#✨│ai-help
brittle wing
viral mason
be more specific
hmm what voice changer are you using, and what's your gpu
most likely you're running a year old version since the demo isn't being updated anymore, but since you don't use Nvidia I don't really know how to help u
but you can read through the first link here
-realtime
viral mason
wokada deiteris works on intel I am pretty sure
viral mason
I do yes, extra at 360.0 and chunk at 2.7 or 3.7 depending on how much your pc can handle
low shard
may u share a screenshot of ur current settings? and which intel gpu?
brittle wing
share ur template with me
mb i had to get off my pc i'll send it when i can and ping u
viscid moss
no problem
gusty bison
yo discord isn't taking the voice from voice changed before it was but not now😭
low shard
!give-media-perms 1h @plain silo
low shard
yo discord isn't taking the voice from voice changed before it was but not now😭
Please Elaborate:
- your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
- your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
- what you want to do? There isn't a program that does everything, there's a program for each thing:
- AI Covers
- Train RVC Models
- TTS
- Roleplay in VC
- Roleplay in Games
- etc
- what tutorial link are you using
- a screenshot of the program
that's not an intel dedicated gpu
that's integrated graphics, it's weak asf
@viral mason #✨│ai-help message please always ask the user gpu before helping him, so we don't have those cases where the user just wasted time
you can delete that, unless you have another gpu, you can't run it
do you may have any other gpus?
delete everything then
Your pc is too Weak for Wokada locally, 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/realtime-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)
- 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)
i would suggest either get a better pc or use cloud
so, what you gonna do?
let me know for any issues
viral mason
<@1023278814752677918> https://discord.com/channels/1159260121998827560/11592901...
my bad
low shard
alright, be sure to use kaggle for free
merry eagle
i'm really sorry buddy i was not there
low shard
i'm really sorry buddy i was not there
it's fine lol
merry eagle
i tried on low settings but still it was not good for me
vast elm
https://discord.com/channels/1159260121998827560/1159290139609137264/14023603896...
aw poor guy 🥲 '
low shard
i tried on low settings but still it was not good for me
what game? can you also share ascreenshot of ur program settings?
you can share a screneshot of ur program
stray bison
my voice is not working in vrchat
low shard
my voice is not working in vrchat
This is a General AI Server, AI has many fields, so we can't know your issue with little info
Please Elaborate:
- your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
- your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
- what you want to do? There isn't a program that does everything, there's a program for each thing:
- AI Covers
- Train RVC Models
- TTS
- Roleplay in VC
- Roleplay in Games
- etc
- what tutorial link are you using
- a screenshot of the program
stray bison
its vcclient
low shard
its vcclient
you need to elaborate everything I asked else it's hard for me to help you
there are thousands of reasons why it wouldn't work
stray bison
i cant send a pic
low shard
i cant send a pic
elaborate also your pc gpu, operating system, what you want to do, and send a tutorial link you're using
not onlya pic
!give-media-perms 30m @stray bison
low shard
that's original wokada, it's not suggested
lemme guess, you used a youtube tutorial with also vb audio cable?
stray bison
mb mb
low shard
many users reported vb audio cable giving issues on windows
low shard
mb mb
elaborate your pc gpu, operating system and what you want to do,
so i can help you
those are crucial info
AI is very intensive, it can't work on every single pc either
stray bison
and phyhon??
low shard
and phyhon??
you don't need python lol, the newer programs come bundled with it
stray bison
oh
stray bison
Tysm for telling me
low shard
Tysm for telling me
if you elaborate what I asked, I can give you newer programs
stray bison
I need i voice changer rilly good i got a 4070 it surper
hushed nest
How to use this discord server
low shard
How to use this discord server
this discord server isn't a tool that you can use, nor a program, it's a community
What are you trying to do exactly?
hushed nest
Ok
low shard
How do i get rid of the cable
stray bison
whats the app called ag
viral mason
u should use my most recent Cyn model
low shard
whats the app called ag
what's your pc gpu, operating system, and what do you want to do?
there's various different programs and versions, it all depends on the things I asked you 😭 this isn't chatgpt 1 click, it's open source AI driven by the community
stray bison
I need just a rilly good ai voice
low shard
I need just a rilly good ai voice
as i said, there's different programs and versions which have different scopes... if you don't elaborate i can't know which one of the thousands programs and which version to give you
@stray bison please tell your operating system, and what you want to do for helpers to help you, if we just "guess", it's going to be a waste of time for both sides
viral mason
u can check gpu with task manager
low shard
Its fine i will do it soonj
alr lmk when you can
low shard
u can check gpu with task manager
their gpu is rtx 4070 ti super #✨│ai-help message, but they need to also elaborate what they want to do and their operating system still
it's not the first time i see people trying to use rvc for roleplay, and wokada for ai covers (which is extremely wrong), so i don't want neither of us to waste time lol
merry eagle
<@&1159293204038955078> is vonovox is realtime voice changer?
low shard
<@&1159293204038955078> is vonovox is realtime voice changer?
yes, similar performance and quality as wokada deiteris fork, but for windows nvidia only
-realtime
low shard
<@&1159293204038955078> is vonovox is realtime voice changer?
also, you may want me to check your wokada deiteris fork first #✨│ai-help message
trim lake
does anyone know why when the vc fork is loading, it kills the audio for everything else? it comes right back or i need to restart programs or re open tabs, but just curious if this is something that can be addressed or it just has to be dealt with
sharp lantern
Yes, it run successful
royal lichen
WHERE I CAN DOWNLOAD AI VOICE MODELS?
low shard
WHERE I CAN DOWNLOAD AI VOICE MODELS?
You can search rvc ai voice models at:
- https://discord.com/channels/1159260121998827560/1175430844685484042
- In https://discord.com/channels/1159260121998827560/1163592055830880266 , Do /find with @earnest musk
- https://weights.com/ (login required)
- https://huggingface.co/models (but watch out cus in hugging face there arent only rvc ai voice models)
- https://voice-models.com/
- https://thevoicemodels.com/ (for Turkish Models, login required with discord and level 2 on their server)
if there isnt one, you can:
- make it yourself with our docs guides
- Ask a free request in https://discord.com/channels/1159260121998827560/1159289738314919936
- Be aware that we don't allow any paid comms in the server
earnest muskBOT
You can search rvc ai voice models at:
- https://discord.com/channels/1159260121...
:wave: @low shard, How can I help?
Available Commands:
• @weights find <query> or /find <query> - Search for RVC Voice Models
sharp lantern
Yes, sure
velvet solar
So, I was wondering. I met a friend of mine that has a really good voice changer, but they couldn't find the name. But, I think they said the file name had something like "kraken". Not entirely sure if thats what it was, I'd have to ask again, but im wondering if anyone could help me find that.
restive rover
guys i am looping and looping and looping
like my voicechanger is applying to others type shit🤣
how do i fix this man
brittle wing
guys
i need help
in that video the link in his describtion dosent lead to the same page he shows
restive rover
what doesnt?
placid holly
i test my VC voice
when output speaker is around 400ms/0.4s to hear the result
but when trying in Discord it around 800ms/0.8s to hear the result (using Let's Check)
any explanation on that?
also i don't know if in video game, does it same like DS 800ms?
mint yew
why cant i hear myself
pastel oak
why cant i hear myself
AI is a very very wide spectrum youll have to be more specific about what program youre using, gpu name, any screenshots can be helpful
pastel oak
i test my VC voice
when output speaker is around 400ms/0.4s to hear the result
b...
Delay from virtual cable and delay going into discord and going into your headphones again, but should not be a difference of 0,4 seconds. Are you using Client/MME or Server/Wasapi
burnt ivy
Hi friends... I want to ask if anyone has a Mikasa Ackerman jp applio model?
placid holly
Delay from virtual cable and delay going into discord and going into your headph...
server, all wasapi
urban forum
Hmmm.
Question
I'm using an AMD GPU and the W-Okada fork is forcing me to use my CPU? Why?
dusty rampart
I'm using an AMD GPU and the W-Okada fork is forcing me to use my CPU? Why?
AMD can only run onnx
No CUDA cores rip
!realtime
simple ore
I'm using an AMD GPU and the W-Okada fork is forcing me to use my CPU? Why?
download DML fork from 1st link
urban forum
DML Fork?
urban forum
download DML fork from 1st link
I'm using the fork right now.
The W-Okada Fork right?
simple ore
i mean DML version is for amd
it should list the AMD GPU in the drop-down
if you got Nvidia version you'll only see CPU
mellow wigeon
Could anyone suggest me chunk and extra amount for my settings?
GPU: 7600XT 16GB VRAM
RAM: 32GB 3200Mhz
CPU: Ryzen 5 5500
(I do have a 9060XT which I haven't installed yet, so another suggestion for that one would be nice)
There's also this that I'm not too familar with, so some insight on that would be great too.
final violet
Could anyone suggest me chunk and extra amount for my settings?
