#✨│ai-help
1 messages · Page 255 of 1
really? last time it was its own app. Just a new concept but ill get used to it
also not recommended to directly transfer loaded models in the old version
as it'd break the compatibility
gotcha ill test some things out. appreciate your help!
Is there a recommended formant shift setting? male to female btw
and should i put protect at 0.33 or keep it at 0.5?
practically it wouldn't fix artifacting on mediocre models
-realtime
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested. GUIDE
ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are way worse. GUIDE
Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated
yes yes
so
you know how
at the top
at the link for delteris fork
when i click that what do i do
i literally never use github im so confused
you dont have to click that
ohhh
read it carefully, dont just click the first link and skip steps, if you do something wrong you can fuck up your whole audio system and not hear anything on your pc
ok so i installed virtual audio cable extracted it and opened setup64
and my audio
is
gone
yep, that's exactly what I was talking about
read all the steps dont skip them
you skippped this step
not related but nick you should participate in the rvc event
it would be fun
(not me trying to get as many people to join as possible so more of these events happen)
it's exactly while opening the setup64.exe, because you will get a popup mid installation where you have to click yes
ok ok wait
im gonna
change it back
ok i have my audio back
do i do it again
and just
do that
what u said
mmm not sure I don't train models since a year
wdym?
you can always get back into it
if you already closed the setup64.exe popup, already installed, did you click yes or no?
what should i do from here
only real requirement is no drum/noise models 
if you already installed it, do the WIN+R method, no need to re-install something you already installed
Me training the duck from REPO
damn too bad was gonna send a Jupiter model
lool
ok well i did that
i dont have a mac
and
i already
have
nvidia
what's your pc gpu again?
which?
4070
download the normal nvidia on windows version
not the 5000 serie one
then after u click that do this
if you downloaded it already, yeah
yes i downloaded it
also tysm for helping me i genuinly appreciate u
okay i extracted it
now go inside the folder, and click MMVCServerSIO.exe
k i have like another tab
and its downloading something
it opened it on browser
is there a way i can get a application
like
my mic chops up a bit when using an ai voice, but it works perfectly fine normally, any fixes? or is that just something that happens
after downloading it the voice changer for w-okada is not working, i am on mac intel , i got "Pytorch is damaged", this message and when i went to the terminal to type the remaining things told in the guide it said not file names that is located
can someone help
it is an application, it just uses a web user interface https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#why-does-it-run-in-a-browser-and-not-its-own-window
Last update: May 5, 2025
show a screenshot
This is a General AI Server, we won't be focused on voices anymore
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
This is a General AI Server, we won't be focused on voices anymore
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
where do i get the help then
it opnes this on google
and after a couple minute it just stops working
i mean the website i dont even know if the
actual purpose works
yep its normal, click start, be sure to not close the CMD
here, we do also voice, just that we do many programs, so you have to specify more
no you shouldn't, it's what its actually running the program
like i was downloading teh voice mod and its not working not opening
gpu: rtx 4070
f0: rmvpe without onnx
chunk: 80ms
extra: 2.7
input: microphone
output: line 1
monitor: optional to hear urself, headphones
on wokada deiteris fork, you can **optionally **use more advanced settings for benefits:
- Force FP32 mode: on (THIS IS OFF BY DEFAULT! Turning this on improves stability. Increases VRAM usage by 200 MB)
- Disable JIT compilation: off for faster loading speed of the program, on for slightly better performance (10-15 ms) for Nvidia only)
- Reduce the delay via: https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#reduce-more-delay
Last update: May 5, 2025
there's many different programs you could mean by that, elaborate all i asked
!give-media-perms 1h @short spire
liek its not workign t this part
i got teh message for pytorch is damaged
but teh command is not workign in teh terminal
what part? send the tutorial link
that doesn't seem to be our ai hub documentation
okay i put all of this
also, what mac m chip do u have?
intel
also be aware that mac isn't the best for AI
ok
this is where i found it
oh that needs to be updated..
the updated one is https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork
Last update: May 5, 2025
gonna update that post later
ok
#✨│ai-help message do you wanna try the optional benefits?
