab site An error occurred during voice conversion. Check command line window for more details. I'm getting an error, what's the solution?
nocturne mural
simple ore
When I make L4 GPU or A100 CPU on Colab site An error occurred during voice conv...
Check command line window for more details
thin veldt
Hi all! is it possible to request help to get feedback from a study about AI?
green canyon
can someone help me download the voice changer im trying to troll but its not working
viral mason
can someone help me download the voice changer im trying to troll but its not wo...
What gpu do u have?
green canyon
What gpu do u have?
i dont know how to see that i want some help on a call ive never done this before
viral mason
i dont know how to see that i want some help on a call ive never done this befor...
Just open your task manager and click the performance tab on it
viral mason
As long as you're not using a laptop you should have a decent gpu
knotty moth
im using intel gpu
intel arc?
green canyon
all it says is intel(R)
viral mason
You're cooked 
green canyon
why
green canyon
You're cooked <:cute_plead:1420419726596771911>
how
low shard
The docs didnt include anything about smart sine, is that important?
Leave it on, it's a smart algorithm that doesn't process noise at all, as said in the 1.6.5 changelog:
One of the most useful features I have developed thus far. This is a replacement for the old VAD and takes its spot as the checkbox. This is a smart algorithm that will disable the SINE generator's ability to translate noise to speech. This means most noise like isolated static and random noises can no longer be mapped to false speech and generate static, NSF generator will not process them. However, extremely loud and consistent noises like slamming on a surface may still trigger false positives , however most normal noise will be stopped. I will continue to improve this, but the first version is very effective so far.
Also, the docs have just been updated on that part
night blade
Anyone has tried any good ai humanizers?
low shard
Anyone has tried any good ai humanizers?
Ai humanizers are just LLMs rephrasing
I wouldn't trust AI Detectors
knotty moth
because it can never be a human
knotty moth
all it says is intel(R)
it is integrated one which won't work
green canyon
ok i already found a different one that will work with the models
cursive oriole
Hi, Can anyone help me?
I'm looking for singer models, but not artists or celebrities, rather singers who have trained their voices and have good quality, similar to the AUDIMEE models.
knotty moth
When I make L4 GPU or A100 CPU on Colab site An error occurred during voice conv...
then stop subscribing colab pro
Check command line window for more details
also you haven't explained the problem and which colab page being used
!howtoask
green canyon
NVM it was a scam you have to, pay money
viral mason
NVM it was a scam you have to, pay money
???
green canyon
they were like ya this is free this is sooo free and it wasnt btw if i had the right gpu i would have this voice changer
crisp slate
Just open your task manager and click the performance tab on it
is this good i dont understand what does it mean
viral mason
is this good i dont understand what does it mean
it doesn't say what the numbers are for ur amd card
weird
did u click the right one
crisp slate
did u click the right one
i went to task manager and porfamnce
ill send full screen ss
viral mason
hmm
I guess u could try wokada deiteris
first is vac lite which is a virtual cable u will need to use it in games second is wokada deiteris (the voice changer)
crisp slate
first is vac lite which is a virtual cable u will need to use it in games second...
oh
crisp slate
I guess u could try wokada deiteris
will it work?
cuz w okada voice chnager didnt
and whats the difference
is quality same and stuff
viral mason
will it work?
it should
better quality than what you would fine on yt
bc the stuff on yt is all outdated
crisp slate
it should
is it the same as the one i downloaded
why would this work and not the one i downloaded
viral mason
idk what the one u have currently is
crisp slate
idk what the one u have currently is
the normal one
crisp slate
viral mason
there is no normal one just different versions
crisp slate
viral mason
anything to do with beatrice means it's super old
crisp slate
anything to do with beatrice means it's super old
what
i downloaded the latest one
wait ill send ss where i downloaded
viral mason
from where
crisp slate
i followed this video and used the link in description https://youtu.be/SxdnGxicJOg?si=HYeg7KXvsrEMvJ2J
crisp slate
that version they use is over a year old
but on the link they gave i downloaded the latest one
viral mason
but on the link they gave i downloaded the latest one
still the code to that is outdated compared to the current stuff
viral mason
idk what that is
crisp slate
viral mason
crisp slate
https://tenor.com/view/marvin-sml-the-ropes-are-coming-what-the-fuck-gif-8234251...
whaa
viral mason
how would that one work does it not require a gpu? and is it that different
if u don't have a gpu ur cooked
crisp slate
if u don't have a gpu ur cooked
i already told u everything before did you forget
i dont have it
viral mason
i already told u everything before did you forget
I have a very bad memory
crisp slate
I have a very bad memory
viral mason
?
crisp slate
my specs
viral mason
is that not a gpu?
viral mason
🤷♀️
viral mason
the one i followed with the video didnt
tbh no clue at this point
crisp slate
tbh no clue at this point
vro
craggy bough
will it work or nah
no
craggy bough
is kaggle my only option
unless you wanna buy a pc/laptop with a dedicated gpu
crisp slate
unless you wanna buy a pc/laptop with a dedicated gpu
i cant until atleast 3 years so yea
my dumbass didnt know gpu stuff
and just bought this
should have spend a little more and got one with dedicated gpu
sharp glacier
(Removed this due to nvm)
mortal nymph
does anyone know some settings to change to make their background vocals less glitchy on rvc
knotty moth
dont expect to run locally with just integrated gpu
and 8 gb ram is too low for most modern applications under win 11, unless you switch to some lightweight linux distro
crisp slate
and 8 gb ram is too low for most modern applications under win 11, unless you sw...
i canb upgrade to 16 gb ram then it will work?
craggy bough
i canb upgrade to 16 gb ram then it will work?
if you arent using it while olaying games and are fine with a massive delay + stutters
then yes
crisp slate
if you arent using it while olaying games and are fine with a massive delay + st...
will it work if im in dc voice call
i dont want too much delay or too much stutters
tawdry dock
I’m having trouble getting mmccs sever to open on mac im using a mac min m4 But it says malware may harm your mac move to trash or done
tawdry dock
Know what ?
craggy bough
i dont want too much delay or too much stutters
expect a delay of at least a few seconds
crisp slate
expect a delay of at least a few seconds
alright
i have downloaded the w okada voice changer latest version
using this tutorial
its not working rn so your saying it will work on 16 gb ram?
crisp slate
https://docs.aihub.gg/
thanks
crisp slate
https://docs.aihub.gg/
its keeps telling me to use online hosts like kaggle
my pc specs are 8 gb ram 5500 u amd ryzen 5 lenovo slim idea pad 1 no dedicated gpu
is it going to work or should i use kaggle
i downloaded the orginal one and it wasnt working is deitris gonna work
craggy bough
its keeps telling me to use online hosts like kaggle
then use an online host
hasty kite
hi, what are some good prompts to generate video game music?
crisp slate
my pc specs are 8 gb ram 5500 u amd ryzen 5 lenovo slim idea pad 1 no dedicated ...
but i got the recommended stuff
the things recomended in the ss
crisp slate
then use an online host
is it the same
craggy bough
why do you keep asking the same questions over and over
read what it says
and decide for yourself
do you want the bare minimum
craggy bough
then chatgpt it
crisp slate
ok
hasty kite
specifically sci fi type music, they all sound generic lol
crisp slate
last thing i downloaded the vb audio cable but the guide says vac lite should i delete the one i have and download the one which the guide sayd does it make a difference
lean mesa
the same volume can not be used as both the source and destination 
trying the 5k series download it said extract the zip folder and it will do the rest automatically but it's not doing that
viral mason
???
lean mesa
that's the error it gave me
viral mason
0 context to this
lean mesa
when i tried to extraxt the zip
viral mason
What zip
lean mesa
trying the 5k series download it said extract the zip folder and it will do the ...