GPU: 7600XT 16G...
why does mine look diffrent
..
mellow wigeon
I'm on AMD maybe that's why
This is ONYX
More specifically: MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a
simple ore
More specifically: MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a
delete this piece of shit and download the recommended one from the link above
mellow wigeon
delete this piece of shit and download the recommended one from the link above
uh ok
simple ore
and dont use crepe
final violet
i lmk need some help
mellow wigeon
This thing right?
mellow wigeon
This yea?
final violet
i wish i had pic perms
so wait
i have
'vcclient_win_cuda_2.1.4-alpha
is that the right one
mellow wigeon
is that the right one
You might wanna do the same thing I did what he told me to do
Go up
final violet
o
mellow wigeon
Click the first link
mellow wigeon
scroll down
mellow wigeon
Yeah, tech forums are like very hard to navigate for first timers
It took me awhile to get used to.
and I still get lost sometimes
But to be fair, the one they made on the link
Is very very well made
Like easier than other places.
low shard
like my voicechanger is applying to others type shit🤣
Please Elaborate:
- your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
- your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
- what you want to do? There isn't a program that does everything, there's a program for each thing:
- AI Covers
- Train RVC Models
- TTS
- Roleplay in VC
- Roleplay in Games
- etc
- what tutorial link are you using
- a screenshot of the program
final violet
Could anyone suggest me chunk and extra amount for my settings?
GPU: 7600XT 16G...
i want mine to look like this
mellow wigeon
Well mine looks like this atm
mellow wigeon
Yeah but it looks like the one on browser is the newer updated one.
low shard
i would swap back
why? even original wokada was using a web browser interface, just "in it's own window"
final violet
oh.
low shard
g oback to the other one
original wokada has worse performance, especially for amd
final violet
i want mine to look like this
final violet
Downloading it now. Thanks.
^^
mellow wigeon
original wokada has worse performance, especially for amd
Hey since you're here, you mind dropping me some tips on the dialing for the settings?
final violet
how do i get the regular ui?
low shard
i want mine to look like this
you can't put it in it's own window
unless you drag out a singular browser tab window
mellow wigeon
7600XT 16GB
R5 5500
32GB RAM 3200Mhz
low shard
how do i get the regular ui?
you don't
brittle wing
My output audio cable is my input audio cable on discord for the ai voice changer but i still hear my voice
final violet
i want this
low shard
7600XT 16GB
R5 5500
32GB RAM 3200Mhz
extra: 2.7
chunk: just know to put it higher than the perf value top left while its running
on wokada deiteris fork, you can **optionally **use more advanced settings for benefits:
- Advanced Settings -> Force FP32 mode: on (THIS IS OFF BY DEFAULT! Turning this on improves stability. Increases VRAM usage by 200 MB)
- Advanced Settings -> Crossfade Lenght: Controls how smoothly the AI stitches different processed parts "chunks" of your voice back together. 0.1 for fastest voice, 0.15 for improved quality but increases delay by ~50 ms
- Reduce the delay on Windows via the Wasapi / Asio Guide
final violet
how do i get that
low shard
that's the same program in the browser, just white mode lmao
final violet
that's the same program in the browser, just white mode lmao
..
i thought it was a application 😭
mellow wigeon
extra: 2.7
chunk: just know to put it higher than the perf value top left while ...
Gonna forward that to a private server thread for myself~
low shard
i thought it was a application 😭
is it that much of a big deal? it still runs on your pc, the only difference is the web user interface
mellow wigeon
Thank you very much.
urban forum
Lmao.
low shard
Thank you very much.
need any other help?
urban forum
I put this on my GPU and my GPU is laughing at this program.
low shard
I put this on my GPU and my GPU is laughing at this program.
huh?
urban forum
GPU usage is at 2%
low shard
GPU usage is at 2%
share a screenshot of ur program settings, i could help if that's what u want
urban forum
Oh. No no. That's a good thing.
I'm getting the full extent of the program without burning up my GPU.
I got a new GPU.
low shard
I got a new GPU.
what gpu u got?
mellow wigeon
need any other help?
Just double checking, This amount of chunk is good enough right?
or do I up it to 900 instead
low shard
Just double checking, This amount of chunk is good enough right?
you can just set the chunk lower to 120, chunk controls the delay and has to be put a bit higher than perf
brittle wing
Is there an easy fix for audio cable
mellow wigeon
you can just set the chunk lower to 120, chunk controls the delay and has to be ...
120 just spiked it to 200
ok 285 looks stable
low shard
120 just spiked it to 200
huh weird, are u running other programs in the background? be sure to close all other programs, and to play on 1080p 60 fps cap,
mellow wigeon
Just my browser
when I turn up the chunks
it goes lower
But when I set chunks low, Perf goes high
low shard
https://youtu.be/We5oYpCR3WQ?si=St1b3ACu3Gm9DfCU
are you trying to use an ai girl voice in the video like duckus to troll/catfish?
brittle wing
are you trying to use an ai girl voice in the video like duckus to troll/catfish...
Not really i want to use a bunch for fun but hes the tutorial i found on youtube
I just wanted a voice changer but its not working outside of the application
low shard
Not really i want to use a bunch for fun but hes the tutorial i found on youtube
great, bc catfishing is illegal
low shard
Is there an easy fix for audio cable
brittle wing
great, bc catfishing is illegal
Of course i have no intent to do so
low shard
Thanks a bunch
yw
sharp lantern
It's working for me 😭
Thank you, after run newest version from tab version, it works very well 👍
viscid moss
Nice
flint vapor
latent kettle
Nice
Hello, sorry to bother you but I want to know which model is best for vocals in UVR UI I'm using "Vocals FV4 by gabox"
viscid moss
Hello, sorry to bother you but I want to know which model is best for vocals in...
That's still the best one as far as I know
latent kettle
That's still the best one as far as I know
For reverb and de noise ?
latent kettle
I'm asking which one is better for reverb and denoise
latent kettle
That's on our docs btw
Anvuew mel dereverb v2 where it is ?
The guide says it's under VR but I'm unable to find it
viscid moss
Anvuew mel dereverb v2 where it is ?
That's "MelBand Roformer | De-Reverb by anvuew"
On UVR5 UI
For the OG UVR name model doesn't matter
Just load the model manually
latent kettle
For the OG UVR name model doesn't matter
I'm on URV Ui but the models names are different. So that's why I'm confused
viscid moss
I'm on URV Ui but the models names are different. So that's why I'm confused
Yeah I change their names to fit my format
latent kettle
Yeah I change their names to fit my format
What about de noise model
viscid moss
What about de noise model
Mel-Roformer-Denoise-Aufr33
latent kettle
Mel-Roformer-Denoise-Aufr33
Lastly what about de echo
idle stag
I am reading the docs for dataset prep, it says to remove ALL silences from the dataset. Just making sure that I should do that right
I have removed the background noise and all, only the vocal are there along with the silences between them.
pastel oak
I have removed the background noise and all, only the vocal are there along with...
Rvc splits up audio by default if theres long silences inbetween sentences so ur generally fine to use it like that. If theres long silence moments just remove them to get a better feel for it and improve
restive heath
can someone show proof this so i can study it(I skipped the class)
pastel oak
There are no filters to search by, youd need to hope the creator mentions it in the description. However, long training time does not mean better model, quality of the dataset is important
viral mason
That's "MelBand Roformer | De-Reverb by anvuew"
this model sometimes just doesn't remove reverb when I try it most likely because I guess it thinks it's echo instead
simple ore
-rt
simple ore
1st link
glacial kelp
does anyone know which stable diffusion is the on where I can select pony, sdxl, sd 1.5 etc in one web ui? it was a fork I think I just dont remember it anymore. I could change it instnaly so I can use diffrent checkpoints that required difrent things
simple ore
both input and output should have the same sample rate
read the doc, it says which one to download
royal lichen
Guys, I have a voice changer running, but it doesn't work on Discord. I can only hear my voice, but I don't know.
idle stag
I am in the training tab inside of Applio and the sampling rate options are 32000, 40000 and 48000 but my audio is 44100 which option should I pick? 40000?
pastel oak
I am in the training tab inside of Applio and the sampling rate options are 3200...
This is a highly debated topic, arguments for either:
- 40k: because the ranges from 44.1k to 48k the ai has to "fill in" from nothing, so for "accuracy" its preferred to go down
- 48k: because those frequencies from 44.1k-48k dont make a huge difference anyway and you capture more depth
I personally would do 40k. However you should check if you actually got 44.1k because your softwares and all always save as 44.1k even if its 32k for example. Use program "spek" for ease
idle stag
This is a highly debated topic, arguments for either:
- 40k: because the ranges...
Spek shows this to me..
pastel oak
Need more info:
- your gpu
- what program exactly you downloaded (full name or a screenshot of the app)
- do you have virtual cable installed?
pastel oak
Spek shows this to me..
Its exactly 40k youre good
idle stag
Its exactly 40k youre good
oh alright, thanks
pastel oak
Need more info:
- your gpu
- what program exactly you downloaded (full name or a...
@royal lichen forgot to tag
idle stag
If I want to use the orignal pretrain I should keep "custom pretrained" off and "Pretrained" on right?
simple ore
If I want to use the orignal pretrain I should keep "custom pretrained" off and ...
this uses default og pretrain
idle stag
this uses default og pretrain
Ty
brittle wing
yo, is there a guide on how to use refineGAN models to work on applio?
simple ore
yo, is there a guide on how to use refineGAN models to work on applio?
none of the existing models are good
they are trained with fp32 and old multi-scale mel loss
brittle wing
they are trained with fp32 and old multi-scale mel loss
ah okay good to know
thank u
simple ore
besides I have a new version cooking
jaunty anvil
knotty moth
can someone show proof this so i can study it(I skipped the class)
do you want gemini pro or claude answer this?
hidden jasper
Where's the entry channel..? I can't find it or don't have access to it :(
(just downloaded Voyages! good luck to everyone who enters the giveaway!)
low shard
Where's the entry channel..? I can't find it or don't have access to it :(
(jus...
