✨│ai-help

[abstract comet]

I’m looking to learn get ad much info as possible on the latest rvc and SVC

[stoic spire]

I had this same problem and the only solution I found was to remove this line

It is in the “Setup runtime environment” cell

[final sierra]

Hello, I just have a question. Right now I’m using Vonovox and I’d like to know what’s best for Discord RVC and in-game use between Okada and Vonovox. I don’t know much about it, sorry. What I use it for is actor voices in all languages, including male voices (deep and high-pitched) and female voices.

[hardy yew]

-rt

[patent trellis]

💻 Local Realtime RVC

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

• Vonovox

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 GUIDE

• Wokada Tg-Develop Fork

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. GUIDE

• Wokada Deiteris Fork

Deiteris' fork (modified version) of wokada that doesn't get updates anymore. GUIDE

⚔️ Wokada Tg-Develop Fork vs Vonovox

For Windows Nvidia, Both Wokada Tg-Develop fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Tg-Develop Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[hardy yew]

There's a "Wokada vs Vonovox" section in the message above. In general, the performance is supposedly very similar, what differs is the interface and some additional functionality. You can go to the links for details

I'm in opposite scenario, currently using Tg-Develop Wokada and I'm wondering whether to switch to vonovox. But it seems like there's no need to jump between the two

[hallow thistle]

This is Tg Develop's W-Okada fork. https://cdn.discordapp.com/attachments/1159290139609137264/1463078807937482817/image.png?ex=6971d752&is=697085d2&hm=4f89a93c83c412a36812ce9b5f7db4ae24a6ebc7399e911bf1dbfbbaf9fc7e03&

[final sierra]

I was wondering because my presses are a bit weird, I don't really know if you have to take one specifically in your language or if you can take one in all languages. because sometimes it's buggy, it's too thick or even it picks up background noise

I also don’t know whether the virtual cable affects how people hear me. I just took the first virtual cable I found and called it a day.

[hardy yew]

Recommended virtual cable is VAC lite (e.g. ver 4.70, there should be a download link in one of the guides), if you have VB Cable I heard there might be issues.

In terms of background noise, I think noise suppression is a common feature between both RVC apps.

In terms of pronunciation and language, depending on the accent and the sound of the language itself the model's influence can be quite heavy. For example Polish language has various sounds that don't exist in English, so the English accent bleeds heavily into my Polish speech and makes it weird. Models trained on Polish data perform much better in that context, obviously. So I'd say this depends on the differences between trained and target languages.

• speaking clearly will always work better than quick mumbling etc.
  It's worth paying attention to what the model handles poorly and avoid it to ultimately make the output sound more genuine. Lowering the index rate also helps in biasing the speech more towards your input, while causing some loss of the trained characteristics.

[ashen patrol]

Why did you have to cancel weights

Tell me why

[analog obsidian]

i can tell you in dm because mods rlly hate when i even dare to mention any other server but this lol

[hallow thistle]

Understand that I'm a helper staff member, not who manages Weights website. There are many potential reasons why Weights is about to take down sooner, though one of known reasons is simply about "money/bankrupt".

[abstract comet]

Hey guys
I’ve trained a model on 48k using hifigan but I am having an issue in that it’s sounding too smooth and not like a real voice or singing

[frigid stone]

Could I also get a DM of the other server? I only know about the RVC Developer server that shows up on Google, but it's not very active

[abstract comet]

https://tenor.com/view/hello-hi-hy-hey-gif-8520159980767013609

[analog obsidian]

check dms

probably you have my dm in the spam folder

[analog obsidian]

check dms, spam/requests folder

[abstract comet]

How did you manage to fix your muffled thing

[frigid stone]

Like I wrote, I'm pretty sure it was the audio quality of my dataset

[eternal harness]

pipeline not installed

[ashen patrol]

WHAT??? are they gonna come back soon

[viral mason]

Ask in the official weights discord 🤷‍♀️

[final sierra]

Wow thxxxx

[turbid blaze]

-rt

[patent trellis]