^^^^^^^
lean mesa
well i can't post screenshots in here so i guess i will be super specific
real time voice changer 5k series fork
viral mason
Ooh cool, same as mine
lean mesa
yeee one sec let me show you what it's telling me
Download all 3 files, then extract the .zip file, it will automatically extract ALL 3 FILES into one. Then open the MMVCServerSIO folder and run MMVCServerSIO.exe (or called MMVCServerSIO if you don't have extensions activated).
viral mason
U should really use Vonovox tho tbh, it's way better than both wokada deiteris and th fork
lean mesa
ah is that something new
viral mason
First guide has it
lean mesa
ok cool i will do that instead then ty
viral mason
Np! If u need help or have questions I'll be here until I fall asleep
Which could be at any time
lean mesa
yeye i'll check out the guiode should be pretty straight forward i imagine but yea the other one said extract one folder and it will do the rest but that was not the case lol
viral mason
This one is just one folder, on the GitHub page it's just a zip called source code, and then when u extract it run setup.bat then start.bat
The layout of it is very different tho from wokada
But it functions the same just way better
lean mesa
oh nice this one can use two gpu's as well
lean mesa
This is ONLY For users with 2 GPUs in the same system, you must do the following:
Open NVIDIA Control Panel
Go to Manage 3D Settings > Program Settings Tab
Add python.exe from the Vonovox runtime folder (runtime\python.exe)
Set both settings "CUDA - GPUs" and "OpenGL rendering GPU" to the GPU you want to use for conversion```Yea ur 5070 is perfect for it
Btw I seem pretty chill would you wanna be friends?
Oh shit you're also in the Film Actors Guild?
I was just about to ask if u had vrchat
viral mason
Ted is awesome
viral mason
Haven't really been showing up to the acting things recently because of work or just not being home when it happens
lean mesa
yeee same i reformatted my pc for my new monitor and wanted to get the voice changer to troll with on my xbox
can route it into console with the remote play thing lol
lean mesa
yeee lol
viral mason
What are u going to use btw, I'm guessing maybe Winnie the Poo?
lean mesa
not sure i started playing overwatch again so might do something on there
lean mesa
yeye true
i also have 50 different avatars on vrchat that i swap into so i could get those voices again and mess around
viral mason
I have vrchat + so I have a lotttt of Avis and voices for most
viral mason
How???
lean mesa
but not going to leak the secrets xd
viral mason
O
lean mesa
not in public anyway but it's a routing thing
viral mason
Fair
lean mesa
can be used on phone calls which is not the best thing for scammers to have access to lol
viral mason
U must be a super genius
lean mesa
only sometimes 
oh this uses asio audio as well what the helly
lots of settings on here
viral mason
I'm too slow to even figure out sometimes how long to cook the chicken nuggies
-# I am 19 btw
viral mason
lots of settings on here
Full screen the app, it is full of settings to mess with, and some you shouldn't
tawdry dock
Ok im trying to figure out how’s this works. I mange to figure out how to disable my mac auto mawle ware and I opened the files but it just keep piping up numbers and trying to connect to sever on my terminal ? Any advice on what to do ?
viral mason
?
tawdry dock
Yea I have to show pictures to explain it
Maybe tomorrow I’ll try again i need my sleep
lean mesa
Full screen the app, it is full of settings to mess with, and some you shouldn't
what settings would max out the quality i see block size and extra time and some advanced settings
viral mason
what settings would max out the quality i see block size and extra time and some...
Tbh just put crossfade at 0.15
And uhhh
Extra at 4.0
I think that's possible
Is it?
viral mason
Ok for block size I'd say keep where it is for now
lean mesa
kk
viral mason
Anddddd did u figure out the mic settings and where the pth and index files go
viral mason
One of mine perhaps? 👀
lean mesa
cassidy from overwatch lol
viral mason
I haven't made any ow models sooo def not mine lol
Btw for female models if you're a dude pitches around 3-10 maybe 12 at the most extreme is best
Try starting at 3 then go up
lean mesa
yea going up by 3 keeps it in the same key as the base pitch ect
viral mason
Huge tip to make it more believable, try to replicate how the character speaks, like if they have an accent or something
Sounding monotone will make it monotone ect
Be expressive
lean mesa
yeee i have been using the wokada stuff for awhile just didn';t know about this new stuff ty for the advice though
viral mason
No problem:3
I have ppl use it after I help them and they say it sounds robotic and either they're using an old crusty e-girl model or they speaking like a regular person into the mic
No life at all 😞
Anyways I'm going to bed now was nice to meet you, can't wait to possibly see u in VRC sometime, my user is the same as my display name here
humble plinth
Google colab is good ?
low shard
you don’t install it, it’s on https://aistudio.google.com
humble plinth
Kaggle or google claoud good ?
hallow thistle
Kaggle or google claoud good ?
Both Google Colab and Kaggle have their own trade-offs, depends on your use cases.
!howtoask
humble plinth
Both Google Colab and Kaggle have their own trade-offs, depends on your use case...
for instant conversation realtime ?
hallow thistle
for instant conversation realtime ?
Be more specific on which you're looking for (e.g. W-Okada, RVC).
humble plinth
E.g. ?
Is this the only app? I'm asking out of curiosity and I feel like an idiot.
@hallow thistle
hallow thistle
The "e.g." in paragraph typically stands for the Latin phrase "exempli gratia", which simply means "for example" in English literary. If you have no idea what any of these words even mean, you can pick up an English dictionary. 
But when you ask too many questions about the same thing (either Colab or W-Okada), I'm not sure what to answer.
hallow thistle
By the way, this is Google Colab. The word "notebook" refers to a type of document that includes blocks of code intended to run on Jupyter Notebook-related website, which itself is also used in Colab. This notebook in my screenshot is not W-Okada or RVC; it's Music Source Separation Training (MSST), a specific type of "Ultimate Vocal Remover" a distinct program used for separate audio into stems.
humble plinth
Oh okey
humble plinth
you dont seem to be fluent in english
İ am turkhis
I am a Turkish voice female package producer
We use the Fivem role-playing game
humble plinth
The site gives the same error, what is the reason?
hallow thistle
humble plinth
Where did you get this Colab notebook from?
from the application on the website
paid version
hallow thistle
Not the correct answer. I mean like, which tutorial or guide did you get this Colab from or were you searching on Google and found one?
hallow thistle
The hell. I literally asked you the question, and you still didn't answer any correct answer.
humble plinth
Not the correct answer. I mean like, which tutorial or guide did you get this Co...
yes
There's a Turkish YouTuber who does this, but he hasn't said anything about it. Do you know the solution?
hallow thistle
For translation from your Turkish to English, I'd suggest DeepL Translator, or Google Translate as a more quick translation. You wouldn't answer anything too simple that would make things more confusing.
If you got a Colab notebook link from tutorial video on YouTube, they outdated. Try this one. https://colab.research.google.com/github/tg-develop/voice-changer/blob/master-custom/Colab_RealtimeVoiceChanger.ipynb
humble plinth
So what is your suggestion?
humble plinth
Is it a definitive solution?
hallow thistle
Whatever you asked.
humble plinth
ok
I trust you
hallow thistle
There's a guide on how to use "Tg Develop's W-Okada Colab". The guide is all English; if you still don't understand English you can use a translator right away. https://docs.aihub.gg/realtime-voice-changer/cloud/tg-develops-w-okada-fork-colab/
humble plinth
There's a guide on how to use "Tg Develop's W-Okada Colab". The guide is all Eng...
ok my bad
Is this a good setting overall?
