#✨│ai-help
1 messages · Page 331 of 1
Where is it Run install. Bat of Applio
The Prompt say Please run 'run-install.bat' first to set up the environment.
i mean my w okada sound or thing wont work anymore after a sudden before it works normally and 5070 and windows 11 and im js using it for normal purposes like changing voice to match the character in game
I assume this is for environment variables? Added them, and now its been stuck like this for roughly 6 minutes.
okay, perhaps building kernels
Where do you suggest that I start, and how?
You can convert safetensors files back to pth apparently, though you do need python installed, and pip install torch and safetensors
I'll share a script I found.
import torch
import sys
import os
from safetensors.torch import load_file, save_file
def convert():
input_path = sys.argv[1]
file_name, file_ext = os.path.splitext(input_path)
if file_ext.lower() == ".pth":
output_path = file_name + ".safetensors"
print(f"Converting {input_path} to {output_path}...")
state_dict = torch.load(input_path, map_location="cpu", weights_only=True)
if "state_dict" in state_dict:
state_dict = state_dict["state_dict"]
save_file(state_dict, output_path)
print("Done!")
else file_ext.lower() == ".safetensors":
output_path = file_name + ".pth"
print(f"Converting {input_path} to {output_path}...")
state_dict = load_file(input_path, device="cpu")
torch.save(state_dict, output_path)
print("Done!")
if __name__ == "__main__":
convert()```
Credit goes to: Codename;0
having the pytorch file format instead would allow you to use the same model in vonovox.
tbh im just trying to have the newest version thats all
You ever got help about W-Okada a year ago. Is your PC GPU still NVIDIA GeForce RTX 4060? And what do you actually use the voice changer for? I forgot.
What is your PC GPU? And what will you use the TTS for?

if u know any AI which can run like online then.. i know a way to get 3050 12gb vram pc for every 1 hr for free
@hallow thistle
u there?
NVIDIA GeForce RTX 3050 that has 12GB doesn't exist; there are 6GB and 8GB variants of desktop RTX 3050. Applio RVC has a built-in TTS feature.
whatever can i use it online?
This is a guide doc on how to use Applio RVC on Kaggle, an online website. https://docs.aihub.gg/rvc/cloud/applio-cloud/#kaggle
Last update: March 24, 2026
arigato
Thank you? You're welcome. 

Why? Applio RVC, as the name says, only accepts RVC voice models; the model file extension is typically .pth.
i need to use my custom voice ...can i use it on anyother AI
What custom voice? You need to be specific.
i have a voice...u can say like a free custom voice model though which u can use that voice model instead of using pre-installed models
im very bad in eng
T_T
No, but like please specific the voice model architecture name, something like GPT-SoVITS.
You can send your screenshot to here now.
its a voice

does anyone know how to hear themselves in vcclient?
Does anyone know how to stop my mic delay going up to 4000ms?
What is your PC GPU? What do you use the voice changer for? Which voice changer? And did you follow any tutorial or guide before?
Which voice changer?
Hello, why I got this error after uploading my newest model? usually it's not happend
is this the reason
I am currently building an AI-based recommendation system where user data (preferences, interactions, new content) is continuously updated in the database. However, because model training is performed only periodically, synchronization with the latest data frequently fails.
Additionally, errors occur frequently when comparing model outputs with the data stored in the database.
How can I resolve these issues? I would be very grateful for any advice or suggestions.
This is a General AI Discord Server, please elaborate:
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
if the user doesn't want to dm, don't push it
RTX 5070
Win 11
RVC for daily use
I use the latest Tg Develop's W Okada Fork
change that line to: cpt = torch.load(modelPath, map_location="cpu", weights_only=False)
should work
how do I edit the line on the console? never do that, sorry for asking!
or I go to the folder path?
I meant, you should edit line 189 in voice_changer/RVC/RVCModelSlotGenerator.py
just edit the file
you need to add the , weights_only=False part before )
and then run the app
yepyep, found it on the folder, I thought I must edit it on the console hahaha
do I make it read only after changing? or leave it as it is?
still getting the error after adding that line
what does the console say?
wait
when i m using rvc live voice cloning if i hve background noise like a laptop fan running will it impact in robotic voice or it will affect the vocie ?
weights_only should be, of course, set to True. Otherwise a specific set of objects that is alowed to load needs to be specified.
I'm trying to train a model, but it's not working. says: An error occurred extracting the index: need at least one array to concatenate
If you are running this code in a virtual environment, make sure you have enough GPU available to generate the Index file.
What did I do wrong?
hi
you did not provide a correct path to the dataset
did not preprocess correctly
did not extract features correctly
epic fail
look at the console, it says 0:0:00 minutes loaded

Sorry, I don't understand. So I change the false to true?
I just bought my new model 🥲
ignore what I said then, listen to Noobies instead
apparently you can reverse what i said previously
False or True doesn't seems to work
and I took the path
Oh 
I already try =True, but still error for uploading my model
let me try it again
scroll up
this is ```if ...
...
else
...
find all torch.load and add weights_only=True to all of them
Okay!
anyway, you're using outdated software that has not been fixed in a few years
I use this one that Namari gave to me on 28 March
Am I doing this right, there's only 1 Line with torch.load

okay
any free local ai that generates consistant images?