what r they
click the link, they are optional settings that can give you benefits like less delay or better quality
its optional tho
does it help alot
depends on which one you do, if you do the 3rd step with asio, yes it can help reduce much the delay, but its more complex
your choice
TL;DR: yes they all help, depends if you want to take 5 mins to read them or are too lazy
can someone help me fix this issue
still gives me this error
oh
well how do you like use the voice changer
add a model, selct it, click start, go in another program like discord, set the voice input of the other program to line 1, youre done
ok i clicked on it and its says waiting for operation to complete
oh it worked
oh the delay is bad
how do i do the other step
close every other program in background, what program are you using it with?
on discord or a game
im probably gonna use it on discord and valorant
how do i fix this error
yeah, go on discord and valorant settings, set the voice input there to line 1, set game graphics to the lowest 1080p with 60fps cap, then show a screenshot of wokada
@pastel oak do you may have any idea?
im not sure if you got a mac, but i dont
sad
does anyone know a good female voice if i have like a deeper voice thats not very feminine
you need some help catfishing lol?
honestly
i just wanna troll on valorant
not really catfishing but i guess that is catfishing
in the sense
but
can u help me
thats honestly real
but yea there is no good way of finding a good voice
you just need to look around find one and try it
you can use the egirl tag in #1175430844685484042
but ye
I dont know much about mac either but just looks like the command is wrong (maybe its mac specific), id suggest asking chatgpt about this step @short spire
ok
does anyone know what this is unhandledrejection error
how do i make it sound smoother
try this
idk what u mean btw with smoother
well are you putting effort in when you speak? speaking like a robot will have it sound robotic :p
but could also be the model
im using E-woman
try lowering the chunk setting until it sounds less choppy and more stable
do i have to like really try
try turning up the pitch if you're a male
@queen valley show a screenshot of ur current wokada
you gotta be slow with them
the ai voice changer
RVC = Retrieval-based-Voice-Conversion, the best Few Shots Speech To Speech AI Models (on v2), Inferences (use models) pre-recorded audio (ai covers) and train (make) models. Technically, Mainline RVC does have a go-realtime.bat (aka RVC-GUI), but it's pretty messy and outdated so it's extremely not suggested for realtime.
Wokada = uses RVC for realtime inference. There's 2 main versions, Original made by Wok, and the most suggested one is Deiteris Fork (modified version)
you're using wokada deiteris fork
me
yes it is
try turning down chunk size
move it to the left a little
<
that way
btw when u click on it u can use arrow keys to move it more precisely
np
when I try to open the okada voice changer, it opens a window but then immediately closes. Anyone know why? I'm using windows 11 and a 3060 nvidia gpu
Elaborate:
- what you want to do
- what tutorial link are you using
- a screenshot of the program
I'm trying to use it for discord, i'm using this video https://www.youtube.com/watch?v=SxdnGxicJOg&t=108s, and the program wont load
i'm clicking the start_http.bat file
i vote we mute anyone linking this video for a week
yeah but wouldnt that just make it worse
probably not
is the video bad?
The voice cable doesn’t work how do I fix that?
it is outdated
ah alright
thats weird cause i sorted by newest in the comment section and they all said it worked
or like 95% said it worked
all of them from today
what tutorial should i use then?
-rt
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested. GUIDE
ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are way worse. GUIDE
Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated
that uses an over year old version of original wokada, and vb audio cable causes issues on windows
every yt tut is outdated
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
people off youtube tutorials dont have up to date info lmao, that has way worse quality and performance
use the wokada deiteris fork tutorial
-realtime
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested. GUIDE
ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are way worse. GUIDE
Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated
1st link
yw and lmk
what browser are u using? what did u do?
im using brave and i opened mmvcserversio.exe
guys if i have like 1:30-2h voice train dataset, what should be an ideal epoch size, batch size and other specs?
it instantly says that on startup?
if so, try using chrome or firefox
well it loaded and that popped up
try another browser
wait is it not an actual app?