@hallow thistle
hallow thistle
Is this a good setting overall?
You sure that's Tg Develop W-Okada Colab? If so, it is supposed to look like in my screenshot. In your screen, it's Deiteris W-Okada, which is a different one. https://cdn.discordapp.com/attachments/1439159243776331899/1439241897498513440/image.png?ex=6921b6f8&is=69206578&hm=971798b80329a7eeac4f6f5a0ddef6fee9e312785cb27a3087bcb3e69cda2fc7&
This is Deiteris W-Okada. https://cdn.discordapp.com/attachments/1425025269949010070/1426904432058171503/image.png?ex=6921a7cf&is=6920564f&hm=4cd2874c36361f41c8f4356cefd6c8c4aa8bb4fd619daefe8fd5a909d0810b9e&
hallow thistle
While both Tg Develop and Deiteris W-Okada forks would work, the Tg Develop W-Okada fork generally has more recent features than Deiteris W-Okada; Deiteris W-Okada, while stable, its latest update was last year.
With "T4" GPU selected, here's your settings:
Chunk: around 80 - 100 ms
Extra: 2.7 s (above 2.8 s might possible)
GPU: Tesla T4
Pitch Detection (F0): rmvpe
Input: microphone
Output: Line 1 (Virtual Audio Cable)
Monitor: this one optional; you can set this to your speakers/headphones to hear the program.
humble plinth
hallow thistle
Oh, so you have the more powerful and expensive "NVIDIA A100" selected in Colab. While that one is powerful for other large AI trainings, it would be too powerful for smaller AIs like W-Okada and RVC than necessary. The T4 is still enough for smaller AIs and eats less compute units than A100, unless you're trying to flex your money that way.
humble plinth
oh okey
humble plinth
Oh, so you have the more powerful and **expensive** "NVIDIA A100" selected in Co...
Why does this voice sound different from before, a bit robotic?
hallow thistle
Additionally, you can go to "Advanced settings" on the program, and try set "crossfade overlap" to 0.15. Otherwise, if the audio still robotic, it can be the voice model itself, so nothing much you can do about it aside from switching to another model.
hallow thistle
Yes, of course. Click "close" to close the setting popup.
humble plinth
okıey
sand flare
@hallow thistle hello what does the sample rate do in vonovox ?
tame oracle
<@561807329977434122> hello what does the sample rate do in vonovox ?
Hi may i ask to not ping just one helper, if you need to please just ping the helper role and when one of us can help you we will
rare tundra
guys what is the best quality female voice model?
mystic salmon
guys what is the best quality female voice model?
there isnt any
search for egirl voices or f4m voices
on weights.gg
viral mason
search for egirl voices or f4m voices
Ew
viral mason
guys what is the best quality female voice model?
🤨
viral mason
How do i install okada?
,
What gpu do you have?
viral mason
I dunno if your GPU is gonna be able to run anything of the voice changers but you can try
viral mason
Vonovox can only run on Nvidia, but wokada deiteris and wokada tg fork possible could work with it
But still u could try Vonovox to see if it even works
viral mason
Crossfade at 0.15
Extra at 4.0
Block size just depends on your gpu but default is usually fine
mystic salmon
can realtek microphones work on this program?
viral mason
And the pitch always have at 0 unless it's a female model, pitches 3-10 maybe 12 are good
mystic salmon
And the pitch always have at 0 unless it's a female model, pitches 3-10 maybe 12...
that depends on my voice
i think
viral mason
can realtek microphones work on this program?
Any headphones work as long as the mic isn't really bad
mystic salmon
Any headphones work as long as the mic isn't really bad
nah dude the thing is
viral mason
that depends on my voice
Yea but most models work best at 0 if they're male and you're male
mystic salmon
the microphone isnt with usb
viral mason
the microphone isnt with usb
Wdym
viral mason
Like the round thing
mystic salmon
ok ill try
viral mason
Lemme know if it works
viral mason
Why are you looking for those kinda voices?
mystic salmon
Why are you looking for those kinda voices?
do i start start.bat or setup.bat then start.bat
viral mason
Setup.bat then start.bat
mystic salmon
ok cool
@viral mason
its saying a new release of pip is available 25.2 -> 25.3
do i wait
viral mason
Should download that version then, unless it's doing it for you
mystic salmon
Should download that version then, unless it's doing it for you
its keep spamming that after a big chunk of texts
nevermind i got it
viral mason
U can download Vonovox since u have a 30 series Nvidia card, just go to that first guide right there and download vac lite and vonovox off GitHub, all u need should be there
-rt
finite raven
U can download Vonovox since u have a 30 series Nvidia card, just go to that fir...
alr thanks
is there possibly a video tut ?
viral mason
Nah it's not that complicated to download, for Vonovox just run setup.bat then start.bat and vac lite run setupx64
finite raven
all good thanks
viral mason
Np!
minor pollen
hey guys can someone whos rly into ai videos help me
frail mica
any one work with livekit ?
i fellow vieso and take smae code but i get this error each time i speak
can any one hlep me
line1 INFO received job request {"job_id":"","room":""}
line2 INFO process initialized {"pid":696263}
line3 ERROR unhandled
{"job_id":"AJ_cE2ZfCy7vvqq","room_id":"RM_X36Vc72HeCj6"}
Traceback:
line4 File "/backend/agent.py", line 32
async with model.start(room=ctx.room) as session:
AttributeError: 'RealtimeModel' object has no attribute 'start'
Nvidia Encoder is supported.
Nvidia Decoder is supported.
ERROR unhandled exception {"job_id":"AJ_cE2ZfCy7vvqq","room_id":"RM_X36Vc72HeCj6"}
Traceback:
File "/backend/agent.py", line 32
async with model.start(room=ctx.room) as session:
AttributeError: 'RealtimeModel' object has no attribute 'start'
viral mason
any one work with livekit ?
i fellow vieso and take smae code but i get this err...
what is livekit?
tawdry dock
Anyone know if weights paid version gives you unlimited hours of audio recordings as well as longer audio files to make. Like 30 minutes audio files?
frail mica
what is livekit?
yes i watch video in youtube but it's a littel bit old
viral mason
Anyone know if weights paid version gives you unlimited hours of audio recording...
don't pay :(
tawdry dock
Question I have a Mac what file should I download? The one that says 3 months ago or the 2.14- alpha zip
viral mason
yes i watch video in youtube but it's a littel bit old
I still dunno what livekit is
fossil sage
GUYS whats the best ai image generator
muted mango
Hello I am looking for an RVC or an alternative for people with an integrated gpu
viral mason
Hello I am looking for an RVC or an alternative for people with an integrated gp...