I want realistic images of a chicken, and I would want different poses for the same chicken
FYI, I don't use model from weights, I bought it from Booth
like my usual rvc model that I used for months
it may be a wrong model with code
but this one error after I bought it today
send it
Permission for DM noobies
yeah
any chance that such errors would appear with RVC v1 models?
ah, alright
but it wont if it is a non-rvc model or something weird
Ah, I give up... I'll try waiting for someone to make it for me... I don't know.
Why would you buy a model? Just make them yourself for free or use any from here
well, I'm just a casual user and use it for as it is. The quality that I bought is perfect for my daily use XD. I can't and have time to learn to make it myself
Turns out the model I bought is the only one that error when upload, the other model is fine
My usual model is fine also, just the newest one I bought
when i m using rvc live voice cloning if i hve background noise like a laptop fan running will it impact in robotic voice or it will affect the voice? can anyone give a suggestion
Heyy does anyon have a link to the RVC? I lost it
Are you wanting to train a model or use the voice changer rvc means "retrieval-based voice converter" not real-time voice changer
It could affect it if it's there the whole time but I think there's an easy way to remove that
like how using nvidia broadcast ?
Im trying to find the link to the Realtime voice changer so I can install it and use it
coulda sworn it came from here o.o
There's a website called MVSEP that is used for removal of background noise, reverb, music, ect
It's free
What gpu do u have (Nvidia or AMD) and what are u using it for? Just curious
Nvidia, and im using it for Roleplay purposes for things like fiveM
What is fiveM is it like D&D?
I'll get the links for u rq
not that i mean i hve an rvc model when i m using that rvc model in a live voice changer like vonov and my input has background noise like a fan will it impact the voice ? not for training the model when using it live .my input mic has noise
No its uh...for gta5, basically an online hub that hosts servers
hard to describe
this is the one where it wants me to download virtual audio cable yah?
Ohhh my bad I misunderstood, I'd suggest not having any loud noises like a fan or people talking behind you as it can really mess it up yea
You need one or else it doesn't work properly
But ye
The voice changer and virtual audio cable
any way to remove them for live use ?
It's recommended to get vac lite over VB cable as it causes weird issues
you sent me two links so what is which xD
Wdym? Just move the fan or smth
its laptop fan so if i put on eco the lag will be high
First is voice changer second is vac lite
oh is that why my voice would like cut out a lot?
sorry ik ur being tossed back n forth
My brain is melting as we all speak
Wait do u mean the voice u are using has a fan noise coming from it or
Btw might disappear bc grandparents
I'll be back soon trust
nah i mean my laptop fan is running so when i speak .there will be laptop fan noise with my voice also
@viral mason when you find some time, would it trouble you to help me refine my settings to make my voice smooth as possible? lmaooo
oh wow this is definitely not the same program I used before haha
the help thing doesnt even te3ll me how to downlaod the voice changer
Ok I'm back
Default voice settings are usually fine but if it's a female voice and you're a male go from 0 and start from 3 all the way up to 12 until it sounds good
Block size at around 0.30 and 0.50 is ok too but increasing it will cause more delay
I'm confused
What GPU do you have (Nvidia or AMD) and what are u using it for? just curious
pitch? is that for pitch?
It is yes
I'm using a voice model in Vonovox, but the output currently sounds bad. I'm using a Maono PD400X dynamic USB mic, and my laptop fan is running loudly in my room. Since the mic is picking up the fan noise in the background, could that be what is causing the voice model to sound bad?
Ohhhh ok ok
I get it now ty
Most likely it could be if it's loud enough the mic you're using would pick it up and interfere with the voice changer being able to output what you're saying
https://discordapp.com/channels/1159260121998827560/1403422896680079390 i m using this model
my main issue is that when I speak the voice changer cuts out, like I'll say "I want a cheese burger" it'll come out "I ant a eese-ger"
this is the output i get
What are you using that model specifically for? e-girl models are usually used for bad things like trolling or catfishing/scamming
nah i m using it for my yt vidoes
That's odd, would you know if anything is in your background causing an issue with the mic picking your voice up?
What kind of videos do you make?
the frst 1 is mine the nxt is the creators
shouldnt be no, but then I turned on the noise supression thing too
Try turning it off then checking again if that may cause the issue
maybe its the model like u said cause I was looking for an e-girly one, this one is taylor swift LMAO
If it doesn't then I'm not entirely sure
this is the output provided by the creator of the rvc model
E-girl models are used for catfishing/scamming normally, you did say you just wanted to use models for fiveM
this is my output
Starting to lose trust in the two of you
you play which rp server ?
Yes, I want to have an e-girl voice for that, one can use e-girl voices for more than scamming/cfing
bruh
<@&1159293140440723499> prob don't ban yet but keep an eye out
go check my yt channel
I don't use rp servers
can you take a look at the voices n tell whts the problem
No need to be rude
there is no need to ping moderators when you dont understand the things people use this for? not everything is a catfish or scam.