it is.. it just uses a web user interface https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#why-does-it-run-in-a-browser-and-not-its-own-window
Last update: May 5, 2025
my mic is not getting detected
This is a General AI Server, we won't be focused on voices anymore
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
gtx 1050ti
windows 10
hear my self when using a vc
yt
i can't send it here
I had to uninstall uvr a bit ago bc of storage issues but now that I dll again it’s not letting me open it, It says error loading python dll
send the tutorial link
!give-media-perms 1h @loud pewter
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
What does my pc gpu gotta do with anything, I’m telling you the error which is a white screen saying “error python dll, cannot be found”
I’ve downloaded uvr before
your pc gpu is crucially important, since you need a good one to run AI, also we would need all the info I asked you to know if you're using an up to date tutorial and all the other things
!give-media-perms 1h @brazen ore
yea but I’ve used this same program which is uvr before
I downloaded from this link hold on
please elaborate all I asked so we can help you the best
This is the link I’m using to download uvr
I install it and then when I am about to open up the program after it’s done installing it tells me the error message I told you
could you also elaborate all the other things? we need to check if it's also worth to use it locally in your case
What other things
My gpu is gtx 1650ti
your pc gpu, operative system
Gpu is 1660ti, I’m running windows 10, I’m trying to open uvr, not using any tutorials, and the program I’m trying to open is ultimate vocal remover
This is what the program look like^^
how do i fix vb cable
if i have to make it work i have to set default to vb cable but i cant hear anything
i need a ai voicechanger for amd
thanks for elaborating, could you try uninstalling again from windows app settings, delete afterwards the C:\Users<YOUR WINDOWS USERNAME>\AppData\Local\Programs\Ultimate Vocal Remover folder, then trying to install again?
also, are you sure you wanna do it locally? there's cloud (remote good pc) free options that could be faster
vb audio cable causes issues on windows, uninstall everything you got off youtube
ALL VIDEO TUTORIALS ARE OUTDATED FOR VOICE CHANGERS
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
Elaborate:
- your PC GPU
- your operative system
- what you want to do
i uninstalled vb audio cable, instead downloaded this thing, and
- rtx 5070
- windows 11
- i want to use it for games and discord with the AI RVC voiec changer
- https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#virtual-audio-cable
- here
and its the same issue
as the vb cable
idk why
because it shouldn't be the default device
you skipped this step
also, be sure you got the nvidia rtx 5000 serie of the wokada deiteris fork
yes i do
wont work
yeah you didn't do it correctly, you need to right click your headphones and set it as default communications device
show also the recording tab
did you also right click the microphone and set it as default communications device?
dude
it won't show the default communication thing
if it is default
and that mic is my main
alr, can you show now your wokada deiteris fork and whatever other program you're gonna use the voice changer on voice settings?
ok ill load it
hold on
did you do the wasapi or asio steps?
you're in server mode
The difference between Server and Client Audio in Wokada Deiteris Fork:
- Client: easy to use, can use echo, sup1, sup2 cancellation
- Server: harder to use, can significally reduce delay
i took from this
+CUDA12.8 Pytorch updated
(Pytorch nightly version)
RTX 5080 test done
Windows / NVIDIA
For RTX 5000 users
thanks to deiteris (https://github.com/deiteris/voice-changer)
that's just the rtx 50 serie wokada deiteris fork
do you want to use the server mode or client mode?
yeah u told me to get the 5000 series one
hmm
^
its totally your choice
even tho with the differneces
depends on your needs
hm
i wanna use client
select client then
input: microphone
output: line 1
monitor: optional to headphones to hear urself
on wokada deiteris fork, you can **optionally **use more advanced settings for benefits:
- Force FP32 mode: on (THIS IS OFF BY DEFAULT! Turning this on improves stability. Increases VRAM usage by 200 MB)
- Disable JIT compilation: off for faster loading speed of the program, on for slightly better performance (10-15 ms) for Nvidia only)
- Reduce the delay via: https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#reduce-more-delay (this is more complex and server mode)
Last update: May 5, 2025
what is this error
chrome
nothing i only added a voice model
why is my audio becoming weirdly low
did you try to rename it on the web user interface?