Hello, sorry for the inconvenience, the channel wasn't correctly mentioned before but now it is #📥│share-your-collections
hidden sky
Does anyone has experience in training models for realtime detection in games? Have some questions about perfomance engine and resolution. 1. Which training engine is the best for such situation YOLO/MobileNet/etc.? 2. When making dataset, should you train on 1920x1080 or you should crop everything to match imgsz?
low shard
Does anyone has experience in training models for realtime detection in games? H...
I never had personal experience for that, but I remember @viscid moss did some detection stuff before
I also remember seeing https://github.com/workofart/brawlstars-ai
viscid moss
I never had personal experience for that, but I remember <@274566299349155851> d...
Gimme 20 mins, I'm in a work meeting
low shard
Gimme 20 mins, I'm in a work meeting
Oh my bad, good luck!
viscid moss
Does anyone has experience in training models for realtime detection in games? H...
I've experience with YOLO
hidden sky
I've experience with YOLO
Nice. So did you successfully made code to detect objects in game realtime? Can you answer on my questions above please?
knotty moth
Nice. So did you successfully made code to detect objects in game realtime? Can ...
im sure it is screen vision
detecting whole game objects would need hacking and be considered cheating
hidden sky
Let's pretend we doing some offline games training, so it's totally fine
knotty moth
Let's pretend we doing some offline games training, so it's totally fine
so what kind of game genre do you want to focus on?
hidden sky
Like I have a problem. I trained model on 1920x1080 dataset screenshots. So after training it with imgsz=480 now it doesn't work for 1920x1080, but for things that near 480x480. I saw many videos on training model and they was using full scale dataset, never saw someone cropping. So by my logic and experience I have to crop dataset to 480, by videos I shouldn't haha
knotty moth
different genre, different model it'd be
hidden sky
so what kind of game genre do you want to focus on?
Let's say FPS
viscid moss
Nice. So did you successfully made code to detect objects in game realtime? Can ...
just detection tbh
not in game/on screen
I made irl stuff detector
viscid moss
Like I have a problem. I trained model on 1920x1080 dataset screenshots. So afte...
for YOLO yes, u need to resize the img to train them
hidden sky
for YOLO yes, u need to resize the img to train them
What about gaming where precision is speed is crucial, but we have good GPU, let's say for avg user gpu 1060. Which training engine would go the best TOLO/MobileNet/something else?
viscid moss
What about gaming where precision is speed is crucial, but we have good GPU, let...
speed it's fine ig. I was running mine on 4060 and detects things every 0.30ms
I used the biggest model aka XL iirc
I can share dataset info and code with u if u want
hidden sky
I can share dataset info and code with u if u want
Would be nice. Thank you!
viscid moss
DMs 
shell kernel
the voice changer isnt working for me
knotty moth
the voice changer isnt working for me
!howtoask
knotty moth
Describe your PC GPU Name, Operating System, the Guide and Step you are on, what you're trying to do, the Program you're using, a screenshot, etc.
shell kernel
i have a nvidea 4070 ti super and i tried the mae vice changer but nothings changed my voice is still normal
i used the novison guide
knotty moth
i have a nvidea 4070 ti super and i tried the mae vice changer but nothings chan...
if the passthru button is red while stopped, click it to change into green
otherwise tell me if you need to post screenshot
also the recommended version and guide is here
brittle wing
!howtoask
brittle wing
how to download rvc
sudden locust
dude I need help with settingss
azure tangle
pastel oak
dude I need help with settingss
Then send your settings and gpu info
pastel oak
how to download rvc
What you trying to do and whats your gpu
bitter cloak
could someone please tell me how do i get started using spin?
bitter cloak
i want to try creating a model with it
solar vine
Hello so I installed W-Okada a few days ago, I tested some voice models particularly the girls because I wanted to troll a bit with my friends but every single voice feels pretty robotic and lags sometimes compared to the ones I seen on YouTube.
PC Specs:
GPU: Nvidia 2060 RTX
CPU: I5 9400
OS: Windows 11
For example this: https://www.youtube.com/watch?v=7upwTBouQn4&ab_channel=fvndz or Duckus
sly summit
PC Specs:
GPU: Nvidia GeForce RTX 4090
CPU: AMD Ryzen 9 7950x3d
OS: Windows 11
So here's my issue, realtime voice changer won't save my settings properly, for some reason they will revert back to default without the display updating
when I launch it says "Output: line 1 (virtual audio cable), however, this isn't true, it's set to my speakers
I've tested this and confirmed it, the only quick fix is changing output to my speakers and back to line 1
This doesn't fix the issue though, the issue that the display isn't showing the actual settings and sometimes reverting some settings back to default
nimble blaze
Anybody here well versed in Python?
simple ore
yes?
nimble blaze
crap I wish I could upload images here
I asked ChatGPT, which is obv not 100% accurate for a roadmap of the stuff to learn to make a custom voice assistant that does things I want. Could you poss tell me whatever's missing/wrong so I could adjust accordingly?
Learning with that goal in mind/building small programs that contribute to said goal would keep me motivated to learn it /not feel like I'm just making random crappy programs
simple ore
voice assistant requires automatic speech recognition (whisper library for example) some kind of LLM to process what you said, as for 'does things' would depend on the things you want to do
nimble blaze
I know
I'm saying I asked chatgpt and I don't know how much of the stuff listed is accurate/if anythings missing/whatevers wrong
simple ore
I'm saying I asked chatgpt and I don't know how much of the stuff listed is accu...
I mean what do you want te voice assistant to do?
just talk back?
tell you jokes? play music?
launch nuclear missiles?
nimble blaze
Open files/apps on my pc/phone, buy stuff online for me, a ton more that I cant think of off the top of my head
simple ore
well, gpt is more or less reasonable
Jarvis, buy me a gallon of lube from Amazon.
what does it have to do with " IZotope Ozone 5 Advanced V5.02"? it is a different product
your mom asks you to buy spagetti, you come home with macaroni elbows
sudden locust
Then send your settings and gpu info
no image perms xd, but ill @ u in my ticket
fleet cedar
why tf when im speaking in a voice message for discord its LOUD af then when i continue to speak its normal
indigo turret
Yo everyone, simple but difficult question : how do I transform a 5-10 mins clean recording of a voice into a huggingface link to put in an ai cover thing ?
low shard
Yo everyone, simple but difficult question : how do I transform a 5-10 mins clea...
you need to train an RVC (STS) AI Voice Model, it can take time
what's your pc gpu and operating system?
flint vapor
why tf when im speaking in a voice message for discord its LOUD af then when i c...
Turn down output volume on voice changer
brittle wing
/collab
indigo turret
you need to train an RVC (STS) AI Voice Model, it can take time
what's your pc g...
If you have the MSI Pulse, your:
GPU is: NVIDIA GeForce RTX 4070 Laptop GPU (8 GB GDDR6)
Operating System is: Windows 11 Home
low shard
If you have the MSI Pulse, your:
GPU is: NVIDIA GeForce RTX 4070 Laptop GPU (8 ...
Nice
idle stag
If I want to train a voice model (Applio) that is unique should I do multi speaker training or individually train them and merge? I am going to be using the voices of three different people, I have 1 hour of audio on each
simple ore
If I want to train a voice model (Applio) that is unique should I do multi speak...
wanna use them separately? train separately
wanna merge them to each other? still train separately
idle stag
K thanks
low shard
be sure to check #📰│dev-updates message
plain island
how to make ai
viral mason
how to make ai
plain island
https://images-ext-1.discordapp.net/external/kJReg1ImBoNDUAQ5SFkEnutaXxw160_t6pq...
😡
are you going to tell me or not.
ember lotus
how to make ai
ai can be made efficiently based on your requirement,
for a instance, if you want ai help you to substitude helpdesk role for your company,
ai can be help as helpdesk chatbot.
low shard
!howtoask
low shard
Please elaborate
low shard
how to make ai
wdym?
low shard
like how do i make a model with coding or smth
which ai model? are you talking about a speech-to-speech voice model ?
there's thousands of different types, models and programs lol
plain island
like a chatbot
thorn timber
Expand-Archive : The path 'python_embed.zip' either does not exist or is not a valid file system path.
At line:1 char:1
- Expand-Archive -Force python_embed.zip runtime
-
+ CategoryInfo : InvalidArgument: (python_embed.zip:String) [Expand-Archive], InvalidOperationException + FullyQualifiedErrorId : ArchiveCmdletPathNotFound,Expand-Archive
I get this error when trying to run setup why
low shard
Expand-Archive : The path 'python_embed.zip' either does not exist or is not a v...
!howtoask
low shard
pls elaborate
thorn timber
I got it working a little better but now im stuck on this error Starting voice conversion...