I agree that you can use them for more than that I just have zero trust of anyone who uses those voices
I'm mainly calling it for the other guy, but I do feel uneasy
bru the frst sample is mine
lord help me
it doesnt sound good like the other 2 samples which i got from the creator of this model . wht will the issue there
even without fan noise it sounds same
Idk the samples sound fine
okk
So do you still need help
I apologize, but yes if you think you can help i would appreciate it
It's alright
wht is the input gian required for rvc ? b/w -24 and -18db volume enough ?
id if you think it would be easier for us to hop in a call i could share my screen of the program, im just kinda slow when it comes to shit like this and its usually easier to show rather than tell
i mean when recording in audacity thats wht i m getting
I understand that, but I could send you the invite to the Vonovox server
I'm unsure what would cause the choppy voice thing that is happening
sure, appreciate it
Dms are fine yes?
ofc
👍
is this enough ?
I'll look at what I have then get back to u
okey thanks for the help man
I've sent you the invite ^^
got it thank u
yo need help
can u people stop getting hacked like a loser? <@&1159293140440723499>
3 in one day is just sad
so went i downlod the app i'm try to cancel the offer the 2.99 or 59.99 or 6.99 but i can't i want to chansel how??
huh?
do i need to pay??
what app are u downloading
ew that thing sucks and is a scam bro
i can't sent the pic
what gpu do u have (Nvidia or AMD) and what are u wanting to use it for? just curious
Which one is the real one?
?
what does this mean
if I don't know your gpu I cannot help you
not currently I am on a call
ohhh
check both gpu 1 and gpu 0
btw i can't sent pic. is AMD Ryzen 5 7520U with radeon graphics
ah ok
but.
so what are u going to use it for btw, like playing as Darth Vader or smth?
i want to prank my best 😂
?
sure
but which voice do i downlod??
ohhh
i downlod and now???
I want the app to put the one I installed and I don't know @viral mason
i'm trying to tell you
@viral mason
hello
can anyone help me with this plss
when i click on download button it says incorrect username and password
what are you downloading?
How do i fix this AttributeError: module 'onnxslim' has no attribute 'simplify'
what gpu do u have(Nvidia or AMD) and what are u trying to do?
can someone help me fix my issue downloading the voice changer vonovox?
What you have is very outdated, what gpu do u have (Nvidia or AMD) and what are you using it for? Just curious
Alr, is there anything in specific u were wanting to do with the voice changer or nah
What

That's now allowed here sorry
<@&1159293140440723499>
Btw before u do get banned I suggest buying a better PC once I get the money, 1050 isn't terrible but I would use maybe at least a 30 series or up
Use Vonovox, I won't be sending links but they're here somewhere
Not for the voice changer stuff no, I can't help since you wanted to do that weird stuff lol
Best voice changer for real-time, Nvidia gpu only

Vonovox is the name
RVC means Retrieval-based-Voice-Conversion
Not realtime voice changer
Are you trying to do TTS, ai covers, e girl trolling / Catfishing or roleplay?
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
What’s the best thing to use for offline/without using ur gpu voice training
My graphics card isn’t the best
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
just give me a link to the best offline voice trainer
that doesn't use gpu
*your
training with CPU will take ages
use a cloud service like Kaggle instead
There isn't a best one for everyone, you need to elaborate
something like applio, that trains without using ur hardware
webui
You have to either use your hardware or use the free limited cloud ones
You can't train out of thin air
That's why it's best to check your hardware first
Use @ helper role to get help about the voice changer.
If you have any dispute on why inferencing/training with only CPU would be better than having a dedicated GPU, here go for it and let me know what you think. https://huggingface.co/IAHispano/Applio/resolve/main/Compiled/Windows/ApplioV3.6.2.zip But if you feel lost otherwise, basically go for an online service at this point.
There are certain people who asked about the voice changer here; some of which asked the same simple questions, never come back or reply anything after, basically leaving me and Nick to ask nobody here, but that's really not that deep I guess. 

You have got help about Vonovox before. What do you mean Windows blocked the program to run?
turn off smart app control temporarely, seems to be a false positive
Try find some ways to fix that; either turning off the Smart App Control entirely, or whitelist the Vonovox and its batch file.
can someone give me a way to apply a bad mic filter? the voice changer quality sounds good but im trying to get a way to worsen the mic to make it sound more real
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
4080 super
windows
get a bad mic filter
didnt use a tutorial
Aside from audio effects. W-Okada or Vonovox? What do you actually use the voice changer for?
W-Okada, and im using it for real time voice conversion
i can send an audio clip so you can see what I mean by it sounds too high quality
im trying to get an audio effect that purposefully makes the mic sound worse
like I want it to sound bad
Experimental, roleplaying or something? Just wanna make sure you don't use the voice changer for something bad. Also, please be specific which W-Okada you're using currently.
roleplaying
how do I check the version of okada im using?
Why roleplaying?
because I do participate in rp servers on discord and I like to use voices for my characters
To check which W-Okada version you're using, look up at the top of GUI, it can be something like "v.1.5.3.7 onnxgpu-cuda" or b2332.
v1.5.3.15
That one W-Okada version is outdated, doesn't have a built-in FX option. Vonovox and Tg Develop's voice changer are only known ones to have FX built in. 
-realtime
Guides for Programs that use RVC Models in Realtime for Calls/Games
A Realtime Voice Changer with similar performance to Wokada Tg-Develop Fork, with extra features, but it supports only Nvidia GPUs on Windows 10/11 unlike other options that have wider support. and without cloud options
A personal fork (modified version) of Wokada Deiteris Fork, it just adds some Quality of Life improvements to it like supporting Spin Embedder and Audio Effects. Don't expect too much about it since the creator made it originally as a personal project.