wdym
did you also rename the model, or just added it?
just added it
when i remove the w okada thing it now makes my audio sound normal
why
should i change the name
no you shouldn't, the issue is because you have to select the model slot then click start
you just added it, you need to select it too
i did that
i did bruh
show a full screenshot of ur wokada
windows,
radeon 7800xt
ai voice changer real time
click on E-woman, then click start, does it have any issues?
yes it doesnt repeat a word i said
yaeh
yeah
roleplay
its a larping
lol
its like a roblox family rp
and my role is mother
-realtime
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested. GUIDE
ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are way worse. GUIDE
Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated
1st link, wokada deiteris fork
oh great then, was asking in case it was catfish
😭
you need to set monitor to ur headphones to hear urself
doesnt work
are you gonna use it just for larping? or catfishing?
its kinda weird that u were asking me that question
but im using it for larping
i do lot of family RPs
on roblox
alright then just making sure since catfishing is banned and we have just banned some recently lol
do you have any friend or alt acc to check if it works on roblox/discord? just to check if it works generally speaking?
well i test it on discord voice settings
can you recommend me the best RVC?
did u test it?
RVC = Retrieval-based-Voice-Conversion, the best Few Shots Speech To Speech AI Models (on v2), Inferences (use models) pre-recorded audio (ai covers) and train (make) models. Technically, Mainline RVC does have a go-realtime.bat (aka RVC-GUI), but it's pretty messy and outdated so it's extremely not suggested for realtime.
Wokada = uses RVC for realtime inference. There's 2 main versions, Original made by Wok, and the most suggested one is Deiteris Fork (modified version)
This is a General AI Server, we won't be focused on voices anymore
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
could you try restarting your pc, then retrying the program, and if still not working trying another browser like edge or firefox?
okay
"You have reached your queued job limit
0" im getting this message on weights. did weights lie and not actually return free voice training or is there a bug im running into
for context. i have not trained a model in 3 weeks
how do i make it load in firefox or smth
@low shard same issue on firefox
weird
weights.com gives one free model training weekly, try reporting this issue in their server discord.gg/weights
pasting the url here
can u check the cmd?
wtf never saw this error before
@crude flame did you ever see this before?
my pc must be special
yea never seen that
very special
@analog obsidian @simple ore yall ever seen this error?
nope
im sure no one have seen it
I would really suggest you to try to uninstall everything and reinstall carefully, and restarting ur pc
what if he tries changing the f0 to crepe
try crepe f0
im curious too
alr
never seen something like that before
where do i change it
f0 det
try full
theres no crepe f0
crepe_full
ok
same error?
Hello, I would like to ask for help
I made an AI model but there are errors (the sound cracks/drops out at high notes and there is a lot of noise during the breaks in the music) and out of curiosity I tried what happens if I convert it to an image and compared it with downloaded models
and it appeared at the beginning and at the end in my case there is something that sounds like noise, while at the beginning there is complete silence in the original file
the downloaded ones don't have it
and I would like to know how to make it disappear
or anything that could be better
(if everything is true the last 2 pictures are mine)
is this the 50xx series version?
yes
show the UI
what UI
i think it's using the cpu somehow
don't use crepe in voice changer, it'll be too slow
input: your mic
hey guys I rly need help, if I have a weights acc with a decently big audience already that's linked to my google account how can I link that to my discord account to get model maker benefits?