What gpu do u have? I'm not sure I understand
low shard
Hello I am looking for an RVC or an alternative for people with an integrated gp...
if you got an integrated GPU like Intel UHD, it's not really that good for Local AI Tasks
I mean it could run on CPU but very slowly
did you check if you have any dedicated gpu?
muted mango
I do have one but my girlfriend doesn't and im trying to find an alternative for her
low shard
I do have one but my girlfriend doesn't and im trying to find an alternative for...
she can either:
- uses your PC
- buys a dedicated GPU
- use cloud: remote good pc
About cloud, there are different services with limited free gpu time:
Train (make) RVC Models on cloud:
- Prepare the Dataset
- Setup RVC:
Choose a cloud way to use RVC,
- Google Colabs (max 4 hours of daily T4 16gb gpu not granted for free, but easy to use, there's a paid tier):
- Kaggles (a bit harder to use and needs phone number but gives 30 hours weekly of better gpus, either T4x2 16gb each or P100 16gb, only free):
- Lightning.ai (Kinda hard, needs login, no issue with web uis or anything, but only free 15 credits monthly, Free Studios run 24/7 but require restart every 4 hours. There's a paid tier):
- Applio (UI)
- Be sure to know about the tensorboard
Google Colab = Easier but risk of getting disconnected
Kaggle = Harder but more gpu time
If you want the easiest way and for free, is using https://weights.com/ which uses RVC
RVC Inference (use models) on pre-recorded audio on Cloud
You can use either:
- Weights.com: Easiest Possible Ever Automatic
- Applio UI Colab: RVC Fork with some extra features like TTS
- RVC AI Cover Maker UI: Automatically Separates the Vocals and Instrumentals, converts the voice and mixes them back
Let me know if you didn't understand something
and nice Alastor pfp lol
tawdry dock
Ok so I got this far and now I’m completely lost I’m have my mic on but I can’t seem to record a voice? Also when I tried just putting a pre recorded test clip it was just my voice. Anyone got any tips and trick to help me out thks
viral mason
Ok so I got this far and now I’m completely lost I’m have my mic on but I can’t...
that's so old
I can see dust on it
tawdry dock
that's so old
Oh is it an out dated model is that why it’s not working ?
tawdry dock
the whole software
So what free soft ware would you recommend then ?
pulsar osprey
i have a problem when i have everything set up my ai voice is laggy but on basic models its working with no issues (RX 6600 XT)
viral mason
So what free soft ware would you recommend then ?
depends, I think I remember ur username do u have a nvidia or amd gpu?
viral mason
i have a problem when i have everything set up my ai voice is laggy but on basic...
there are no basic models that come with any of the voice changers, what u have is the original wokada which is super old and outdated
u should try out woakda deiteris
-rt
tawdry dock
depends, I think I remember ur username do u have a nvidia or amd gpu?
lol nope a filthy Mac user with an m4 Mac min
feral crater
Are paid monthly subscriptions worth it for those who don’t have ultra end GPU’s to vocal train?
I’m currently running an RTX 2080 Ti, probably meh for today’s standards
viral mason
Are paid monthly subscriptions worth it for those who don’t have ultra end GPU’s...
just like, use kaggle
it's free and has 30 hours a week for free users
feral crater
just like, use kaggle
Interesting, how would I implement it with RVC for training?
viral mason
Interesting, how would I implement it with RVC for training?
all u really need to do is make an account with kaggle, phone verify and do this to set up the kaggle applio space for model training
this doesn't really explain in detail how to train but more of just what to do
kinda messy
feral crater
all u really need to do is make an account with kaggle, phone verify and do this...
That’s really cool!
I wish there a more in depth about it within Kaggle.
I understand Applio, just the whole Kaggle I need more learning.
However I appreciate you sharing, extremely helpful.
Thanks!
viral mason
That’s really cool!
I wish there a more in depth about it within Kaggle.
I und...
np I'm currently not mentally well enough to properly explain how to use it in depth ^^
I could show you how to use it but not explain
prime badge
Has any one else experienced the RVC voice sounding good when listening to it via monitor but when it comes out of the other people’s speaker it sounds horribly like ai/robot? and where you able to resolve it?
static blade
where i can download rvc?
subtle cedar
there's still Gradios? I mean unfortunately that Weights needs pay now and that Astra Labs already gone so there's no create AI Covers in easier way anymore ☹️😭😢
-colab
subtle cedar
Which is colab for making AI Covers the most among these?
twin falcon
my voice changer is more using my cpu than my gpu
craggy bough
my voice changer is more using my cpu than my gpu
switch it from cpu to gpu mode
subtle cedar
can anyone pls answer me which is colab for AI Covers and I hope it's free even if it's limited thanks
tame oracle
can anyone pls answer me which is colab for AI Covers and I hope it's free even ...
the one that says "AI cover maker"
subtle cedar
the one that says "AI cover maker"
Yeah but I mean there's another colab for AI covers as well because probably some other Gradios are better
tame oracle
Yeah but I mean there's another colab for AI covers as well because probably som...
So, you're telling me you know about that one and that there is better ones?
subtle cedar
So, you're telling me you know about that one and that there is better ones?
Yeah hehe
But Idk because I'm not familiar with newer/modern/current Gradios or Colabs yet
tame oracle
But Idk because I'm not familiar with newer/modern/current Gradios or Colabs yet
neither am i, but i do know that Kaggle has been said to be better than colab mainly for more free GPU time
-kaggle
subtle cedar
neither am i, but i do know that Kaggle has been said to be better than colab ma...
Yeah thanks but I think that's only works for laptops or PCs especially Windows
tame oracle
Yeah thanks but I think that's only works for laptops or PCs especially Windows
Kaggle is in a way a upgraded better version of Google Colab, you can only use Colab or Kaggle on a PC
subtle cedar
Kaggle is in a way a upgraded better version of Google Colab, you can only use C...
Yeah but I only have a phone and tablet so idk if that Kaggle works for mobile or Android
tame oracle
Yeah but I only have a phone and tablet so idk if that Kaggle works for mobile o...
sorry, don't know
subtle cedar
sorry, don't know
Oh okay, I think I just use Gradio or just make a "How Would__sing" videos that aren't AI Covers that using original version of the songs then or I will use Kaggle when I have a laptop if that's not work for mobile
tame oracle
Oh okay, I think I just use Gradio or just make a "How Would__sing" videos that ...
sounds like you got a plan
subtle cedar
sounds like you got a plan
Yep hehe
hallow thistle
gt 730
NVIDIA GeForce GT 730 is a decade-old GPU from 2014, has just one GB of VRAM although some might have 2GB, and attempting to get either W-Okada or Vonovox to work with that GPU would likely result not working afterall or run on your PC CPU instead. There have been several members here trying to run with GeForce 7xx GPU which won't be very ideal. Besides that, neither guide says GeForce 700 as a minimum. 
AMD Ryzen 9 9950X3D is a CPU released in early 2025, several years ahead of GT 730. While it sounds possible to run W-Okada with only CPU, it would perform much slower than a dedicated GPU like AMD Radeon RX 9000 series or NVIDIA GeForce RTX 50.
hallow thistle
<@561807329977434122> hello what does the sample rate do in vonovox ?
To know about the term "sample rate", look up on Google. The "sample rate" is something like 44100, 48000, 96000 and 192000, all of which typically expressed in Hertz (Hz), although some might express in khz (kilohertz; 44.1khz, 48.0khz). It indicates how much a digital representation captures the original analog signal in a second; the more hz number the more audio quality, while lower is lower quality. The "sample rate" is not just in Vonovox but also found in other audio-related softwares, audio files, and even your speaker/microphone settings.
knotty moth
amd ryzen 9 9950X3D
that's a terrible setup, overkill cpu and too weak gpu that has no resell value. sounds like you could exchange it with 5800X3D + RTX 3060 12 GB / 4060 Ti 16 GB setup (prob in used market pricing) which is more capable to run voice changer and also for gaming
hallow thistle
For a recommended number in sample rate, either 44100 or 48000 will work but the latter is more preferred. 96000 and 192000 are typically used in professional audio, where the quality is more important than general listening. 
sly cosmos
Is there a free ai thats good at naturally removing objects
hallow thistle
Is there a free ai thats good at naturally removing objects
Google Gemini's Nano Banana can also do that, though you can try this one and spot the result.
sly cosmos
Google Gemini's Nano Banana can also do that, though you can try this one and sp...