💻 Local Realtime RVC

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

• Vonovox

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 GUIDE

• Wokada Tg-Develop Fork

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. GUIDE

• Wokada Deiteris Fork

Deiteris' fork (modified version) of wokada that doesn't get updates anymore. GUIDE

⚔️ Wokada Tg-Develop Fork vs Vonovox

For Windows Nvidia, Both Wokada Tg-Develop fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Tg-Develop Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[magic lynx]

guys what is this 😂

[simple ore]

you're trying to use a contentvec model with spinv2 embedder for inference, which as expected does not work

[magic lynx]

why so muffled tho

[final sierra]

What's the best version of Vonovox? I'm on 1.6.9

[viral mason]

the beta is the current best although it's still being updated which means it could be unstable

[simple ore]

contentvec modes does not understand spinv2 phonemes so it produces garbage output

[magic lynx]

Thanks

[real depot]

Hi everyone!
A quick question out of curiosity:
I noticed that, depending on the size of the dataset, the index file ends up being larger or smaller. The smaller the dataset, the larger the index; the larger the dataset, the smaller the index. Why does this happen?

[lusty holly]

for the voice changer whats the best option for RTX 50 series

[viral mason]

vonovox

[final sierra]

Do you have a link? For example, the download and changelog, if you don't mind.

And what do you mean by unstable? Have you had any problems?

[viral mason]

it's recommended to go to the official discord server for vonovox there's too much tto explain

[final sierra]

Thank you so much, I didn't even know there was a server, I'm just starting out in this world, thank you for the help

[hallow thistle]

That's not how it is, unless they gonna make a U-turn of their site but with limited basic features come back.

[quiet axle]

Hi i was wondering if its possible to run vonovox on a 2nd pc?

[full wadi]

Hello, I'd like to ask is there a way to make W-okada Voice changer run smoothly on AMD, I've been experiencing alot of delays despite following tutorials on how to make it run smoothly

[bold venture]

Did you try the Deiteris one? I'm on AMD and it worked for me

[full wadi]

Can i DM you later for help, I've never heard of it before

[bold venture]

i can send you the link

[hallow thistle]

What is your AMD Radeon RX GPU?

[hallow thistle]

Yes, you can run Vonovox on any PC, as long as that one PC has NVIDIA GPU. The question is, how will you share Vonovox from one PC over LAN like what W-Okada voice changer does?

[full wadi]

AMD Ryzen 3 2200GE with RVG

[hallow thistle]

"AMD Ryzen" is an AMD CPU. What does "RVG" mean?

[full wadi]

Radeon Vega Graphics

[hallow thistle]

"AMD Radeon Vega Graphics" refers to an integrated GPU found in some Ryzen system, unlike AMD Radeon **RX **which is a dedicated GPU. So, while AI program like W-Okada voice changer, especially its DirectML variant, might work with an integrated GPU, the performance will be lower or the fraction of your CPU.

[full wadi]

What does that mean

Was that why W-Okada has been constantly stuttering on me

[hallow thistle]

Likely. The older W-Okada version (like v.1.5.3.18a) DirectML is very buggy for AMD/Intel GPU, it would likely fall back to CPU instead. Newer W-Okada forks from Tg Develop/Deiteris can detect and use your AMD GPU but that's about it.

[full wadi]

I See

[hallow thistle]

You can try W-Okada fork from https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/ or go for an online cloud service for more performance.

[full wadi]

i did install tg-develops but im having a lot of stutters and delays with it but it wasnt as bad as without it but im still having issues

[tepid moth]

yo, so im just tryna get help installing the right versions and then setting it up

thats about it @hallow thistle

[hallow thistle]

!howtoask

[patent trellis]

❓ How to Ask for Help

✅ Before You Ask!

1. Check Docs & Guides: Your answer may already be in the AI Hub Docs or the https://discord.com/channels/1159260121998827560/1159513888199540817 channel.
2. Search the https://discord.com/channels/1159260121998827560/1192011222023950368 : Look for existing posts that solve your issue. Do not invade someone else's post.

📝 How to ask?