@low shard lol the issue was i had to increase my chunk
change to server mode
😭
ok
change in and out to WASAPI mic / line 1
click initialize
ok
dosent work
but i got new issue now
the virtual cable line 1 thing isnt showing any audio
and it has lot of delay
imo vb cable audio is better
lol
does anyine have a new way to make ai covers cause the one i usually use is not good
i have officially fixed it, thanks guys @crude flame @analog obsidian @low shard @simple ore
😭
that was stressing
hey how do i fix it when my voicemod is not reading the cable
um this is not a voicemod server
its a AI discord server
ik but it should work if i put the cable in voicemod
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
dos anyone have anew way to make ai covers. The ai gen hub ui doesnt work anymore
voicemod isn't even ai, and dont use youtube tuts they are all old
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
idk i was using ai gen hub ui and that process led to gradio and now the colab links dont work
elaborate all i asked
it's 3 am lmao
either elaborate what I asked or hope there's an helper that can help you in vc
oh its 8 for me
you should try the better one instead of that sponsored stuff
Last update: May 5, 2025
how to use RVC for free? im brazilian, idk english
what's the best way for me to make models? my GPU is an Intel(R) HD Graphics 520
heyy guys im trying a voice changer with kasane teto to troll with my friends but for some reason my voice is like VERY VERY delayed. do you know whats that about?
anyone got a tutorial on how to install/how the ai stuff works?
Do you guys use Index? I've heard from the guide (the docs) that it make's the voice higher quality but I'm not sure if it's too different
it affects the accent precisely
is there any ai thing out that can let you upload audio files & it makes songs similar to that audio?
can someone suggest me a good lora train model - im wondering if i should use fluxgym/diffusion-pipe/flux
yo
hi guys, know this isnt the super nerdy ai place but does anybody know how semantic chunking for llamaindex indexing is priced? i couldn't find any mention of it on the pricing page and can't use it since its grayed out
guys i need some i idea for a project on fraud detection of finger prints or iris scan with gen ai
to train a good singing ai model, how much falsetto should there be
my voice can't glide from normal voice to falsetto, it's rather a large jump between notes so like should i include any falsetto at all or none
to make a good model you need to include the whole vocal range right, but since it's a large jump and there are notes missing, how would that affect the output quality?
how do you actually continue training on applio ?
how i can make ai covers now whiout weights because there is 3 minute limits?
This is a General AI Server, AI has many fields
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
This is a General AI Server, AI has many fields
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
This is a General AI Server, AI has many fields
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
THERES THOUSANDS OF AI PROGRAMS, WE DONT KNOW WHICH YOURE TALKING ABOUT
RVC = Retrieval-based-Voice-Conversion, the best Few Shots Speech To Speech AI Models (on v2), Inferences (use models) pre-recorded audio (ai covers) and train (make) models. Technically, Mainline RVC does have a go-realtime.bat (aka RVC-GUI), but it's pretty messy and outdated so it's extremely not suggested for realtime.
Wokada = uses RVC for realtime inference. There's 2 main versions, Original made by Wok, and the most suggested one is Deiteris Fork (modified version)
This is a General AI Server, AI has many fields
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
That is integrated graphics, it's shitty and wouldn't even be recognized for local training
Aren't you a model maker already? What did you use before? Do you.maybe have another GPU?
This is a General AI Server, AI has many fields
Elaborate:
- your PC GPU
- your operative system
- what you want to do
- what tutorial link are you using
- a screenshot of the program
That is an over year old version of original wokada, and vb audio cable causes issues on windows
uninstall everything, all YouTube tuts are outdated
This is a General AI Server, AI has many fields
Elaborate:
- your operative system
- what you want to do
Maybe try weights.com or udio
ohh thank you but dont worry, for some reason if i wait a bit after it loads up without cilciking begin or start, it just works but after some time, it crashes
Share a screenshot of your wokada
Bro don't use that program, it's outdated asf
Don't even bother using it
what do i use then 💔
It's like using windows XP in 2050
Not video tutorials for voice changers for sure, if you ever want one just check this server, because there's no updated one and will prob never be one, ai moves at Sonic speed
-realtime
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested. GUIDE
ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are way worse. GUIDE
Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated
1st link, wokada deiteris fork, read it carefully
Uninstall everything you got, it was just wasted time with ancient info
@pure nest Let me know
Not sure about it, but maybe @simple ore got some ideas about that
Also this is trying a more general ai server now, just we would need more staff that isn't voice focused
You will have to delete the original wokada folder totally, then uninstall vb audio cable from the windows app settings
The 3 minutes limit has been increased to 5 minutes for free recently
Hey would anyone help me build a simple boxing app?