I need a ai where I can circle the thing I wanna remove. I have a ai picture that has too many hands and I wanna show it which one to remove
knotty moth
I need a ai where I can circle the thing I wanna remove. I have a ai picture th...
stable diffusion/flux inpainting could work but gemini nano banana or flux kontext could do better to preserve untouched details
just tell which part to edit/remove in a prompt
if not sure, crop to the part to edit, process it, and put it back on the original image
solid smelt
just installed the wokada , its not working . like no output or input coming .
tame oracle
just installed the wokada , its not working . like no output or input coming .
!howtoask
solid smelt
!howtoask
NVIDIA RTX 5050 , windows 11 , just tried wokada voice changer it has delay in voice too
tame oracle
NVIDIA RTX 5050 , windows 11 , just tried wokada voice changer it has delay in ...
well you have a NVIDIA 50 series card, you should use vonovox
-rt
solid smelt
well you have a NVIDIA 50 series card, you should use vonovox
yeah have installed
it
how do i use it ?
tame oracle
yeah have installed
didn't you just say you have o okada installed?
solid smelt
added the voice models
tame oracle
O Okada and Vonovox are different, what does your UI look like? Did you get it from a yt video?
tame oracle
<@1416622683163983902> dahh
What is your question? We'll see if i can answer it
tame oracle
You mean Text to speach?
tame oracle
O okada is a real time local voice changer program
prime ridge
Okay, and how many minutes of voice can it generate in one go
solid smelt
didn't you just say you have o okada installed?
yes i did it first . it was not working . then installed vonovox
solid smelt
O Okada and Vonovox are different, what does your UI look like? Did you get it f...
yes from yt video
prime ridge
yes i did it first . it was not working . then installed vonovox
Does it work
solid smelt
i cant share pic here
tame oracle
Okay, and how many minutes of voice can it generate in one go
O okada doesn't have a limit other than how strong your GPU is, its not a TTS program either
prime ridge
O okada doesn't have a limit other than how strong your GPU is, its not a TTS pr...
Okay,
tame oracle
yes from yt video
ok, well anything from a yt video is outdated
prime ridge
What a waste of time. Soab
tame oracle
delay
it will have delay for processing of your voice and depending on how good your GPU is
solid smelt
ok, well anything from a yt video is outdated
downloaded vo. from here
tame oracle
downloaded vo. from here
Ok, and im sorry, your issue was what again?
solid smelt
it will have delay for processing of your voice and depending on how good your G...
yes , how can i run it on vo
solid smelt
Ok, and im sorry, your issue was what again?
you told me i got rtx 50 series so should use vonovox
like to know how that works
tame oracle
ok, so you downloaded it, run "setup.bat" before anything
solid smelt
ok, so you downloaded it, run "setup.bat" before anything
i did it and added voice model .
tame oracle
i did it and added voice model .
Okay, so now you should just need to start the voice changer and you will have to mess with the settings as things vary
guide here too
solid smelt
https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
have downloaded and installed precompiled setup
need to setup the audio
tame oracle
need to setup the audio
you installed the virtual cable right?
tame oracle
thats an driver ?
? The virtual audio cable is for routing the output of the voice changer to discord and games and such
solid smelt
? The virtual audio cable is for routing the output of the voice changer to disc...
yes have installed it
solid smelt
Okay, so what do you need help with? You seem to have it all under control
voice change is not working
input and output should be our devices right
tame oracle
input and output should be our devices right
Input should be your Mic and output for now should be your headphones/speakers while we get it working
solid smelt
Input should be your Mic and output for now should be your headphones/speakers w...
ahh
2sounds at time , how do i fix that one
tame oracle
2sounds at time , how do i fix that one
2 sounds at a time? Never heard of that, what are your settings?
solid smelt
2 sounds at a time? Never heard of that, what are your settings?
yes my orginal sound and model voice
tame oracle
tame oracle
yes my orginal sound and model voice
ur settings, what are they?
tame oracle
Okay, say something again
solid smelt
you hear ?
tame oracle
Yea, ive never seen that before, i honestly don't know
sounds like your chunk might be a bit low
solid smelt
tame oracle
i apologize as i feel im not being able to do much
solid smelt
sounds like your chunk might be a bit low
how do i change it
tame oracle
in Vonovox its called block
solid smelt
ahhh
tame oracle
under voice settings
turn it up by like 2-3
maybe a little more, sounds better to me
solid smelt
ahh
tame oracle
definitely better, seems you just need to mess with the block size and extra to work well with your GPU
solid smelt
definitely better, seems you just need to mess with the block size and extra to ...
is it working on idscord ?
tame oracle
is it working on idscord ?
If ur talking right now then i don't hear it but i was before yea
solid smelt
If ur talking right now then i don't hear it but i was before yea
ahm
which input device should i select in discord
tame oracle
which input device should i select in discord
Discord input should be the virtual cable output
solid smelt
Discord input should be the virtual cable output
how do i turn on off the my own voice
solid smelt
how do i turn on off the my own voice
mean whenever i start am hearing it like echo
tame oracle
idk what that is, Vonovox doesn't have any kind of voice passthru so that's what i mean when i say idk
im so very sorry
never seen that before
solid smelt
ahh okie
solid smelt
idk what that is, Vonovox doesn't have any kind of voice passthru so that's what...
voice model voice is am getting
in echo
tame oracle
i understand that, i just don't know what it causing that as it shouldn't be happening
solid smelt
any alternative
tame oracle
The only other thing would be W okada eithe deitris or Tg-develop, but Vonovox should really b working fine on your system, can't garentee the others wont do the same if you try them
solid smelt
The only other thing would be W okada eithe deitris or Tg-develop, but Vonovox s...
its working but thing is echo inapp
like when we testing audiio in discord
tame oracle
weird, so when you use it not on Discord its clear?
tame oracle
i mean inapp sound
in app sound? Im sorry im not understanding something here, might be cuz its past midnight
ur saying the actual voice changer?
solid smelt
ur saying the actual voice changer?
yes , inappp voice
in wokada these is an option turn it off
tame oracle
? The way you're saying that is like a order, a command, im sorry im tired, can you elaborate more or worst case you may have to wait for another helper
solid smelt
? The way you're saying that is like a order, a command, im sorry im tired, can ...
when i talk i hear it from inapp . like an acho
tame oracle
when i talk i hear it from inapp . like an acho
well im usless for that then cuz idk, im very sorry
solid smelt
lol diff id
solid smelt
hallow thistle
<@1416622683163983902> dahh
!howtoask
hallow thistle
Make sure to read help guidelines before start asking anything back, but if you see this as a waste of time, it might be your issue rather than the program itself.
hallow thistle
? The way you're saying that is like a order, a command, im sorry im tired, can ...
Let's make thing look simple. Because words alone can't make one to imagine as image, if you send some visual examples like what I do, like the screenshots of Vonovox UI, that would make things much easier to communicate. Otherwise, you can expect a very long conversation that has little progress to go here. 
hallow thistle
when i talk i hear it from inapp . like an acho
What are your settings on Vonovox? Because Vonovox looks like this.
jade hedge
can someone help me with why my real time voice changer sounds like its having voice cracks all the time
hallow thistle
can someone help me with why my real time voice changer sounds like its having v...
Make sure to read help guidelines before start asking anything about W-Okada realtime voice changer.
jade hedge
nvidia Geforce rtx 4060 ti, windows 11, when i speak it just sounds like im having a ton of voice cracks on real time voice changer
hallow thistle
nvidia Geforce rtx 4060 ti, windows 11, when i speak it just sounds like im ha...