A Realtime Voice Changer with similar performance to Vonovox & Wokada Tg-Develop Fork, with extra features.
Deiteris' fork (modified version) of wokada that doesn't get updates anymore.
These options are not recommended for use.
Not suggested, older versions in youtube tuts are even way worse. GUIDE
The program is worse compared to the ones above, and much less updated. GUIDE
Whats the best way to find the best settings for your vocie and voice model
For voice models, look up in #1175430844685484042.
Yea but the settings inside the app
whats the best way to tweak those
just trial and error?
What is your PC GPU? Which voice changer or version are you using? And what do you use the voice changer for?
On Vonovox, try set crossfade duration to 0.15, extra time as 2.0.
im trying the tg-develop fork and it actually sounds worse than it did with original wokada
Do i ened premium?
All it says for me there is block size
The "Premium" one is optional, not needed unless you want to support the author with money.
Just asys block size tho
New design?
Is it a bug though? I don't know, I can't identify the issue from your words alone. You can either try report the bug in Vonovox's official Discord server, the invite said somewhere in Vonovox guide doc on AI Hub website.
What do you mean by new design?
Namari is there abetter voice changer then vono vox?
A new design for the voice-changing app? Or is this a different Chinese program? By the way, is the 5070T a powerful enough graphics card for voice-changing?
Bud, that's Vonovox, another voice changer program that uses RVC voice model. It's not W-Okada.
VCClient Yes, but he eats a lot
will it make my voice changer sound better, and is a 5070 good anough?
1.6.9 is a stable version of Vonovox but that's about it. There's Tg Develop's voice changer.
Please, please listen to me.
I can't say for sure, but on the 3060 it worked about half the time.
So will vonovox 1.6.9 sound better then 17_11
WIll it sound better if i use vcclient?
No. Vonovox 1.6.9 is basically a downgrade of Vonovox 17_11; the 17_11 one is a beta one.
What do u think abiout vcclient?
VCClient is literally W-Okada but older version.
This is the guide doc for Vonovox. https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
Last update: March 30, 2026
What makes Tg Develop's W-Okada fork to sound bad is when you didn't set extra to 2.7 s, you only left extra default to 0.5 s.
Paradoxically, both Vonovox and W-Okada forks have different trade-offs.
iF I get tahat extra time and crossfade duration will my voice changer sound alot bhetter
You should try test by yourself, that's the answer. 
yea u gotta test a lot
These 2 sliders are removed in latest vonovox beta.
That's why @white cairn couldn't find it
They work differently in vonovox than in w-okada so the values shouldn't be directly compared
Anyway, the sliders were removed because in Vonovox there's no reason to adjust them anymore, so everything is fine
whats the best voice changer
What's up, Demon Slaying profile user. What is your PC GPU? And what will you use the voice changer for?
um i dont know my gpu but using voice changer just for fun
To check your GPU name, open Task Manager. 
ah gpu is graphic
Full name?
why i should ?
Because you should. W-Okada and Vonovox work best if there's a dedicated GPU; otherwise you'd expect them to run CPU only.

???????
actually i just need a voice changer where i can import some voices and it need to be good the quallity
The full name of one GPU is something like AMD Radeon(TM) or AMD Radeon RX 9060 XT, this might give you an idea. On Task Manager, go to Performance tab, spot GPU 0 or GPU 1, click either one to reveal its full name on right panel.
amd raedeon tm
Anyways, that's an integrated GPU. Not really recommended to run either voice changer with that GPU, you know.
Am I supposed to know? 
what app? this is a general ai server
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
What model gtx 1650s 4gb vram super can run?
Because some people here have been telling me to go for Vonovox 17_11 instead of 1.6.9, so I finally did. This is the same laptop that has no GPU by the way. This time, I tried to recreate the error running from Downloads folder, but then Vonovox opens just fine.
How does it work with no gpu?
Mine is just a dummy interface with no actual function, so while Vonovox would open anyway, the actual voice changing might not work.

Ah ok 
Why are u using it in Roblox? As long as you're being a character like Goku or whatever it's ok
Roblox has a lottt of weirdos
worm canu help me
How far will my AI voice model go with 4 seconds data set😭😭😭😭
Will it be some what decent
e girl trolling / catfishing?
get a bigger dataset
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
please elaborate:
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
probably not even possible, the least amount of audio I've ever been able to train is 32 seconds
What if I loop it
I don’t got more but what if I loop it
it wouldn't be able to speak well probably
Will the voice come out good?
what is the 4 second audio anyway?