I am new to programming
had a course in C# last semester tho
hey guys
i need help
im super confused with the ui
as its nothing like this videohttps://www.youtube.com/watch?v=SxdnGxicJOg
so ihvae 0 clue
can someone guide me through this?
ah, i was confused why everything was about voice changers though
in the meantime are there any other ai dev servers that are well established? most ones i find are client facing product servers and not general dev help so i usually can't find help 
all video tutorials are outdated
can i g
oh
No need to ping me lol
Yeah that tutorial uses an over year old original wokada
how do i do stuff then
It's outdated asf
I also wrote it in the comments
this one
The performance between the one used in the video and that one is the same
RVC = Retrieval-based-Voice-Conversion, the best Few Shots Speech To Speech AI Models (on v2), Inferences (use models) pre-recorded audio (ai covers) and train (make) models. Technically, Mainline RVC does have a go-realtime.bat (aka RVC-GUI), but it's pretty messy and outdated so it's extremely not suggested for realtime.
Wokada = uses RVC for realtime inference. There's 2 main versions, Original made by Wok, and the most suggested one is Deiteris Fork (modified version)
what do i do
Read it up, it isn't just called voice changer
Voice changer is a wayyy too broad term
yeah i know
There's thousands of programs that could be named like that
can someone please share a workflow for flux lora training that works
can you tell me your PC GPU and operative system?
AMD radeon TM graphcis
and WIN11
That's not good at all, it's integrated graphics, low low power
Do you have any other GPUs?
yikes
nop
this is all the stuff i have
https://x.com/cocktailpeanut/status/1832084951115972653 you may wanna check this for flux gym, tho I never really made a flux Lora and can't help much
mid tier laptop
the fluxgym is broken bec of deleted huggingface
That isn't mid tier at all, you have no dedicated GPU, only integrated one lol
oh
so your telling me i got scammed 💔
i spent like 65k my currency on this laptop
which is like 500-600 dollars
somewhere in that range
You got 3 options:
- Buy a better pc
- Run it locally (on ur pc) using the CPU mode of the wokada fork which has better performance https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/ (but this isn't suggested as it could be unstable)
- Use **cloud **(remote good pc):
About Cloud, there are different services:
- Google Colabs (4 hours daily of free T4 gpu, easy to use, require only a google account) :
- W-Okada's Deiteris' Fork Voice Changer Google Colab (currently works only on google colab PAID tier)
- How to use Original W-Okada's Voice Changer Google Colab (has a Guide) (currently broken)
- Kaggles (30 hours weekly of better GPUs, T4x2 & P100, harder to use, requires an account and a phone number):
- W-Okada's Deiteris' Fork Voice Changer Kaggle (the best and only working one currently for free)
- Original W-Okada's Voice Changer Kaggle (currently broken)
The ram and CPU are nice, but you got no dedicated GPU, which is the most important component for any complex intensive task like gaming, ai and 3d modelling
oh also, when wokada says your CPU can be damaged if you use CPU-only voice conversion what kind of damage does ts mean
also does it perform differently if i use a non-gaming GPU
Oh ig thats why blender lags for me so much
Ai is complex and intensive, it needs good hardware
i see
can i use a cloud pc?
do you use instancing on blender lmao
and how would that work and stuff
i wasnever really able to learn it so
colab and kaggle, scroll up
Might be, without a decent dedicated GPU you can't do much, the CPU is nice but that's meant for running the OS and task that don't require complex computing
100% learn it it reduces workload by huge amounts with repetitive models
how do i get started.and how does cloud pcs even work?
cloud pcs are like someone else with a big fat pc doing the computation for you and just sending you the results
if you scroll up the tutorials for colabs and kaggle were posted
For that price range it's normal, it's a nice laptop, dedicated GPUs are expensive and can cost up from 300 ninimum to 2k+ dollars
Also of course laptops aren't as powerful as desktops
ok but like how would i use it for my lapgtop and stuff
like all my data
Yeah, it runs on a remote good PC with limited free time
You just need to read the wokada deiteris fork Kaggle guide
can u give it here sob
It would be better to make an issue on their GitHub
and wha do i do about the thign i downlaoded
#✨│ai-help message I did already
You can uninstall everything and read only the Kaggle guide
i saw theyve released a flux lora training workflow and will try that rn
what did you download (ngrok? you need the key from that anyway so dont delete it), you can delete the rest
also chat 
please answer
why everytime i open the voice changer file start_http it doesnt open cmd?