Did you follow any tutorial or guide before this?
supple patio
Rtx 5070, Ryzen 7 7800x3d, the voice changer works in the cpu mode and doesnt with my gpu.Thisis the error code
: NVIDIA GeForce RTX 5070 with CUDA capability sm_120 is not compatible with the current PyTorch installation.
The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_61 sm_70 sm_75 sm_80 sm_86 sm_90 compute_37.If you want to use the NVIDIA GeForce RTX 5070 GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/
jade hedge
no i just got the link from a friend and copied his settings
hallow thistle
no i just got the link from a friend and copied his settings
b2332, b2377 or v.1.5.3.18a? I'm more to be specific.
jade hedge
v.1.5.3.18a
hallow thistle
Rtx 5070, Ryzen 7 7800x3d, the voice changer works in the cpu mode and doesnt wi...
The best working W-Okada voice changer doesn't need you to make a bit complex tweak in the program by yourself. Rather, try Vonovox, as this one is also made for GeForce RTX 50 series GPU. https://github.com/dr87/Vonovox/releases/tag/v1.6.9 https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
supple patio
The best working W-Okada voice changer doesn't need you to make a bit complex tw...
Okay perfect! tysm
hallow thistle
v.1.5.3.18a
That one is the original version of W-Okada and is outdated. Try Tg Develop's W-Okada (b2397). https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/ https://github.com/tg-develop/voice-changer/releases/tag/b2397 The b2397 is now marked as a latest version on GitHub, while b2377 is a bit behind.
hallow thistle
Okay perfect! tysm
Let me know if you have any issue using Vonovox, a complete alternative to W-Okada.
jade hedge
Okay thank you ill try it and see if it works
supple patio
Let me know if you have any issue using Vonovox, a complete alternative to W-Oka...
Will do tysm! 
jade hedge
the files that is says to download for nvidia arent unzippable t
knotty moth
the files that is says to download for nvidia arent unzippable t
use 7-zip or winrar, not the built-in windows one
hallow thistle
the files that is says to download for nvidia arent unzippable t
You have to download two files, as the zip file split into two pieces.
And make sure to use 7-Zip or WinRAR instead of the built-in archive extractor.
jade hedge
it sent me to rvc voice changer website is that supposed to happenn http://127.0.0.1:18888/
hallow thistle
it sent me to rvc voice changer website is that supposed to happenn http://127.0...
It's completely normal for this specific W-Okada version.
jade hedge
oh okay perfect. how do i make it so i can hear myself on this version?
jade hedge
Oh tysm, is it normal for it to sounds almost like its very raspy is that normal
solid smelt
What are your settings on Vonovox? Because Vonovox looks like this.
it looks exactly like this
everything was on default
no changes made,and fixed it now
hallow thistle
Oh tysm, is it normal for it to sounds almost like its very raspy is that normal
Here's some more settings if you haven't yet:
Chunk: around 60 - 90 ms
Extra: 2.7 s
GPU: NVIDIA GeForce RTX 4060 Ti
Pitch extraction (F0): rmvpe
barren acorn
Here's some more settings if you haven't yet:
Chunk: around 60 - 90 ms
Extra: 2....
what voice changer should i get for my gtx1660
hallow thistle
what voice changer should i get for my gtx1660
https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/ https://github.com/tg-develop/voice-changer/releases/tag/b2397
barren acorn
https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/ htt...
witch one?
hallow thistle
witch one?
The highlighted ones. https://cdn.discordapp.com/attachments/1159290139609137264/1441738090560491631/image.png?ex=6922e2ba&is=6921913a&hm=c4a73bdb84e710baf7ea70686fda1c2f15cc827991da60d658586f46f7676a31&
barren acorn
The highlighted ones. https://cdn.discordapp.com/attachments/1159290139609137264...
both?
barren acorn
Yes, two of them.
okiiiii tym!
hallow thistle
Both zip files have to be in the same folder, and use WinRAR or 7-Zip to open or extract the ".zip.001" one.
barren acorn
Both zip files have to be in the same folder, and use WinRAR or 7-Zip to open or...
its not letting me extract
it let me extract now
weird
@hallow thistle what next
hallow thistle
<@561807329977434122> what next
Chunk: around 160 - 180 ms
Extra: 2.7 s
GPU: NVIDIA GeF0rce GTX 1660
Pitch extraction (F0): rmvpe (not rmvpe_onnx)
Input: microphone
Output: Line 1 (Virtual Audio Cable)
Monitor: optional; you can set this to your speakers/headphones to hear the program.
If you have set the audio processing as "server" mode, selecting audio driver as "WASAPI" is more preferred than "MME". Otherwise, switch back to "client" mode for much simpler settings and ability to select noise reduction features.
hallow thistle
In "server" mode, if you set it to WASAPI, the sample rate on W-Okada has to be 48000 Hz, and your speakers/headphones also has to be 48000 Hz too.
barren acorn
In "server" mode, if you set it to WASAPI, the sample rate on W-Okada has to be ...
problem still
barren acorn
that one
hallow thistle
If using W-Okada on Opera/Opera GX has some issues, try another web browser.
Always make sure to send the full screenshot of your W-Okada, because when you send simply just error messages, I won't be able to see your current settings.
barren acorn
Always make sure to send the full screenshot of your W-Okada, because when you s...
on google too
hallow thistle
When I said "48000 Hz", you still have sample rate being set to "44100 Hz" on W-Okada. Click red "stop server" button, set them and try again.
hallow thistle
It's how WASAPI's "shared mode" works, typically found in many audio-related softwares. You can't mix up different sample rates in the same audio system. The another mode in WASAPI and also ASIO, "exclusive mode" can be of any sample rate number, although it will mute other program audio and leaving W-Okada the only program to output audio, which I don't think it would be ideal for use on Discord or another program where they all use the same audio system.
barren acorn
It's how WASAPI's "shared mode" works, typically found in many audio-related sof...
i want it for discord tho
so what should i put there?
hallow thistle
so what should i put there?
Sounds good now. Always check perf number in the performance stats to make sure it stays green.
patent rapids
I don't understand how to download and install this
hallow thistle
I don't understand how to download and install this
!howtoask
hallow thistle
Make sure to read help guidelines before start asking anything instead of a small rant.
To know about your PC GPU name, open Task Manager, go to Performance tab, spot GPU 0 or GPU 1 and click either one to reveal its full name on the right side.
sand flare
@hallow thistle hiii
the voice output from vonovox is stuttering
it was working fine before
hallow thistle
<@561807329977434122> hiii
You wouldn't always ask a simple question and then going right away like this. For a general question about Vonovox or W-Okada, you can simply explain about your issue and it doesn't have to mention/ping me like if I'm only a helper you think you only familar with. 
hallow thistle
That doesn't mean anything. As what other helper said to you earlier, you can use @ helper role ping if necessary. 
!howtoask
sand flare
arent u a helper
hallow thistle
Make sure to read help quidelines before start asking anything. 
sand flare
ok
ive read it
rtx 4070 super
windows 11
vonovox voice output stuttering
no error message
<@&1159293204038955078>
hallow thistle
rtx 4070 super
windows 11
vonovox voice output stuttering
no error message
You can send your Vonovox screenshot to here now.