And on a different voice thing there’s some guy that posts studio sessions but they got hella background noise what if. Use them
It’s a take I got from a session
just remove the background noise with MVSEP
I tried didn’t work
oh
The beat is bleeding from his headphones and it’s too quiet I guess
There’s food MJ models here tho?
if it's reallllly quiet it might not impact the model too badly
Good*
It is noticeable with no headphones
yea there's plenty, not all of them are good tho
Yea
oh dang
Does applio know how to deal with that or I can give up
not really it just trains whatever u have
l
help me
wasssup
The app is on a white background and doesn't load further after I selected my voice (file)
[VCClient] wait web server...10 http://127.0.0.1:18888/
[VCClient] wait web server...20 http://127.0.0.1:18888/
[VCClient] wait web server...30 http://127.0.0.1:18888/
[VCClient] wait web server...40 http://127.0.0.1:18888/
[VCClient] wait web server...50 http://127.0.0.1:18888/
[VCClient] wait web server...60 http://127.0.0.1:18888/
[VCClient] wait web server...70 http://127.0.0.1:18888/
[VCClient] wait web server...80 http://127.0.0.1:18888/
[VCClient] wait web server...90 http://127.0.0.1:18888/
[VCClient] wait web server...100 http://127.0.0.1:18888/
[VCClient] wait web server...110 http://127.0.0.1:18888/
[VCClient] wait web server...120 http://127.0.0.1:18888/
[VCClient] wait web server...130 http://127.0.0.1:18888/
[VCClient] wait web server...140 http://127.0.0.1:18888/
[VCClient] wait web server...150 http://127.0.0.1:18888/
[VCClient] wait web server...160 http://127.0.0.1:18888/
[VCClient] wait web server...170 http://127.0.0.1:18888/
that's super old
I'll get u the newer thing
what are u using it for btw, like playing as Goku or smth?
I watched a video titled 'SPEEDRUN ON VOICECHANGER LIKE BRATISHKIN' (this video is in Russian - https://youtu.be/rLIzH6CcIUg?si=-yUfwfXccV1BByzg) and tried to do it like he does, but it doesn't work for me.
what should I download from this?
?
If you need urgent help, please:
- Checkout our AI Hub Docs.
- Ask for help here elaborating your issue and following the [Guidelines](#1402790586028789830 message).
THIS IS IMPORTANT, IF YOU IGNORE THIS YOU WILL GET LESS CHANCES OF BEING SUCESSFULY HELPED SINCE WE CAN'T HELP WITHOUT INFO
which one to run, start.bat or start_hidden.bat
both, for vonovox run start and for vac lite run setup64 then install driver
stop getting hacked it's so obvious #✨│ai-help message <@&1159293140440723499>
those models are usually used for scamming or catfishing, why would u need one
die
<@&1159293140440723499>
i saw it on a yt vid
that it was used for trolling
yea that's weird and not allowed here
you could use something like Goku, or Darth Vader, Omni man
literally anything but that weird stuff
yea that's disgusting, this server isn't for that
it's for general ai talk as well as a server that has many voice models
ohr fairs
Anybody got a model of The Storm King (My Little Pony: The Movie) from Weights before it shutdown?
so what happened to all the pretrained voices they are like gone?
i have a old computer but what would be good settings for vonovox with Intel(R) Core(TM) i7-9700 CPU and RAM 16.0 GB
right here
vonovox can only run on nvidia I'm sorry
i have nvidia
anything intel really can't run this stuff at all
hollup
thank you very much I seemed to have been blind I appreciate it!
ur welcome :3
i have nvidia geforce GTX 1650
o
in that case then yah it can run it
so what are u using it for btw, just curious
like goku or Darth vader?
o I dont even have a model yet idek what to use
To talk to my friends in voices
alrighty, do u have the most recent beta and vac lite?
Whats that...
oh Virtual audio cable?
ye
whats the beta part
and yeah i have recent
kk
Vonovox 17_11 (or 1.7.0_BET) is the most recent beta version. Vonovox 1.6.9 is the recent stable release but older than 17_11 one.
if u need help or anything just message here btw
i guess i need help
What help?
i wanna start learning AI and all
and to be honest, I am really new to this Discord thing
AI means artificial intelligence.
Oh yeah, I know that
i mean i wanna learn how to use it for and good and making content
i mainly want to make content using it
There are many ways to start.
are you on telegram?
No.
I texted you check
I use Adobe Premiere Pro to edit videos. Not so strictly AI related, but I use the software to edit videos. 
yeah i have heard of it
i heard that it is the best software for editing
but i also head that it is very complex to use
Not always complex, I just know how to use it.
I didn't follow any tutorial. I open Adobe Premiere Pro, and learn to use features at my own.
Should we move this conversation to #🧬│ai-chat or something? This one doesn't look like a help-related topic at least to me.
yeah okay how do we do that but?
What do y'all think about how I launch both Vonovox and W-Okada voice changer fork at the same time? I know I won't get any answer, but I do this in case I run out of idea to help people about voice changer.
I run both simply because Vonovox doesn't have a passthrough feature so I use tg fork for it
Vonovox is for the ai tg fork is for the real me
The specific setup I have virtual cable wise doesn't allow easy mic swapping
The .py file doesn't run because there are no libraries.
Hello
Why?
start.bat also doesn't start, that is, it just writes "launcher.py"
?
That's where it froze, and that's it - C:\Vonovox_beta_17_11>runtime\python.exe launcher.py
help me pls
Try extract to somewhere like "D:\Vonovox" or "E:\Vonovox" and run the batch file from that. Also, don't run as administrator.
Oh, it just takes a long time to load, I am using Vonovox_beta_17_11
how to set it up
This is guide doc for Vonovox. https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
Last update: March 30, 2026
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
Having trouble with Applio AMD.