Starting warmup sequence
Traceback (most recent call last):
File "core\inference\cuvc.py", line 225, in core.inference.cuvc.RVC.init
File "C:\Users\username\Downloads\Vonovox-1.5.0\Vonovox-1.5.0\runtime\Lib\site-packages\transformers\configuration_utils.py", line 781, in from_json_file
config_dict = cls._dict_from_json_file(json_file)
File "C:\Users\username\Downloads\Vonovox-1.5.0\Vonovox-1.5.0\runtime\Lib\site-packages\transformers\configuration_utils.py", line 786, in _dict_from_json_file
with open(json_file, "r", encoding="utf-8") as reader:
FileNotFoundError: [Errno 2] No such file or directory: 'C:\Users\username\Downloads\Vonovox-1.5.0\Vonovox-1.5.0\assets\contentvec\config.json'
Error: RVC model not properly initialized. Check above error for more details.
Failed to initialize audio processing
Critical error in start_vc: Failed to initialize audio processing
Traceback (most recent call last):
File "gui\gui.py", line 1339, in gui.gui.GUI.start_vc
File "gui\gui.py", line 1363, in gui.gui.GUI.initialize_voice_conversion
RuntimeError: Failed to initialize audio processing
I downloaded a faulty pytorch version whichh didn't support my gpu
but then I got a 2.9 12.8 nightly build or something like that which didn't show any errors
asking chatgpt it says im missing the contentvec config json
but I don't know what had happened or where to get it
im gonna read the tutorial incase i missed something
simple ore
Vonovox comes with all you need, no extra torch install is required
thorn timber
Setup.bat never worked for me
so I needed a manual install
maybe ive done it wrong
but Idk how to fix it right
every requirement seems satisfied but this error is always there Expand-Archive : The path 'python_embed.zip' either does not exist or is not a valid file system path.
At line:1 char:1
- Expand-Archive -Force python_embed.zip runtime
-
+ CategoryInfo : InvalidArgument: (python_embed.zip:String) [Expand-Archive], InvalidOperationException + FullyQualifiedErrorId : ArchiveCmdletPathNotFound,Expand-Archive
simple ore
thorn timber
may I ask where it is
pure ledge
Hi guys , does anyone use Ultimate vocal remover 5 ?
simple ore
every requirement seems satisfied but this error is always there Expand-Archive ...
for this it downloads this file
thorn timber
but where is the rtx_5000 setup
thorn timber
https://www.python.org/ftp/python/3.12.8/python-3.12.8-embed-amd64.zip
I downloaded this and renamed
still error
simple ore
you dont need rtx setup, the setup.bat installs the correct version
runtime\python.exe -m pip install --no-warn-script-location torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
thorn timber
but the setup bat give me that error
thorn timber
runtime\python.exe -m pip install --no-warn-script-location torch torchvision to...
this is all satisfied
how do I make a path for the embed
simple ore
I've deleted 1st line of setup.bat to see the outputs
are you for some reason running setup.bat as admin?
thorn timber
no
simple ore
or using a modified windows version that runs everything as admin?
thorn timber
I might have done something while ago forcing powershell to run commands as admin via registry
I can try to return it
simple ore
you can download the embeds and unzip them into runtime folder manually
then delete that part from setup.bat
runtime should look like this when you done
thorn timber
then delete that part from setup.bat
what does this part mean
thorn timber
you can download the embeds and unzip them into runtime folder manually
I tried this and I had already done it before it just asked to replace the files
odd smelt
what the best tts rvc without a gpu 😭
viral mason
what the best tts rvc without a gpu 😭
you're cooked/j
simple ore
what does this part mean
the part of the bat file that downloads and unzips the embeds, lines 5-10
odd smelt
😭
pure ledge
guys is google collab overdated for making ai covers ? I cant find any good link , thats what I used 1 or 2 years ago
sharp marlin
I downloaded an AI voice model of Wallace Breen from HL2. Can I use that in ElevenLabs, and if so, how do I import it?
royal gull
I just had a couple questions for anyone that knows!
- How can I get the voice changer to be almost in real time without much delay?
- Is there a way to fix the almost robot sounding noise it makes at times when you talk?
Let me know if you guys have a work around! Thank you ❤️
rose meadow
Which ai Voice changer should I use
low shard
Which ai Voice changer should I use
!howtoask
low shard
pls elaborate
low shard
I just had a couple questions for anyone that knows!
1. How can I get the voice...
pls elaborate which program and other crucial info
!howtoask
low shard
what the best tts rvc without a gpu 😭
There are different Text To Speech (TTS) AIs:
- GPT So Vits: Great Few Shots (needs a lil training) TTS, its only limited to: english, chinese, Cantonese, japanese & korean, if you wanna check gpt so vits instead, read https://docs.aihub.gg/tts/gpt-sovits/
- 11labs: Easy way to do TTS is https://elevenlabs.io/, its a mostly premium easy way for good quality TTS
- FishSpeech: FishSpeech is a 0 shot (no explicit training needed) TTS, if you got a good pc you can use it locally else use their site
You can check other TTS in our tts index
With RVC Models:
RVC is natively for Speech To Speech, but forks such as Applio have built in tts (using Microsoft Edge TTS to make a tts audio, i suggest you to choose a tts model that is the same gender and language of the rvc model you wanna use, and then convert it with rvc)
If you don't got a good pc you can do tts with RVC Voice Models on cloud:
-
Use Applio UI Colab (with google colab T4 free daily limit gpu)
-
You could try another tts from our tts index and use the output as an input in rvc
low shard
I downloaded an AI voice model of Wallace Breen from HL2. Can I use that in Elev...
I don't think so, RVC is STS, 11labs is STS, 2 different programs, and i don't think their other STS part is related to rvc at all
dark ginkgo
it occcured to me just now but is it possible to use my unused rtx 3060 12gb with my rtx 5060 ti 16gb system for ai useage? how does multi gpu setup look like in a small form factor pc? (I still have two pcie slots btw)
also, is this true?
mint yew
why cant i hearmyself with the voice changer
molten dirge
Does anyone know what the new zentreya tts they use?
idle stag
I use UVR5 to clean my audio for the most part (bg noise, music other stuff) and then I open it in audacity and go through it just to remove anything that it couldn't.
My question is what should I remove?
I currently remove these:
Bits with filters on audio
coughs
extreme increase in pitch (I keep the mild increases in)
Should I be removing them? Or should I leave them so that the model learns how to handle coughs, drastic increase in pitches, laughs etc.
simple ore
Does anyone know what the new zentreya tts they use?
probably a trained voice on Azure speech studio
royal gull
pls elaborate which program and other crucial info
Give me one sec, just finishing up a few things
simple ore
it occcured to me just now but is it possible to use my unused rtx 3060 12gb wit...
you can pair 3060 and 5060 for more vram
as long as you have a decent PSU and a free 4 lane pcie slot
next mist
Has anyone here managed to use local inference using Intel GPUs?
or is there no support?
simple ore
should be possible with convering pytorch model to openvino or onnx
simple ore
realtime voice changer does conversion to onnx
low shard
Give me one sec, just finishing up a few things
Alr
royal gull
Alr
The voice changer is just called Voice Changer Client Demo but some of the settings are in japanese, I downloaded this VC a long time ago
low shard
The voice changer is just called Voice Changer Client Demo but some of the setti...
!give-media-perms @royal gull 30m
Show jt
I feel like you're using an over year old version of original wokada
Wait hold on
!give-media-perms 30m @royal gull
Okay it works now
royal gull
This one
low shard
This one
That's outdated asf
Delete the folder zip
And uninstall vb audio cable from windows app settings
royal gull
Ohhhh, give me a sec, bare with me
royal gull
From your models, are you trying to do e girl trolling/catfishing?
Just vtubing
low shard
I see
royal gull
So delete the entire folder?
low shard
So delete the entire folder?
Delete everything
Everything you got is outdated
I'm guessing you're on windows 11, what's your PC GPU?
royal gull
RTX 2070 Super
low shard
Please read the message above
royal gull
Okay I will
low shard
Okay I will
Lmk
lament bobcat
does anyone know a program that uses the voices we use for real time conversion but as a tts model that sounds good
one that i can put our models to work for but in a tts mode
royal gull
Lmk
It's downloading now, Vonovox
I am at the setup phase now
Setup is complete, do I open Vonovox now?
viral mason
That's "MelBand Roformer | De-Reverb by anvuew"
what's the best crowd removal?
craggy bough
what's the best crowd removal?
chlorine gas
viscid moss
what's the best crowd removal?
UVR-MDX-NET_Crowd_HQ_1
Mel-Roformer-Crowd-Aufr33-Viperx
Try those ones
viral mason
Yea gabox made new peak and I wanna test
swift spade
im tryna make a voice changer model of a rapper so i can use as an index/path file, does anything have a guide on how i can get started
hoary mirage
hi
brittle slate
should i use vonovox or wokada deiteris fork
exotic ridge
Which intel version works best for realtime voice changer? On laptop
fathom mason
how do you add a voice for the voice changer
spring socket
@low shard Sorry for mentioning. I want to ask something about Wokada
The normal setting for RTX 4060 is: (right?)
extra: 2.7
chunk: 90ms
f0: rmvpe without onnx
Is there any optimal setting (lower than that) that I can lower other than the normal settings If I use Wokada while playing game that's a bit heavy (Triple AAA games) so the Wokada still running smoothly?
My Procie is i5 12400F btw.