]
you can destroy it all
i just made a acc
just make sure you keep ngrok and chocolatey
then you didnt install ngrok
unless you used a different method
so i delete everything?
or did you use the other thing h something
more or less
k'
i pray for a response
@stoic emberits not letting me delete the zip file
huh? show the error box
if it says task is running or sm shi go to task manager and kill it
nvm]
also you cant preview the zip while deleting it -_-
i run 4 vms with my puny iris xe
It could damage the CPU because it might overheat and also it's not really meant to use a CPU in AI
Wdym with non gaming GPU?
nick i need help
i didnt find the
guide
rtx 6000
@wise fiber tutorials
just clone the notebook brotato 🥀
its like 5 steps in total
also is that kaggle
kaggle is harder to setup
use colabs if you're dumb like me and can afford performance losses
im broke ashell
and it is
what notebook 🥀
i didnt mean it as money, both are free
click the blue hyperlink
done
now click the thing at the top right labeled copy and edit
i did copy into kaggle notebook
yes now open the copy
Yeah you don't have to worry at all, that GPU is meant for workstation, you will have 0 issues
are you able to edit lmao
if you can you already are in it
It's the only FREE working one
It's an hyperlink, blue text that redirects you to a link when you click it
im dense
can u go to vc2
ill screenshare
i clicked it
bt now
its sayijg save version
whar
i got two options to edit in kaggle or colab
do you not see ts
alr
i dont understand shit anynore
my brain
got overloaded
augh
and i have a physcis exam tmrq
but im ttying to make a fake identity online sob
@stoic embermay i get assistance
where do i verify my phone nunber
Meow
I would like to ask for help with this.
@stoic emberAUGHH
@low shard sorry for the ping
WHWRE DO I VERIFY MY PHONE NUMBER AUGH
@low shard im facing a issue,its not letting me verify my phone nubmer
its not letting me verify it
does it send you a text or call
nuthing
nah
fuckkkkkkkkkkkkkkkk
just search up "burner phone sms or sm shi"
dont verify with persona unless you want to be tracked down for the rest of your life btw
what country are u from?
it's a google service bro
those won't work, it's google
i aint trust it
does work funnily
bd
just tried
bangladesh
a free or paid one?
free
wtf did u use it didnt work for me, u sure u mean on kaggle?
yeah, the phone verif
try with ur phone number
it doenst work
might be better to contact their support
not all countries are supported
is kaggle available in your country to begin w
@wise fiber try https://www.kaggle.com/discussions/general/300902#2641240
https://www.kaggle.com/contact this is their support
or just buy a better pc
they are hella expensive
liek i said
one costs here is like 100k
(which is like work of 10 months)
damn ur currency is cooked
then import one brotato
😭
same thing vro
importing wont do anything
the cost is the same
yeah
thats 800 bucks
whats ur annual income brotato
ur only option seems to be contacting kaggle support
welp
im 13
i thought u were a game dev bc of ur bio
uhhhh if you dont wanna use kaggle you can pay for colab tbh
i am learning to be one
i dont think they would want to pay for colab tbh 😭
why have you been on discord for 4 years
dont ask
great observation..
honestly go find some trash 660s for 20 bucks on ebay and use them
and run it locally
a gtx 660 wouldn't work
ebay isnt supported here
and i use a laptop
dont they have SLI
you need min a gtx 900 serie to make it work with cuda and be the minimum decency
even paypal doesnt work here
sad
fire can you pay for nvidia gpu uses directly? its mostly for devs tho
the reason why i want a ai voice is for another identity,mainly for spying on my friends
sob
i cant pay for anything online
free or its a no go
just learn how to do it yourself
wdym? are you trying to catfish?