Gpu: rx 6750 xt
CPU: 5800x3d
rocBLAS error: Cannot read C:\Program Files\AMD\ROCm\6.4\bin/rocblas/library/TensileLibrary.dat: No such file or directory for GPU arch : gfx1031
rocBLAS error: Could not initialize Tensile host:
regex_error(error_backref): The expression contained an invalid back reference.
where can i use the voice models
I dont understand the guide doc for vonovox
just download QWEN 1.7b locally in ur pc it sooo good af
Hello, how can I fix the robotic voice issue? I'm running version 1.6.9. At random times, the AI decides to lower my voice, making it sound squeaky. 5070 Ti + R7 7700
what do you need it for
if you mean vonovox, then try the most recent beta in the first place, it's much better
The beta is available here: https://huggingface.co/dr87/vonovox/resolve/main/Vonovox_beta_17_11.zip
(if you want to use index, it also needs to be patched with the following command: runtime\python.exe -m pip install faiss-cpu)
but in general, weird robotic sounds etc. are usually produced by the model itself
and no program will fix it
downloaded after this
if im making my own voice model and it sounds robot what's usually the cause? i think it was like 8 mins of talking could it be not enough maybe background noise or could it be any of it?
Is the first generation in applio generally suppose to take long?
Is it possible to set a longer delay? Or is there a way to get rid of this? Because I don't need it to bbe real-time.
if you don't need realtime, why use the realtime app at all? You can use Applio instead and run inference on audio files. It will work much better than realtime apps
I need this for voice messages. But thanks for letting me know that’s possible! By the way, the beta version works pretty well. Thx
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
could you explain what you don't understand?
which qwen version?
This is a General AI Discord Server, please elaborate:
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
then you should use Applio and not a realtime program like Vonovox
Is there a good AI bot that you can consult for help with gaming? Like a companion who helps you learn and get better? Everytime i ask anything on official discord or google, i get very different results.
I learn to play Minecraft at my own.
Yeah but i'm not talking about Minecraft, i'm talking about something like HOI4 or Frostpunk
I need a companion because i dont have anyone whos human to speak to for help
Does first compilations usually take long
Where can I get the voices I want?
Look up in #1175430844685484042.

Anyways, you're banned.

hello, do you need any help?
im trying to figure it out xd
Figure what out
i need help with finding a good ai with no usage linit that i can download on my windows pc and can control my computar
Just how it all works but i think im getting an idea of it
^^
How what works silly
how what works? because this is a server for general ai
there's no ai without usage limit completely free unless you run it with your own hardware
perhaps you could try OpenClaw with local LLMs
just be aware of the security risks
My bad sorry
i can install it into my device
What's your PC GPU btw?
x1 85
We'll never know what Samse is trying to do because they have yet to say it
trolling
what kind? 
you need to download the set of libraries for 6700
Push to talk for W-Okada?
are u looking for the voice changer
im having a strange issue with the app.. it works for ~20 seconds and does not change my voice. and then the application becomes unresponsive for some reason... no edits i make are applied..
i have an astral 5090
9950x3d
64gb 6000 mz ram
duno what the issue is
is it an amd gpu? i've not heard of this
Guys, I can’t train the voice model on Applio’s Colab it’s giving me an error. What should I do?
this is nvidia
rtx 5090 is what u use right?
yes
oki, what are u using the voice changer for?
personal reasons, if its required id prefer to say in dms
as long as it's not for egirl trolling
opposite actually
are image perms given in this server?
would make it a lot easier on my end
this is the version i downloaded
this is old
wdym
which one do i download then?
^^^^^^^^^^
send in dms then ^^
sent...
whats the up to date version then?
these two links right here
i read in the guide about a specific 5000 series version of the app?
or am i miss reading it
the beta works fine with Nvidia
not running....
Vonovox, what are you wanting to use it for?
that's not allowed here sorry
<@&1159293140440723499>
egirl trolling is related to scamming/catfishing
that's why
im noticing the delay is very low and not much hardware is being used.... using here (red mark, start and end)
voices sound very low quality..
an example is here https://discord.com/channels/1159260121998827560/1359651876924293160 (sound file here) but more robitic/low quality. Ive tried multiple models and im kinda lost on how i can make it work... I feel like its using no hardware and thus sounds bad
Wdym? I'm following through the applio guide on the ai hub.
It launches and I'm able to browse through
but when I try to convert an inference, it's stuck on processing compilation
if you're using zluda guide, you should've downloaded a set of libraries for gfx1031
did you install hip sdk 6.4.2?
Yes
Is this the hip sdk or a completely different thing I gotta install?
yall need to make a bktherula voice model
Ignore the the dll file and only use the lib folder right?
use both
where does it say ignore?
in hip_sdk / bin replace rocblas.dll, then go into rocblas folder and replace the library folder with what is in the archive
that is a little different, but same overall
dont need to reboot, just restart applio
I restarted the entire process of installation. Do I use hip62.bat or hip64.bat file?
hip64 right?
hip 64
do not re-run patch 64 without re-install torch again
ok
Is this okay? It says system cannot find the file specified though
because there was no old zluda folder, all good
did you add hip sdk /bin to the Path variable?
Yeah\
there should be _6.dll
make a copy with _7.dll
and delete the _6 dll?
no
so it's fine to have both dlls there?
yes
are you running two instances of applio?
Once Applio RVC restarts, try inference another model and see if problem persists.
hmm tried another model but it still persists
going to try a newer model
Yeah same result
show the error
can you open the .pth with winzip/7zip?