I'm scared that my rig can't handle it like in the past in the middle of streaming hahaha
severe mango
has anyone here been able to make RVC audio work on linux? if so, which distro? Linux Mint refuses to work.
brittle slate
so im using vonovox and when i try using it in games or discord it doesnt work, even though i used VAC and made sure the input in discord was vac and the output was my audio
sharp marlin
I don't think so, RVC is STS, 11labs is STS, 2 different programs, and i don't t...
Shit.
winged garnet
about the training resources. is it necessary to split it? or if the audio is already separated, is it okay to just leave a 3-4 hour long wav file?
low shard
does anyone know a program that uses the voices we use for real time conversion ...
There are different Text To Speech (TTS) AIs:
- GPT So Vits: Great Few Shots (needs a lil training) TTS, its only limited to: english, chinese, Cantonese, japanese & korean, if you wanna check gpt so vits instead, read https://docs.aihub.gg/tts/gpt-sovits/
- 11labs: Easy way to do TTS is https://elevenlabs.io/, its a mostly premium easy way for good quality TTS
- FishSpeech: FishSpeech is a 0 shot (no explicit training needed) TTS, if you got a good pc you can use it locally else use their site
You can check other TTS in our tts index
With RVC Models:
RVC is natively for Speech To Speech, but forks such as Applio have built in tts (using Microsoft Edge TTS to make a tts audio, i suggest you to choose a tts model that is the same gender and language of the rvc model you wanna use, and then convert it with rvc)
If you wanna do tts locally with RVC Voice Models (if you got a good pc):
- You can get Applio in our docs
If you don't got a good pc you can do tts with RVC Voice Models on cloud:
-
Use Applio UI Colab (with google colab T4 free daily limit gpu)
-
You could try another tts from our tts index and use the output as an input in rvc
low shard
Setup is complete, do I open Vonovox now?
yeah
low shard
im tryna make a voice changer model of a rapper so i can use as an index/path fi...
what's ur pc gpu and operating system?
low shard
should i use vonovox or wokada deiteris fork
what's ur pc gpu and operating system and what u want to do?
low shard
Which intel version works best for realtime voice changer? On laptop
!howtoask
low shard
elaborate more pls
low shard
elaborate everything
low shard
<@911742715019001897> Sorry for mentioning. I want to ask something about Wokada...
you can lower the extra for less delay, sacrificing a bit the quality
also you should play only in 1080p 60 fps cap lowest graphics
low shard
Damn... thanks though! 🙏 Don't worry love i got it fixed
Reminder to not use yt video tutorial and to use wokada deiteris fork b2332
I'm hoping u have a dedicated Intel GPU rather than integrated
tulip cobalt
dark hatch
Anyone here worked with AI Agent and its free will? Mine after few hours of working on its own is doing whatever he wants even after using constrains. Should I make a Guardian Agent or not?
molten dirge
Is there any kaggle ai image/stable diffusion?
next mist
!colab
swift spade
what's ur pc gpu and operating system?
win 10, rtx 4060, i5 12400f, yesterday i separated like 50 vocal splits with demucs and tried using rvc but it was so confusing i didnt understand nothing
low shard
win 10, rtx 4060, i5 12400f, yesterday i separated like 50 vocal splits with dem...
did u read https://docs.aihub.gg/essentials/how-to-make-voice-models/ and got applio?
swift spade
no i didnt ill check it out
swift spade
did u read https://docs.aihub.gg/essentials/how-to-make-voice-models/ and got ap...
im isolating vocals with uvr5 right now and ill follow the guide wiht applio after
swift spade
alr
after isolating the vocals can i just go train the model or do i needa do the spek/audacity stuff?
low shard
without huggingface
low shard
after isolating the vocals can i just go train the model or do i needa do the sp...
yeah
vapid gust
https://docs.aihub.gg/rvc/resources/dataset-isolation/#vocal-isolation--cleaning
best vocal modeL? : :Vocals MDX-Net Gabox's voc_fv4
winged garnet
about the training resources. is it necessary to split it? or if the audio is already separated, is it okay to just leave a 3-4 hour long wav file?
@low shard
simple ore
about the training resources. is it necessary to split it? or if the audio is al...
one file is fine, depending on the splitting method you gonna use
3-4 hours is a an excessive amount of audio to train a model on
variety and quality > quantity
winged garnet
variety and quality > quantity
3 4 hours is quality sound, not just for show. Is that okay?
odd shale
3 4 hours is quality sound, not just for show. Is that okay?
Honestly, i think you're good to go with just 1 hour.
After all, all what matters is how clean your dataset is and avoid overtraining it.
winged garnet
okay. tysm
simple ore
3 4 hours is quality sound, not just for show. Is that okay?
30 minutes of expressive audio-book reading is better than 4 hours of monotone podcast talking about Donald Trump
odd shale
okay. tysm
You're welcome buddy.
low shard
True
11.6
NVIDIA GeForce RTX 4060
could you show a screenshot of the issue?
swift spade
well i stopped the training but when i was training, 20 mins had passed by and it only went thru 4 epocghs and i checked task mgr and my cpu was being used and my gpu wasnt
simple ore
well i stopped the training but when i was training, 20 mins had passed by and i...
check task manager
GPU is being used and within normal VRAM
swift spade
i said i did, my gpu was at 1%
simple ore
you can see GPU being using with this metric
also on the training tab there's Advanced Settings at the top
expand that and it should show the GPU
swift spade
yes its detected
and selected
yea i trained again and now its 7.2/7.9 shared gpu memory
14.7/15.9 gpu memory
simple ore
batch size? checked [x] cache dataset in gpu?
if with shared memory being used the training is 4-6x slower
so use a resonable batch size, dont cache data, check [x] Checkpointing if you still exceeding VRAM use (it will be a little slower, but would use much less VRAM)
swift spade
yea i just got cuda outta memory
simple ore
so answer my questions
simple ore
I mean what batch size you're using that leads to OOM?
simple ore
what version of applio you got?
swift spade
3.2.9
simple ore
do you have hardware acceleration enabled in the browser / discord?
swift spade
both yes
simple ore
well, you're very close to VRAM limit with batch 8
so either lower it to 6 or 4, or enable checkpointing in the training options
swift spade
ill try 6
OsamasonModel | epoch=1 | step=687 | time=12:19:30 | training_speed=0:04:49
and i gotta go thru 650 epochs
its overrr
😭
simple ore
why are you going 650 epochs?
swift spade
saw in the guide to do aroudn 100 since its using tensorboard
viral mason
OsamasonModel | epoch=1 | step=687 | time=12:19:30 | training_speed=0:04:49
Just use applio on kaggle mr deer if you're not already
swift spade
tried it on kaggle it was a lil confusing
simple ore
normally 30-60 minutes is more than enough
swift spade
i downloaded 103 songs
by a rapper
lol
should i do like 30 songs then
30 vocals
viral mason
i downloaded 103 songs
Holy
swift spade
only ai training ive done before was image recognition and sometimes 1k wasnt even great
so i assumed this was the same
simple ore
should i do like 30 songs then
if it is the same rapper, dont go over 1 hour total
give it a try
maybe even 30 min
swift spade
alr how many epochs
simple ore
alr how many epochs
normally 200-300 is enough
viral mason
How will I know if I've been accepted as a Quality Control unit
I wanna sweep away the garbage in case someone posts something yucky
low shard
How will I know if I've been accepted as a Quality Control unit
You will get a DM about the application result
torn junco
Can anybody help me run stable diffusion on my Intel arc A750 for image upscale/enhance? I'm not an expert and upscalers like krea ai are paid
royal kettle
hello, i was sent here by a tut. The tut kinda sucked and never explained how models were actually made or where to go to find one. So that brings me to my question; how do i make a good quality voice model? I only know the bare basics of audio, like compression is bad for quality.
low shard
hello, i was sent here by a tut. The tut kinda sucked and never explained how mo...
lemme guess, you used a youtube tut for realtime voice changer?
all yt (video) tuts are extremely outdated, they prob made u download vb audio cable along an over year old version of original wokada, am i right?
royal kettle
all yt (video) tuts are extremely outdated, they prob made u download vb audio c...
Yes, but i didnt download anything because i know stuff in ai moves fast.
But that doesnt answer my question of how do i make a quality voice model
low shard
Yes, but i didnt download anything because i know stuff in ai moves fast.
forget every info on youtube, every single one is old
low shard
But that doesnt answer my question of how do i make a quality voice model
what's your pc gpu and operating system?
I'm guessing you want RVC Speech-To-Speech Voice Models
royal kettle
forget every info on youtube, every single one is old
yea, i assumed so
royal kettle
what's your pc gpu and operating system?
i got a 4060 and im using win 10
low shard
i got a 4060 and im using win 10
rtx 4060 8gb desktop?
low shard
no, sorry its a 4060 ti 16gb
that's great, you can use Applio https://docs.aihub.gg/rvc/local/applio/, an RVC Fork with some performance improvements and extra feature
royal kettle
that's great, you can use Applio https://docs.aihub.gg/rvc/local/applio/, an RVC...
A fork? So other than perf improvements is there any difference between the original version and Applio?
i dont care about bloat features
low shard
A fork? So other than perf improvements is there any difference between the orig...
fork = modified version
Applio has better performance and it's more maintained (as the original RVC project devs kinda left it to rot in 2023 to go work on other things not related to voice models, like GPT-So-VITS), along with an easier ui, with features i didn't mean bloat, i meant actual useful things such as TTS
Quality? Nope, the rvc quality is the same as the one in 2023 with rvc v2, no one can do anything about it it seems
but applio is more suggested for the things i mentioned
royal kettle
fork = modified version
Applio has better performance and it's more maintained ...