no
just to see what they think about me
catifsh is bum
you're 13
make a 2nd email and burner phone and try again w a vpn
😭
use one in britain, i think the american ones are dried up on kaggle regs
ohyeha
im tryna become a cc aswell
and i dont reallllt wabba yse my
actual voice yk
use tts then
also i think there was a vid that a guy did a diy voice model
using stt and tts
sounded like an ai bum but yeah
if you know how to use vocaloid software im sure you could plug it into real time stt
although vocaloid software is paid (hasnt stopped most people from torrenting)
but thats unrelated, just a seperate solution
unless kaggle saves you basically
you could always risk it and run it locally anyway
wdym risk it
and how would i run it locally
its gonna fry ur computer
tutorial was alr posted
w the kaggle and colab tutorials
you got big big scammed, you couldve gotten this PC instead https://www.youtube.com/shorts/DjEJMXjZ6mc
Video Link: https://www.youtube.com/watch?v=d-csyC6tm28&t=7s
📩 Business Contact: eddiesdlbusiness@gmail.com
Join this channel to get access to perks:
https://www.youtube.com/channel/UCWzLIO13cVIw0-y1B_bgytA/join
Make Sure to check out my Socials!
🔴Gaming Channel: https://www.youtube.com/@UChWXgjiJY4pFG0n...
um im in another country
what country
bangladesh
should i remove these bits in a dataset?
i can't hear anything when i play and applying 'truncate silence' according to the guides settings doesnt remove them
Check our guides about RVC model creation
was that a reply to me..?
oh, no but... yes u should remove that
tell me
Deitier's one
hmmmm
according to the other guys i talked with
U need a GPU
ah okay thank you. should i do truncate silence with a more agro setting?
what's ur current setting?
im using the guides settings
That's correct!
If there are parts that are not being removed that means there are noise in there
u can remove that parts
U can use colab, but I'm not sure if there is one working
nah, 1 pass it's fine
ig i can use colab
sorry for being dumb but how should i remove those sections then? just manually?
oh yeah is it free?
@viscid moss
yep
yeah, manually
I hope there aren't many
one more question, is it better to remove breathing sounds?
no
its a bad idea to remove them
rvc needs them to learn how to breath
smh spreading misinformation
oh no haha.. i will keep them if thats what the qcer says
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested. GUIDE
ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are way worse. GUIDE
Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated
You're getting sent to the pit with the rest of them
:3
You are safe for now
ik but
the tokens
and how do i set it up
and do i do this
i need hellpp
where do i find the live rvc live voice chager
thanks
that's not a link to teh voice changer
double thanks ?
np
the first link wasn't for u btw lol but that one is
btw @wise fiber if u need more help dm me so nothing gets lost or confusing with all these ppl
kk
ppls help
will do
is their an older version you can give me thats not on the browser
can you not run that version or do u use amd
what's ur pc gpu?
yeah but i had a desktop ver i want to find again
4060
-realtime
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested. GUIDE
ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are way worse. GUIDE
Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated
1st link
wokada deiteris fork
you need more than just the download link
here you can read how to use it and also get vac lite
ok
it just uses a web user interfac elike most ai programs: https://docs.aihub.gg/rvc-voice-changer/local/deiteris-w-okada-fork/#why-does-it-run-in-a-browser-and-not-its-own-window
Last update: May 5, 2025
read that part to get more info about it
anyways it's not a bad thing at all
it's outdated
it is still an application, literally nothing changes except the interface
if you mean the one that opened in a single window, no, that's old original wokada
using a web browser can also help
please just read the link i sent above to understand more
ok
thank you for helping
yw, did u get vac lite?
yeah
be sure u uninstalled the old one, and vb audio cable
@low shard are you able to show me what the C:\Users<YOUR WINDOWS USERNAME>\AppData\Local\Programs\Ultimate Vocal Remover folder should look like for everyone?
the path depends on your windows username, the path is different for everyone
any updates?
no i mean the actual ultimate vocal remover folder
it should be deleted if you're trying to re-install the program
hey where can i download the latest version of the voice changer?