They came in a zip
I just used whatever the first one was in the voice model channel
oh
so you placed the unzippped pth into applio/logs?
yes
I extracted the zip the download came with and put it in the pth and index into applio
this is super weird
can you show a screenshot of unzipped files?
I'm redoing in the install to check if there was anything wrong within the cmd for any of them
does anything look off here?
Should I set the path for that? I was only aware we only set path for hipsdk bin only
no. it is fine
Guys wha are we doing with our life rn🤣😭😭🤣🤣🙏💀
Rn I'm trying to sleep
I like sleep
up
what app? This is a General AI Discord Server, please elaborate:
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
thanks for reporting
For real, actually? Well, in AI Hub Discord server, I've been helping people here on how to use the voice changer, RVC and thing things. I don't really like to keep certain knowledges to me, I'd like to share for others to know.
I know I might sound like I'm writing a mail to someone. Besides that, I make some voice models, some of which are funny and experimental, but I always make sure those models I made should sound good to satisfy those who use them. I may be accept too many requests in #1159289738314919936, I know I promised so many people genuinely, but then I do try my best to complete them, anything takes time.
What is your PC GPU? Are you following tutorial or guide? And what do you use the voice changer for?
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
Hi! I need to find a completely uncensored ai chat bot that can use anchor image does anyone have such tool
Why?
is there anyway to do voice change (not locally) thats not through a scam website :/
Sure. There are Kaggle and Google Colab. But what will you use the W-Okada for?
I want to put an artist's voice on a different song.
There's Applio RVC.
-rvc
I looked for the cloud option... It's giving a 404 error
Last update: March 24, 2026
This is the most confusing shit I ever seen in my life.
Let me help
Just follow this video
@vital vessel
It's kaggle
Hello, nice to meet you. do you need developer? Please send me DM.
<@&1159293140440723499> hacked account
i'm already in love
if you plan on trolling it's not allowed here
Shes mine you cant have her 😡
we can share
hmm sure 🙄
you know what I mean 
hey theres another website like weights?
ai influencer
what was the name of the voice changer again?
Whenever I run w-okada it just makes a log and doesnt open does anyone know why?
it depends on what gpu u have(Nvidia or AMD), what are u using it for btw?
is there anything better than okada?
vonovox, it only works on Nvidia gpus tho
oh , ty, is easier than okada or not?
I would say so yea, the current beta only has two settings you would need to mess with, Block size (controls delay) and the pitch
nothing difficult
ok , i will try, ty for the help
is it the same as on the ai hub website?
it's a little different than on the guide as the current vonovox guide is about the latest non beta version
but the beta better
That sounds awkward.

There are Vonovox and W-Okada.
Not mine.
Which Vonovox version are you using?
when I run start_http.bat for wokada it just makes an empty log and doesnt open does anyone know how to fix this
hi this is my 3rd or 4th tme trying this, but im trying to install the ai voice changer. Each time I tried it seems to not be able to function correctly. I use windows 11, and my cpu is a ryzen 7700x with a 4070tis
is this related to NSFW?
there isn't just one
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
are you tryin gto do e girl trolling / Catfishing?
This is a General AI Discord Server, please elaborate:
- your pc gpu
- your pc os
- what are you trying to do: TTS, AI Covers, E Girl Trolling / Catfishing or Roleplay
- the tutorial link used
Is your gpu the 4070? Use Vonovox
What link are you trying to use and what are you wanting to use it for, just curious
what do i do?
Why?
?
Hey, I haven't been on over a year.
Is KLM pre-trainer still getting any updates?
Or has the dev stopped?
Or are there any good pre trainers?
use legacy core 1.5, or 1.6
1.6 is best at more higher pitched voices and such but struggles a lot with rough gravely voices like Batman or Kratos
legacy 1.5 is good with that kinda stuff
pretrains can be found here https://discord.com/channels/1159260121998827560/1235952130855010365
Which pre-trainer would you suggest for female singers?
legacy core 1.6 should work fine, if you dislike the results you can look at legacy 1.5 as well or just use og (default one used)
I’m trying to clone a voice, but it’s been a long time since I last did it, and I’m not sure which 40k pretrain model is the best nowadays. If anyone knows, please let me know. I’d really appreciate it.
Im using UVR5 eddy and im getting The system cannot find the file specified this error
Hi, how are you? Sorry to bother you, but I wanted to make an AI cover of a song, using an AI voice model of Caine from The Amazing Digital Circus, for a meme with some friends, and due to a lack of safe alternatives, I ended up here. I would appreciate any guidance on how to make this AI cover, if it's not too much trouble.
Why? 
Why are you saying like you're the problem here? To do AI cover, there's Applio RVC.
Base GeForce RTX 4070, Ti or Ti Super? Either NVIDIA GPU would work with Vonovox.