Is the TTS any good? Is it anything like F5, Kokoro, and Zonos?
low shard
Is the TTS any good? Is it anything like F5, Kokoro, and Zonos?
since RVC is natively Speech To speech, the TTS works via using Edge TTS as input for RVC, meaning it will be multilingual and high quality, even tho not emotional
royal kettle
Quality? Nope, the rvc quality is the same as the one in 2023 with rvc v2, no on...
Has no one been working on RVC or something? Why hasnt there been improvements since 2023?
like nothing from the community/new devs
low shard
Has no one been working on RVC or something? Why hasnt there been improvements s...
RVC isn't easy, there are alot of devs who tried experimenting with it #🔊│ai-development , but no one is seeming to make an "rvc v3", the code is very complex and you can't easily just upgrade the quality
royal kettle
RVC isn't easy, there are alot of devs who tried experimenting with it <#1159290...
hmm, ok.
low shard
hmm, ok.
There have been alot of experiments though, like using community made pretrains, refinegan and other things (which you can use in applio), but still it won't help with quality like an "rvc v3" would
This is also why this place isn't rvc focused anymore, and more general, as RVC itself is dead by its original devs
royal kettle
https://docs.aihub.gg/essentials/how-to-make-voice-models/ can help too
is there a more advanced guide on how to make quality voice models that i can look at later? The stuff there seems basic, unless RVC is easy and its that simple.
low shard
is there a more advanced guide on how to make quality voice models that i can lo...
that guide links to the 3 steps to make a model, each step (well except downloading rvc lol) has a more complex guide, like the tensorboard and for datasets using rx11
I'm not sure what you expected it to be though
Training an RVC model isn't super difficult, that's why there's thousands of them
royal kettle
that guide links to the 3 steps to make a model, each step (well except download...
Just skimming through links that link to more guides at the moment but it just seemed so simple. Is there really not much difference between a model made by someone new and one made by someone whos been using this ai for a while?
I am very willing to read through many guides if it means i can get a good voice model
royal kettle
since RVC is natively Speech To speech, the TTS works via using Edge TTS as inpu...
Also since Applio is using Edge as the 'base' for the TTS does that mean the outputted audio will carry all of the flaws of Edge?
low shard
Just skimming through links that link to more guides at the moment but it just s...
I mean that's the way you train rvc models,
you could also use applio's main branch which contains more experimental features, use also other community pretrains (explained in the guide), could also use refinegan
but those are just experimental, it won't guarantee you that using experimental stuff will help for quality like @simple ore (engineer) would say
low shard
I am very willing to read through many guides if it means i can get a good voice...
it would be better you read through all the stuff
be also aware that rvc isn't that good at non speech sounds, so for example it can't always do super realistic laughs, even tho training on them could help for certain laughs
royal kettle
be also aware that rvc isn't that good at non speech sounds, so for example it c...
Why is that? Is it tech limitations or just how RVC is built?
royal kettle
I mean that's the way you train rvc models,
you could also use applio's main bra...
After googling RefineGAN it seems to good to be true. How can you create something better than groud truth?
wispy glen
Can someone please help me in DM on how to use the AI voice changer plsss
simple ore
After googling RefineGAN it seems to good to be true. How can you create somethi...
easy
royal kettle
easy
So it is just filling in the areas taken by compression with synthetic stuff?
What if the audio isnt compressed and is taken care of properly?
how does Refine benefit that?
royal kettle
Hmm, based on the overtraining information "model is unable to produce high end harmonics" and "when the graphs in the Tensorboard are going up" does that mean i should watch the harmonics of the audio to make sure they dont vanish to find when my model is 'done'? Or should i just watch the graphs for when they go up? This guide seems pretty basic (https://docs.aihub.gg/rvc/resources/training/#tensorboard), im finding it hard to believe that this is what experts with RVC do.
Also why doesnt this ai use some sort of validation like other ais? Has it just not been attempted yet? Why not use something like NISQA?
simple ore
there are no validation since there's no ground truth for audio converted to another voice
glacial aurora
BRRO
royal kettle
there are no validation since there's no ground truth for audio converted to ano...
Why not just grab a small chunk of the dataset and use that as validation? It wont be a unseen voice but it will give you an idea, no?
simple ore
the only true validation is listening to the result of conversion and deciding whether it is good or not
simple ore
Why not just grab a small chunk of the dataset and use that as validation? It wo...
using 100 VCTK speakers to train and 10 speakers to validate wav2mel2wav is possible
because they are all unique and separate speakers
when you have a single speaker it is all "seen"
so the best validation you can do is whether the model can closely reproduce its own dataset
royal kettle
Why not do that? its better than no validation.
simple ore
if I take 10$% of random characters from "Why not just grab a small chunk of the dataset and use that as validation? It wont be a unseen voice but it will give you an idea, no?" how big is the overlap with the rest of the sentence?
i'm not going to repeat myself
royal kettle
so thats a valid reason to not add any validation? ok i guess.
simple ore
royal kettle
https://discord.com/channels/1159260121998827560/1159290096458149938/14008851164...
no access
swift spade
normally 200-300 is enough
im doing 58m of vocals at 260 epoch, 1:20 each epoch
😭
7 batch
opaque marlin
holas alguien que hable español que me resuelva algunas dudas?
quartz moon
yall uh how do i know if im using chatgpt 5 or not
it said introducing chatgpt 5 when i opened a chat but when i asked it it said the model is chatgpt 4o
and i also asked it to make a game of tetris and it made the exact same one as the gpt 4o unlike the vids
low shard
it said introducing chatgpt 5 when i opened a chat but when i asked it it said t...
try opening it directly from the "try chatgpt5": https://chatgpt.com/?openaicom-did=daa5b0ad-c3e0-4764-aa20-26bfc12114fe&openaicom_referred=true
low shard
holas alguien que hable español que me resuelva algunas dudas?
please elaborate in english
opaque marlin
please elaborate in english
I have a question about the voice changer.
low shard
Can someone please help me in DM on how to use the AI voice changer plsss
This is a General AI Server, AI has many fields, so we can't know your issue with little info
Please Elaborate:
- your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
- your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
- what you want to do? There isn't a program that does everything, there's a program for each thing:
- AI Covers
- Train RVC Models
- TTS
- Roleplay in VC
- Roleplay in Games
- etc
- what tutorial link are you using
- a screenshot of the program
low shard
I have a question about the voice changer.
you can ask it, please also elaborate, the term "voice changer" can refer to multiple programs, and be sure to not use youtube video tutorials as they are old for voice changers
opaque marlin
you can ask it, please also elaborate, the term "voice changer" can refer to mul...
Well, I don't know, here it says it's called voice-changer-windows-nvidia-b2332 and what happens is that I don't know, but in some games where I want to use it, it doesn't work, meaning the audio stops just like that.
And it only resumes when I enter Google again
low shard
Well, I don't know, here it says it's called voice-changer-windows-nvidia-b2332 ...
oh that's the latest wokada deiteris fork, good
low shard
And it only resumes when I enter Google again
you should never close the browser, nor the command prompt
what game are you playing? what's your pc gpu? i'm guessing you're on windows 11
opaque marlin
For example Red Dead Redemption, also with Days Gone, and several more, my PC is an Azus Tuf Gaming A15 2023 RTX4050, AMD Ryzen 7, and I don't close either the command prompt or Google is the audio
The one who stops out of nowhere
quartz moon
try opening it directly from the "try chatgpt5": https://chatgpt.com/?openaicom-...
oh cool it says that now ty
low shard
For example Red Dead Redemption, also with Days Gone, and several more, my PC is...
play on lowest graphics 1080p 60fps cap
show a screenshot of ur program settings
quartz moon
yall when uh i run out from responses from chatgpt 5 wil it return me to 4o?
low shard
yall when uh i run out from responses from chatgpt 5 wil it return me to 4o?
nope
it seems like they deleted all previous models
quartz moon
i ran out of messages but i can still talk to him
ig they made a dumber chatgpt 5 or smthn
low shard
ig they made a dumber chatgpt 5 or smthn
idk if they added older models back
but there's a super recent article about it
it ended like 2 hours ago
quartz moon
THEY R MAKING CHAT GPT 6?
low shard
THEY R MAKING CHAT GPT 6?
not yet
opaque marlin
show a screenshot of ur program settings
low shard
extra: 2.0
uninstall vb audio cable, get vac lite https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#virtual-audio-cable
uncheck sup1
on wokada deiteris fork, you can **optionally **use more advanced settings for benefits:
- Advanced Settings -> Force FP32 mode: on (THIS IS OFF BY DEFAULT! Turning this on improves stability. Increases VRAM usage by 200 MB)
- Advanced Settings -> Disable JIT compilation: off for faster loading speed of the program, on for slightly better performance (10-15 ms) for Nvidia only)
- Advanced Settings -> Crossfade Lenght: Controls how smoothly the AI stitches different processed parts "chunks" of your voice back together. 0.1 for fastest voice, 0.15 for improved quality but increases delay by ~50 ms
- Reduce the delay on Windows via the Wasapi / Asio Guide
opaque marlin
extra: 2.0
uninstall vb audio cable, get vac lite https://docs.aihub.gg/realtime...