I'm more familar with W-Okada, so anything about Vonovox I'd have to rely on the guide as a base knowledge.
sand flare
sand flare
I'm more familar with W-Okada, so anything about Vonovox I'd have to rely on the...
i have that one too
should i use that to see if the problem persists?
hallow thistle
For W-Okada, I'd suggest Tg Develop W-Okada fork over the "original" one from tutorial from YouTube.
sand flare
low shard
are you running a game too meanwhile using vonovox?
sand flare
are you running a game too meanwhile using vonovox?
it happens regardless
of whether i run a game or no
it used to run perfect before
even when i ran a game with it
low shard
it happens regardless
do you get like delay or does it have issues with specific voice models?
sand flare
all models have the same type of stuttering
there is an update tho
it prompted me to update
low shard
there is an update tho
ofcourse update vonovox to the latest version
windows too
and ofcourse the gpu drivers too
sand flare
okay
sand flare
after that set extra time to 3.0
its fixed
by setting it to 3
didnt evne have to update
low shard
didnt evne have to update
you deffo should btw
it's better to avoid any bugs and get the best
sand flare
you deffo should btw
Ok
unkempt aspen
whats this
viral mason
whats this
That is wokada tg fork
mossy patrol
Any advice on how to train an RVC voice model? How many epochs should I use for a dataset of 15-40 minutes in general?
viral mason
There is no specific epoch count for any dataset length, usually a model is fine after around 100-200
Just check each saved epoch with a short sample
barren acorn
what to put for a male
and female?
viral mason
what to put for a male
and female?
if you are a male and using a male voice for most of them keep pitch at 0
if u are a female using a male voice model tune the pitch down some
and reverse it for other way around
viral mason
female voices are usually good around 3-10 pitch wise
try going up till it sounds good
barren acorn
try going up till it sounds good
Formant Shift?
viral mason
nope, pitch
barren acorn
nope, pitch
i meant what about it
viral mason
nope just changing the pitch will be fine
barren acorn
nope just changing the pitch will be fine
but if it sound bad still its voice problem?
viral mason
most likely either the voice model is old and not really made well, or it could be ur mic and or background noise/people talking in your background getting picked up by the mic
tight linden
Can't send pic here so Idk how to describe my issue but on kaggle the applio import keeps loading and is taking way longer than2-3 minutes this is ridiculous idk wth it isn't working for me.....
timber token
-colab
viral mason
tame oracle
@strong nova What is your GPU?
strong nova
@tame oracle Nvidia Geforce Rtx 3070
tame oracle
<@1416622683163983902> Nvidia Geforce Rtx 3070
great that's more than enough for Vonovox
-rt
strong nova
Thank you
tame oracle
No problem
viral mason
how do i upload models to kits.ai
why would u use that awful website :<
fringe remnant
help
cant send picture
https://youtu.be/8Hq4PEZ30oY?si=v6tBH6DyUiCtjKJZ how to do this
nocturne mural
apparently it's a problem with all the Dione apps.
Try installing Applio manually
neon valley
ai voice changer for intel?
hallow thistle
ai voice changer for intel?
Intel Arc A/B, Intel Arc Graphics or Intel UHD Graphics? These are Intel GPUs, with the first one being a dedicated GPU while the others are integrated.
neon valley
Intel Arc A/B, Intel Arc Graphics or Intel UHD Graphics? These are Intel GPUs, w...
not quite sure but i think its an integrated
Processor Intel(R) Core(TM) i5-3210M CPU @ 2.50GHz 2.50 GHz
Graphics Card Intel(R) HD Graphics 4000 (32 MB)
hallow thistle
That laptop must be hella old, as your laptop Intel Core CPU is third generation, but a bit step up from my laptop which has second generation Intel Core i3. 
vapid gust
burnt zealot
does any1 konw why my voice changer not working
just ranomly sotpped 5 minutes ago
iahave everything set perfectly
it just doesnt talk or anything
hallow thistle
does any1 konw why my voice changer not working
There have been newer version and an alternative to this one. Tg Develop's W-Okada fork (b2397) and Vonovox, which is a complete alternative to W-Okada, are known to work with NVIDIA GeForce RTX 50 series GPU.
Vonovox can give better audio quality although its UI can be less familar than other W-Okadas. https://github.com/dr87/Vonovox/releases/tag/v1.6.9 https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
Or opt for Tg Develop W-Okada for more familar W-Okada interfaces. https://github.com/tg-develop/voice-changer/releases/tag/b2397 https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/
VB-Cable can be very confusing to set it up to work, where you would encounter audio not coming out from the program, while Virtual Audio Cable lite is much simpler to set up, as it as only one virtual line to use. https://software.muzychenko.net/freeware/vac470lite.zip
manic vector
does anyone know why mine voice changer is cuttting out everytime i talked
feral crater
Are AMD GPU’s supported for training?
burnt zealot
There have been newer version and an alternative to this one. Tg Develop's W-Oka...
so vonovox is better than okada? and also does the same models work with vonovox
as if i were to be on okada
hallow thistle
so vonovox is better than okada? and also does the same models work with vonovox
Both Vonovox and all W-Okada versions support "RVC" voice models. While I haven't seen any benchmark between Vonovox and Tg Develop W-Okada fork on the exact system, many people prefer Vonovox for audio quality, and that's what I know.
burnt zealot
Both Vonovox and all W-Okada versions support "RVC" voice models. While I haven'...
can i just drag and drop the model folder from rvc to vonovox or do i have to add it 1 by 1 manually
hallow thistle
Are AMD GPU’s supported for training?
It is possible that AMD Radeon RX GPU can be used to train RVC voice model, especially with Applio RVC fork, and Stable Diffusion, but only specific versions from both programs would work, as many main programs that made for NVIDIA GPU would often perform poorly or not work at all with AMD GPU.
hallow thistle
can i just drag and drop the model folder from rvc to vonovox or do i have to ad...
RVC doesn't always mean realtime voice changer; RVC would be more of "Retrieval-based Voice Conversion". Import by dragging the RVC model either from MMVCServerSIO's model folder or directly onto Vonovox program don't sound like something you could possibly do with either Vonovox or W-Okada.
burnt zealot
RVC doesn't always mean realtime voice changer; RVC would be more of "Retrieval-...
so if i already have a ai model on rvc i cant use that model with vonovox?
hallow thistle
so if i already have a ai model on rvc i cant use that model with vonovox?
I'm more of focus on settings, but when you ask too many questions with little or no progress to go, there's the guide doc for Vonovox. https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
burnt zealot
I'm more of focus on settings, but when you ask too many questions with little o...
thats why im asking bc im confused it sasy add .pth but rvbc okada is using .safetensors so i dont see a rvc model when i try to add it
hallow thistle
You can't always make things more complicated than necessary. Either Deiteris or Tg Develop W-Okada would convert pth into safetensors file by default, which is completely normal for these specific versions. To upload a model into Vonovox, you simply upload the original "pth" file rather than the one that already in W-Okada.
burnt zealot
okay i see i got it
sucks i lost most of my original uploaded models but i have my most recent
hallow thistle
For voice models, check #1175430844685484042.
burnt zealot
For voice models, check <#1175430844685484042>.
nah i make my own bro i usually just delete them
i only care for my recent one so its fine
thx though im ab to test this is and i think from what im hearing already the quality is better
i do feel like rvc fake laughing
is better
could just be the model but idk
woven pumice
Hello everyone! I’ve been using the Hina Modified Realtime Voice Changer Client on Colab for a long time, but recently I’ve been having a problem where every model gets stuck at 96–99% during loading, even when using GPU and even with models that used to work perfectly.
I deleted my old notebook and re-uploaded it, but the issue still happens.
Could someone please share the most up-to-date and stable version of the Realtime Voice Changer Client, or let me know if there’s a newer working Colab version?