Hey so, I have ran into an issue I am sure I have not done something to cause this, but when I go on discord or roblox my voice goes through, and after that the voice changer it self, I am using Tg Develop W-Okada
hi everyone ik that this may be a dumb question but what is the best ai (as of now because ai keeps evolving) to run locally for coding, then the best for just like chatting (like gemini), and then for videos and pictures
Download Vonovox. https://huggingface.co/dr87/vonovox/resolve/main/Vonovox_beta_17_11.zip
Can anyone recommend which program is best for logo generation right now? (Midjourney, Grok, Chatgpt etc?)
elaborate
Try to sell PC part online, notice a lot of scams when selling components. Any advice to avoid them
What gpu do u have? (Nvidia or AMD) and what are u using the voice changer for? Just curious
I have a geforce 1650 super (its old ash ik) and I jus use it to troll with my friends
I fixed the issue with just disabling a settings in the mic anyways
What gpu do u have (Nvidia or AMD) and what are u using it for?
What do u mean by trolling btw?
Like Goku or Darth Vader or smth?
we find diff goofy voices and just talk
to people
Ok good good, you're safe
I want to make a crazy project which will be fully made with ai is their anyone who is ai specialist and can help me related to that
Sooo the problem may just be your gpu, I've tried having a friend of mine download and use any voice changer on this GPU and it didn't recognize it but I can still give u Vonovox to try
nah its all good after I disabled listen to this device
it fixed
Oh
U should still be using Vonovox it's better than tg fork
Unless you tried it and didn't like how it sounded
I'm not really sure tbh but you could always try if you want
u thought I was using it to sound like a girl didnt u
I aint weird 😭
Dude
You don't know how many of those weirdos I have to talk to every day
It's sad
Any reason you need Vonovox in particular or
No e-girl models right?
Where does bro keep going 
I value my ideals I aint weird
but I can see why
you would ask
people r freaks
Yea what you're doing is normal
Hi everyone :), im new :D
I appreciate you being one of the good people using the voice changer
Hello! Welcome:D
This is a help channel do u need any help?
🫡
we voiced over a dragon ball episode it was so hilarious
why cant i send an image?
sometimes the ai would bug out so u would jus hear a grown ass man crack his voice mixed with the ai
enchancer
User: hey
AI: Hi!, what can i get for you?
User: i have some question
AI: Hello!, How can i help you today?, do you got any qustion?
User: what is 1 plus 1
AI: Hello!, do you need any help or something?
User: how are you?
AI: I'm feeling... computational! And pretty good about it. I'm all about the "hello world" feeling. It's a cool vibe. How are you doing today?
User: are you dumb?
AI: I'm not dumb, I'm a simple AI who can see the world in just a few seconds. It's a skill that I have. I'm Zin AI, created by Ytiomar Dev. I hope you're doing well.
User: who are u?
AI: I'm a human with a purpose. I'm ready to help and willing to learn. I hope you're enjoying your stay here.
Btw why is my AI kinda dumb?
Image perms require being leveled up or limiting time access from a mod (usually only done for help)
U can level up by talking
ok
😭
The ai thinking it's a person is kinda funny, but what one are u using?
Idk why it didn't answer your 1 + 1 question
Everything is funnier at night
You seem like you're in a rush and it's hard to communicate with limited words but ok
Just don't use it for weird stuff
Did you mean to link the actual model or the voice models page?
Try make an AI that could play Minecraft.
Umm like how
I am making mods and builds with ai currently
Both.
TBH an AI claiming to be a person is another kind of potential scam
I'm pressing separate and it gives me a Roformer separation failed
Here
Hi does anyone have some advice about what setting should i use on the vonovox basic effects (noise gate, eq band 1, eq band 2) to improve quality?
They don't improve quality. Noise gate is for cutting off noise during silence, and EQ is for slightly adjusting the sound to what you like.
The source of the "quality" per se is the model
Best you can do with EQ is boost/reduce some frequencies, e.g. add more bass, or highs etc
@low shard need help here
<@&1159293140440723499> freaky hacked acc
<@&1159293140440723499> ban the weirdo
oh my days
<@&1159293140440723499>
The voice changer isn't made for that purpose
That's weird
its already gone, they specifically dmed you
make sure to read when those messages were sent, I already knew he was gone since u got him earlier and told me about it
someone having blue screen on nvidia w okada
?
So since weights is down as a website, did the RVC models get moved somewhere else? (Sorry I've been away from here for so long_
The RVC models hosted on Weights were deleted, so some people were trying to download them before the site closed. However, it’s worth noting that most of the models on Weights were just re-uploads from this server.
Can anyone recommend an AI for manga cleaning, specifically for removing text? I tried using NanoBanana, but it doesn’t preserve the original resolution. I’m working with 2040×2880 images. Is there any local editor available?
Elaborate
weights isn’t temporarely down, it completely shutdown on the first of april, and everything got deleted
You can find models in #1175430844685484042
@marble sky don’t promote
Is this a real server?
no
weights is dead
this is someone hijacking the name
i have it
really?
yes
link maeko (ajthefunky) voice rvc on discord
no i dont feel like it sorry :(
please
no…
i want to have maeko voice on weights.gg: https://www.weights.com/models/cmi286s1801x70ab6nxkg19u4
i want it now
please @proven hill
you have to wait
okay, now please faster @proven hill, i want maeko rvc on weights.gg in replay
I'm looking to run MentaLLaMA-chat-7B on llama.cpp, but I can't find a pre-converted GGUF version online
faster?
click
link to screenshot
yes!! exaxtly
whats the command again for the voice changer thing
like to bring up the embed with links
-rvc
found it
really @restive kiln ?
Aye, just having a bit of trouble, it's voice isn't returning