I didn't understand anything haha help
low shard
I didn't understand anything haha help
first of all uninstall vb audio cable and get vac lite
opaque marlin
first of all uninstall vb audio cable and get vac lite
And since I uninstalled it, I went to settings and uninstalled it, but it still appears and within the files of that vb I don't see any app that says uninstall.
low shard
And since I uninstalled it, I went to settings and uninstalled it, but it still ...
you need to delete any folder or zip related to vb audio cable
then do this
then, get vac lite from the guide link i sent
opaque marlin
then do this
I already did that but it's still there. I just need to drag the root folder to the trash, right?
low shard
I already did that but it's still there. I just need to drag the root folder to ...
where is it? the zip and folder of vb audio cable?
opaque marlin
where is it? the zip and folder of vb audio cable?
This is the folder I have to delete, right?
swift spade
instead of using my model just for real time voice changing how can i change the voice of an already recorded file?
low shard
Now he left me 3 different ones according to
wdym?
low shard
Which of these do I choose?
try restarting your pc first and see if vb audio cable is still there
opaque marlin
opaque marlin
I already did it and one is like this
I restarted it 2 times, uninstalled the app from settings, apps, installed apps
And I deleted the root folder of that virtual cable completely but they are still there
low shard
search for vb
low shard
reduce extra to 2.0
uncheck sup 1
select line 1 as the output in wokada, and select line 1 as the input in games
opaque marlin
Protocol: sio
Crossfade length: 0.15 s
SilenceFront: on
Force FP32 mode: on
Disable JIT compilation: on
Protect: 0.5
Skip Pass through confirmation: No
this is how i set it up
low shard
Protocol: sio
Crossfade length: 0.15 s
SilenceFront: on
Force FP32 mode: on
...
you could set crossfade lenght to 0.1 for lower delay (will slightly impact quality)
test it and lmk
be sure to play 1080 60fps cap lowest graphics
tame jewel
how do i fix a 2 second delay in the software
austere cipher
-rt
low shard
how do i fix a 2 second delay in the software
what software? elaborate, this is a general ai server
!howtoask
drowsy veldt
can somebody help me rq
low shard
can somebody help me rq
sure, elaborate your help request
drowsy veldt
I use weights for simple voice conversions, but before that i would use a "hina mod" simple collab notebook, unfortunately something in it messed up and it hasnt worked since, ive been using weights but i did prefer the google collab because i could customize accent, pitch, etc. I asked some guy on here for help and he said google collab conversions would NEVER come back, i didnt think of it too much and continued to use weights but i recently saw someone on tiktok post an AI cover and said it was made by gradio RVC notebook on collab, so im troubleshooting/asking for help if anyone knows if its still possible to use gradio/collab, im not very skilled in the conversion AI usage so it would be a dream to use it again
low shard
I use weights for simple voice conversions, but before that i would use a "hina ...
you're talking about hina mod AICoverGen, I was the dude you talked to #✨│ai-help message and yes AICoverGen is abandoned, but not google colab as a whole, colab is just a cloud (remote good pc) service, everyone can code their AI notebooks,
https://docs.aihub.gg/ you can check the docs for RVC Cloud guides (not the same notebook you used, but there are other ones)
low shard
I use weights for simple voice conversions, but before that i would use a "hina ...
pitch
btw, you can costumize the ai covers in weights via basic and advanced settings, yk that right?
drowsy veldt
ahh, my bad im still not the best with conversions and yes i know about weights advanced settings but the one i had was accent and a whole bunch of other stuff, so in simple terms, its over for me?
last time i talked to you i had a trash laptop but i bought a PC so i think im good now
ill peep th elink u sent
low shard
last time i talked to you i had a trash laptop but i bought a PC so i think im g...
what's ur pc gpu?
low shard
ahh, my bad im still not the best with conversions and yes i know about weights ...
AICoverMaker is a "continuation" of AICoverGen, and it's in the docs if that's what u want
drowsy veldt
what's ur pc gpu?
im not sure if this is right or how to get to it but i think its GPU 0
AMD Radeon(TM) Graphics
PCI bus 14, device 0, function 0
Utilization 4%
Dedicated GPU memory 427/512 MB
Shared GPU memory 0.6/15.5 GB
GPU Memory 1.0/16.0 GB
is this terrible
low shard
is this terrible
if there's no other gpu 1, then yeah
that's still integrated graphics
drowsy veldt
aw man
low shard
the thing you need the most for AI is a good dedicated gpu
like an rtx 5060
which ofc is expensive
drowsy veldt
aw hell, weights it is
chilly furnace
yall whag
simple ore
tf you doing?
brittle slate
what's ur pc gpu and operating system and what u want to do?
RTX 4080, and windows 11 and i wanna use a RVC
simple ore
rvc as realtime voice changer or as retrieval-based voice conversion?
brittle slate
rvc as realtime voice changer or as retrieval-based voice conversion?
realtime voice changer
chilly furnace
tf you doing?
applio kaggle ??
low shard
realtime voice changer
rvc doesn't mean realtime voice changer
brittle slate
oh
low shard
rvc means retrieval-based-voice-conversion
low shard
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. There also updated forks with extra features like Applio.
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)
low shard
i have a problem with vonovox
what's the issue?
brittle slate
what's the issue?
so i tried using it in games and discord but i cant hear anything, i can only hear the ai voice changer work in the app
i tried setting the input in discord to the VAC
it still didnt work
low shard
so i tried using it in games and discord but i cant hear anything, i can only he...
show a screenshot of ur settings
!give-media-perms @brittle slate 30m
brittle slate
in discord or vonovox?
low shard
!give-media-perms 30m @brittle slate
low shard
in discord or vonovox?
both
brittle slate
one sec
simple ore
i'd change the input to actual microphone.. instad of voicemod whatever that is
brittle slate
i'd change the input to actual microphone.. instad of voicemod whatever that is
its a soundboard app but ill try doing that
low shard
also use rmvpe
brittle slate
also use rmvpe
why?
brittle slate
i'd change the input to actual microphone.. instad of voicemod whatever that is
tried that but it didnt work
viral mason
never put voicemod in the audio things dum dum, where that digital audio thing is for me, u put just ur regular mic
brittle slate
best precision and robustness, tho it is more heavy
okay that somewhat fixed it, thanks
i can hear it working in discord
pulsar anchor
What can GPT-5 do?
empty leaf
Hello, I installed W-Okada today and it had a very long delay, even on 320 setting. Everything below that cuts off for me.
Is Wokada Deiteris Fork the go to option for me if I want real time voice changing ? I have an AMD gpu and cpu so Vonovox wouldnt work according to this chart
tepid cargo
Hello, I installed W-Okada today and it had a very long delay, even on 320 setti...
Which AMD GPU do you have?
empty leaf
RX 6650 XT and ryzen 5 5600X cpu
tepid cargo
Weird. What OS and what version?
empty leaf
Windows 11
It was the basic W-Okada I found on some old youtube tutorial from 1 year ago
tepid cargo
Can you check if it is 100% using your RX 6650 XT when you are using the vc? You can see in task manager when you are using it
I think the terminal may also say something as well
empty leaf
I'm currently installing the other version, I already nuked the previous versions folder 😬
The Deiteris version, idk if that's better
tepid cargo
I see. I believe there are separate requirements/dependencies when using an AMD gpu. Make sure you are installing those rather than for nvidia (not sure if this was/is the problem, but possible)
empty leaf
Might have been, I used a strang egit hub link for the basic W-Okada
Now im using this servers link for the Deiteris one, I think I did everything correctly. Trying rn
"Windows protected your PC" do I just ignore that?
I installed it from the -rt link
-rt
empty leaf
It opened the prompt then vanished
tepid cargo
"Windows protected your PC" do I just ignore that?
Yes, it should be fine
tepid cargo
It opened the prompt then vanished
When?
tepid cargo
Can you show me ss of what you ran in https://discord.com/channels/1159260121998827560/1192011222023950368
empty leaf
I put it there
urban wraith
can someone help me get this ai stuff working?
all i want rn is to sound like market pliers
tepid cargo
can someone help me get this ai stuff working?
https://discord.com/channels/1159260121998827560/1393389200862089240
Get the suggested local client and market pliers model from https://discord.com/channels/1159260121998827560/1175430844685484042
placid holly
i just Install UVR5 UI
now any recommendation what model should i download?
- Separate room noise (fan, pc, keyboard, tapping etc) & voice
- separate music & Vocal
hallow thistle
Anything catfishing is banned here. Sorry with that.
paper widget
how to use this https://huggingface.co/Abhinay45/XTTS-Hindi-finetuned
fierce ivy
Yo guys
Please open the following URL in your browser.
https://<IP>:<PORT>/
In many cases, it will launch when you access any of the following URLs.
https://127.0.0.1:18888/
https://192.168.0.2:18888/
Booting PHASE :MMVCServerSIO
[Voice Changer] VoiceChangerManager initializing...
[Voice Changer] VoiceChangerManager initializing... done.
[Voice Changer] MMVC_Rest initializing...
[Voice Changer] MMVC_Rest initializing... done.
[Voice Changer] MMVC_SocketIOApp initializing...
[Voice Changer] MMVC_SocketIOApp initializing... done.
what i can do
hallow thistle
Please open the following URL in your browser.
https://<IP>:<PORT>/
In m...