I would really appreciate any help. Thank you!
hasty barn
Can't I sing with my own voice? Can I only sing certain sounds?
hallow thistle
Hello everyone! I’ve been using the Hina Modified Realtime Voice Changer Client ...
If you have Google Colab Pro tier, you can run this voice changer notebook; the free tier otherwise can cause issues running either Deiteris or Tg Develop W-Okada fork on Colab. https://colab.research.google.com/github/tg-develop/voice-changer/blob/master-custom/Colab_RealtimeVoiceChanger.ipynb https://docs.aihub.gg/realtime-voice-changer/cloud/tg-develops-w-okada-fork-colab/
hallow thistle
Can't I sing with my own voice? Can I only sing certain sounds?
Make sure to read help guidelines before start asking anything about RVC or W-Okada.
!howtoask
hasty barn
Make sure to read help guidelines before start asking anything about RVC or W-Ok...
I couldn't do it again. My problem is that I have an audio file that I want to model with RVC GUI, but the ready-made audio files have pth and index extensions, and since my audio file doesn't have them, I can't add them. How do I add them?
hallow thistle
I couldn't do it again. My problem is that I have an audio file that I want to m...
What is your PC GPU? And which RVC fork are you trying to run? Because when you say " I couldn't do it again", it can either mean you didn't read the help guidelines how to explain about your issue.
Otherwise, it's hard to identify the issue and a potential solution if any.
hasty barn
RTX 5060TI 16gm vram
hallow thistle
Simply as that. The original and mainline RVC GUI, while stable for a few scenarios, wouldn't work with NVIDIA GeForce RTX 50 series GPU. Applio RVC fork, a superior RVC to the mainline, can work with RTX 50. Download Applio from Hugging Face. https://huggingface.co/IAHispano/Applio/tree/main/Compiled/Windows https://docs.aihub.gg/rvc/local/applio
-rvc
hallow thistle
Attempting to run original RVC-GUI with RTX 50 would likely run on your PC CPU instead.
Because your initial query is confusing to understand, so that's what you would expect. If you meant something else, let me know.
woven pumice
If you have Google Colab Pro tier, you can run this voice changer notebook; the ...
I have a question: how long does this last?
I usually used a file from Google Drive that I opened and it configured itself automatically, and I could use the voice for about 4–5 hours. So, approximately how long can I use the website?
hallow thistle
I have a question: how long does this last?
I usually used a file from Google D...
Usually 4 hours a day for "free users". The question is, again, "how can you run W-Okada notebook on Colab with free tier"?
woven pumice
Usually 4 hours a day for "free users". The question is, again, "how can you run...
I’m just running the notebook normally on Colab Free. I open the link, connect to the runtime, and it works for me. I’m not using anything special.
hallow thistle
I’m just running the notebook normally on Colab Free. I open the link, connect t...
Beware of any random disconnection while running W-Okada on Colab with free tier, as some people here have got their Google accounts terminated from using the service on Colab. For the more runtime and fail proof against random disconnection, Google Colab Pro or "100, 500 compute units" is more preferred than the free one. Kaggle is another free option that works similar to Google Colab, provides 30 hours a week with "T4 x 2" enabled within a notebook.
signal crater
guys i installed deiteris-w-okada-fork for linunx as per the guide
i even downloaded port audio as well
my question is
which mic do i use there in linux
wary badge
i tryed to run the https and then it didnt do anything and then i just tried to open the voice changer it self it opened then didnt show anything
lavish isle
Is there a Colab where I can create a model from scratch?
signal crater
its stuck like this, what do i do?
hallow thistle
Is there a Colab where I can create a model from scratch?
-colab
signal crater
its stuck like this, what do i do?
help me @hallow thistle
hallow thistle
help me <@561807329977434122>
There has been a newer version of Deiteris W-Okada but developed by a different author. Tg Develop W-Okada (b2397) has more recent features than the Deiteris (b2397) one. I'm more familar with Microsoft Windows, so anything about Linux distro would have to rely on guide as a base knowledge.
signal crater
@hallow thistle which one i choose here
hallow thistle
If you know how to download files or use a command in Linux, download these three zip files from GitHub. https://github.com/tg-develop/voice-changer/releases/tag/b2397 Use WinRAR, 7-Zip, or zip and unzip commands in terminal to combine and extract the ".tar.gz.aa" one and all.
hallow thistle
<@561807329977434122> which one i choose here
Did you not pay attention to my statement?
signal crater
If you know how to download files or use a command in Linux, download these thre...
i did all these steps bro
my question is which mic i choose
hallow thistle
my question is which mic i choose
More like "HD-Audio Generic: ALC257 Analog (hw:2,0)". Because that one says "analog", so that one must be the microphone you plugged into your PC. The Windows equivalent is something like "Microphone (Realtek High Definition Audio)" which is named to be more friendly.
signal crater
what about virtual audio cable
currently im using my headphone blackwire
but for output
hallow thistle
hallow thistle
I'm not sure about the "virtual audio line" in Linux, as I don't see "portaudio" or something related in your screenshot.
signal crater
so should it show port audio in voice changeR?
lavish isle
@hallow thistle Can you check your DM ?
hallow thistle
@lavish isle Instead of hopping into my DM. For the best working RVC Colab notebook to train a voice model, there's Applio RVC. https://colab.research.google.com/github/iahispano/applio/blob/master/assets/Applio.ipynb https://docs.aihub.gg/rvc/cloud/applio-colab/ The no-UI one is also possible. https://colab.research.google.com/github/IAHispano/Applio/blob/main/assets/Applio_NoUI.ipynb
signal crater
anyone who helps me fix me my virtual audio cable in linux? i can pay for the help
hallow thistle
so should it show port audio in voice changeR?
Whatever you done in W-Okada, as you have only said you have followed all step froms the guide, and that's all. Rather, I think you might need to configure the PulseAudio or PipeWire for use with PortAudio, as what these softwares hinted as the potential solution that could work similar to Virtual Audio Cable or VB-Cable in Windows system.
mint plank
when i change on discord to line 1 nobody hears me why??
orchid shore
i need help
voice chnager wont even open
when i try to start it up it opens the command prompt like usual but tehn it says a bunch of stuff and closes itself and doesnt even start the vc
orchid shore
@hallow thistle
lime smelt
So, I've been playing Games with 3090, a Primary card for Games and Using RVC(Vonovox) on 2070, a secondary card for Voice Changer.
I've noticed that RVC, despite running on a Separate GPU in a Same PC rig is experiencing a Performance issue and the sound become choppy when Game's Heavy load on GPU or CPU become apparent, despite the game running on 3090 and 3090 ONLY.
Can you help me with this? Thanks.
mint plank
why does it say trial in a dc call
the other person hears trial every 5-10 secs with the ai voice
knotty moth
why does it say trial in a dc call
either pay for it or uninstall it then install vac lite
mint plank
oooh it says trial cuz its not the lite one?
viral mason
the other person hears trial every 5-10 secs with the ai voice
You have the most outdated voice changer ever
Unless that's not the issue and I'm stupid
orchid shore
can u guys answer my question too while ur here 😛
orchid shore
when i try to start it up it opens the command prompt like usual but tehn it say...
^
viral mason
when i try to start it up it opens the command prompt like usual but tehn it say...
What gpu do u have
And did u get the voice changer from a YouTube tutorial
Btw if u have Intel you're cooked
orchid shore
What gpu do u have
nvidia
mint plank
how do i delete the previous one again 😅
orchid shore
ive been using it since july and its been working