Which W-Okada are you using for this multi-PC LAN setup?
fierce ivy
Im not using multi-pc lan
terse jetty
hello, i just downlaod the deiteris-Fork RVC file,as i extract the folder, it is saying that it cannot failed to extract/load the pretrain model, after i delete it and install it again, it still failed regardless, is there a way to fix this? is the download wrong? i used the link provided in the channel
2025-08-09 17:38:58,068 ERROR [WeightDownloader] Failed to download or verify pretrain/fcpe.onnx
2025-08-09 17:38:58,069 ERROR [WeightDownloader] Cannot connect to host huggingface.co:443 ssl:default [The semaphore timeout period has expired]
NoneType: None
Traceback (most recent call last):
File "client.py", line 22, in <module>
File "asyncio\runners.py", line 190, in run
File "asyncio\runners.py", line 118, in run
File "asyncio\base_events.py", line 654, in run_until_complete
File "main.py", line 91, in main
File "downloader\WeightDownloader.py", line 88, in downloadWeight
Exceptions.PretrainDownloadException: 'Failed to download pretrain models.'
it says something like this, i am using a rtx 2080 super, win 10, just on the extracting phase of the folder
hallow thistle
Im not using multi-pc lan
So which W-Okada version are you using anyways? And what is your PC GPU?
hallow thistle
2025-08-09 17:38:58,068 ERROR [WeightDownloader] Failed to download or verify...
If you see this error, it means the "pretrained models" download have failed and your internet was slow. You either connect to a better network, delete "pretrain" folder in MMVCServerSIO and try run the program again, or follow this solution link. https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#how-to-fix-failed-to-download-or-verify
gloomy lynx
Trying to use UVR, but the problem I'm having is that the reverb/echo seems really stuck on the song, even max aggression on the de-reverb model isn't doing much
And Idk if it's an issue on my end, because I've had about as much luck with 5 aggression as I had with 100
terse jetty
If you see this error, it means the "pretrained models" download have failed and...
Thankyou for the help, I extracted everything and out in those pth and index, I was trying to get my mic connected to my voice changer, I check the inputs and outputs which is fine, and I can't seems to see what had gone wrong, I tried discord input voice device to see if it's the problem of my mic, but it isn't, the voice test is fine, it is just on this voice changer that it's not receiving or tramsitting any voice, is there a fix to that?
hallow thistle
Thankyou for the help, I extracted everything and out in those pth and index, I ...
Did you use Virtual Audio Cable or VB-Cable/Voicemeeter on this one?
terse jetty
Did you use Virtual Audio Cable or VB-Cable/Voicemeeter on this one?
Let me clarify sorry, So you are suppose to hear the after-process voice in your headphones(monitor) , but not even that is currently happening for me, so I suspect it didn't even translate any voice, that's my concern, I am aware of a virtual audio cable and have it downloaded
2025-08-09 18:54:11.4453769 [W:onnxruntime:, transformer_memcpy.cc:74 onnxruntime::MemcpyTransformer::ApplyImpl] 2 Memcpy nodes are added to the graph main_graph for CUDAExecutionProvider. It might have negative impact on performance (including unable to run CUDA graph). Set session_options.log_severity_level=1 to see the detail logs before this message.
hmm.. is this one of the problem?
ok i figured it out ,thx for the help though !
simple ore
how to use this https://huggingface.co/Abhinay45/XTTS-Hindi-finetuned
install coqui, use a script
simple ore
why is the TITAN pretrain hated now?
it is a finetune of original pretrain with unknown number of speakers, which potentially downgraded the model's generalization
surreal vessel
Wouldn't it be a good idea to release a new compiled version of Applio (for beginners)?
The last one is from April, if I'm not mistaken....
nimble blaze
What's the best TTS to use with RVC for real time conversion, for an assistant?
royal kettle
In the docs why does it say to remove all the silence? It just says to do it and doesnt give any reason as to why? Is it just for a smaller file size or does this affect something in some way?
https://docs.aihub.gg/rvc/resources/dataset-isolation/#step-2-truncating-silence
safe kayak
how do i use W-Okada Fork
low shard
Please open the following URL in your browser.
https://<IP>:<PORT>/
In m...
Please Elaborate:
- your PC GPU (Such as Nvidia RTX 3060, AMD RX 9070 XT..)
- your operating system (Like Windows 10, Windows 11, Linux, MacOS 26..)
- what you want to do? There isn't a program that does everything, there's a program for each thing:
- AI Covers
- Train RVC Models
- E girl trolling/catfishing
- TTS
- Roleplay in VC
- Roleplay in Games
- etc
- what tutorial link are you using
- a screenshot of the program
low shard
how do i use W-Okada Fork
!howtoask
low shard
please elaborate
untold axle
Setting up a micro data centre would cost? Realistically?
hallow thistle
how do i use W-Okada Fork
The thing is, what do you use fork W-Okada the realtime voice changer for? And what is your PC GPU?
hallow thistle
Setting up a micro data centre would cost? Realistically?
Micro datacenter or a small-scale server? Both are costly to set up, which include hard drives, and a host device or computer especially.
formal ermine
-realtime
tidal parrot
hi
tidal parrot
Using Intel(R) UHD Graphics 600, Windows 11, I'm in weights.com at the screen with 6 options (Image, Video, Voice Cover, Song, Conversation, Character). I think the option I need is Voice Cover, as I want to modify the voice of an existing song. Program? What? I'm using Google Chrome
Selected Ganyu japanese voice, until now I think it's correct
the song
and then
this screen showed, I kept in the middle (0) and when I listened to the result the voice was still the same, that's the problem
Should it be negative or positive? <@&1159293204038955078>
low shard
this screen showed, I kept in the middle (0) and when I listened to the result t...
a lower pitch makes it sound more 'masculine'
an higher pitch makes it sound more 'feminine'
you may need to play with that setting till it feels right, and also try other models
tidal parrot
a'ight
but it's weird how it was in the middle and didn't turn into Ganyu voice, nothing. Shouldn't it do as I selected the model?
still I'll try and make positive, feminine, hope Ganyu's voice appears this time
hallow thistle
this screen showed, I kept in the middle (0) and when I listened to the result t...
If the vocal audio is extracted from start, you can select "Pre-Stemmed". If you leave it turn off, the website would attempt to extract vocals and instrument from your given vocal audio.
tidal parrot
I tried +7 selecting pre-stemmed and the voice remained the same as the original -_-
hallow thistle
Additionally, you can now upload other full mp3 or mp4 files into Weights, the website will automatically extract into vocal and instrumental tracks and then convert AI cover by steps as usual.
tidal parrot
that's gold
hallow thistle
After it finishes processing, there you can download converted vocals and other stems from there.
tidal parrot
trying +12 now, if it doesn't work I'm trying another model but I really love her jp 🇯🇵 voice
hallow thistle
Most of female Genshin Impact and Blue Archive characters, if the original vocal audio is male, I always set pitch up to +12.
tidal parrot
After it finishes processing, there you can download converted vocals and other ...
my download options are less than that
tidal parrot
Most of female Genshin Impact and Blue Archive characters, if the original vocal...
important info
hallow thistle
my download options are less than that
Whoops. I forgot this feature is for premium users. I have Weights Premium, so I didn't see the difference. 
Even so, you can go for extracted vocals instead.
tidal parrot
mmmmmmm the guiding of this site is very good and intuitive, really hope I keep testing and get my result
+12 also didn't work
+18 also didn't, so weird how it doesn't seem to change
actually it did ^^
I was playing only the input as the output option wasn't showing before. I knew something very basic was missing
wonderful it's easier than RVC website * -*
@low shard @hallow thistle Thank you so much for guiding the first steps to create, less than 1 hour and it's done
I'll make more later
low shard
<@911742715019001897> <@561807329977434122> Thank you so much for guiding the fi...
have a nice day
tidal parrot
😹 definitely using
candid meteor
I have a dataset now on audacity what sample rate do i export it in?
low shard
I have a dataset now on audacity what sample rate do i export it in?
https://docs.aihub.gg/rvc/resources/dataset-isolation/#step-1-find-the-sample-rate in the same sample rate that you got by spek
candid meteor
thanks
viral mason
😹 definitely using
2 years ago? Yikes
viral mason
After it finishes processing, there you can download converted vocals and other ...
If u spend money u get this
hidden sky
Does anyone know why once I convert model from onnx to trt, few classes stop being detected? Like 2 out of classes are not detected at all (while in pt/onnx works perfectly)
keen tangle
I want to download W okada for fivem? can i get help?
viral mason
What is fivem
viral mason
Does anyone know why once I convert model from onnx to trt, few classes stop bei...
What does class mean?
hidden sky
What does class mean?
Class/Label
viral mason
I still don't get it, I'm not gonna be able to help here
low shard
I want to download W okada for fivem? can i get help?
!howtoask
low shard
please elaborate
feral mural
When i pick an RVC sound to the app there are some explosive sound the app doesn't catch it.
viral mason
what
vast elm
yo how do i get the voice changer to work
!howtoask
viral mason
yo how do i get the voice changer to work
!realtime
viral mason
follow the guid to the one u are using, if it's neither of these it's outdated and uninstall it completely
low shard
When i pick an RVC sound to the app there are some explosive sound the app doesn...
!howtoask
low shard
elaborate