Tell your:

• Full GPU Name: (e.g., NVIDIA RTX 4060 8gb vram desktop)
• Operating System: (e.g., Windows 11)
• Detailed Description: What were you trying to do and what went wrong?
• Tutorial Used: Link to the guide you were following.
• Screenshot: A picture of the full error message.

🚫 Prohibited Topics (We Will NOT Help With These)

To maintain a legal, safe & ethical community, we will NOT provide help for:

• ANY illegal activities.
• NSFW/Porn.
  Requests for these topics may be ignored, not helped and result in moderation action.

<:matsuripray:1159685390156967936> Community Expectations

• Be Polite & Patient: Our helpers are volunteers. You may ping the Helpers role once.
• English Only: Please keep all conversations in English.
• Don't Ask To Ask.

[tepid moth]

windows 11,
AMD Radeon (TM) graphics
Im trying to install the software itself and then set it up without messing anything up or breaking anything.

i didnt use much of a tutorial, i just joined the server and downloaded: https://github.com/deiteris/voice-changer/releases/download/b2332/voice-changer-windows-amd64-dml.zip
it wont let me send any screenshots here, so yea

[hallow thistle]

AMD Radeon(TM) Graphics is an integrated GPU found in AMD Ryzen systems. See #✨│ai-help message on why integrated GPU might not fit for a voice changer.

[tepid moth]

aw dangg

well

it worked before

if that helps

but now its not

[hallow thistle]

You could get W-Okada voice changer fork to work with an integrated GPU but that's about it. For full performance, an online website is considered.

[tepid moth]

Yea, im using theonline website rn lol

i could dm u the screenshot unless theres a channel here dedicated for media

@tame oracle gurt

[tame oracle]

wgat

[tepid moth]

gurt: yo

ts tuff guys trust

[tame oracle]

i dont follow but okay

[tepid moth]

idk im spewing nonsense

[hallow thistle]

Are you follow this https://docs.aihub.gg/realtime-voice-changer/cloud/tg-develops-w-okada-fork-kaggle/ guide? Because this Kaggle is more recent than Deiteris' W-Okada fork.

[tepid moth]

nah im using uhh

honestly idk what im using

but il try to use that one

nevermind xd

wait okay whatever im using rn which is the green anime girl looking thingy (idk what to call it) is changing my voice

i gotta adjust the settings i think and then figure out how to output it to dc

[full wadi]

i just installed deiteris and i cant hear my voice

[hallow thistle]

This image is "Tg Develop's W-Okada fork". The Deiteris' W-Okada fork (b2332) has a different interface. https://cdn.discordapp.com/attachments/1159290139609137264/1463078807937482817/image.png?ex=697328d2&is=6971d752&hm=cc46a2b2724a938513f6e3cd2fcc0ba805d46f8d5b57b4a269c96b347f555a5d&

[tepid moth]

this isnt the one im using but it looks way better, the one im using is basically the regulaur exe where it would show you all those premade japanese models

but on the web

[hallow thistle]

[tepid moth]

this

got a high enough level to send smth

[hallow thistle]

Still going for locally run?

[tepid moth]

uhhh

whut that mean 😭

im completely new to this btw

i just wanna know how to be able to make it work through discord

if that makes any sense

[hallow thistle]

When you a .exe file from your PC, it called "to run locally".

Running on Kaggle/Colab means online.

[tepid moth]

ohhh

well im currently running it online

but whichever one works best for discord il go for

since even so, online i dont use discord on a browser, i use the app

[full wadi]

i cant hear my voice on deiteris even when i have the correct input/output/monitor device

[hallow thistle]

Look around the current discussion in #✨│ai-help.

[full wadi]

i have the voice changer client open on my browser (opera gx)

[full wadi]

Ah nevermind running the Client outside the browser fixed my problem

[lapis kiln]

Is there an update?

for voice changer

fork

@full wadi

[full wadi]

I Dont know i just installed deiteris fork

[lapis kiln]

Give me the link

[full wadi]

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

where do i find the force_gpu_clocks bat file

[earnest niche]

im trying to download this https://discord.com/channels/1159260121998827560/1327262488277815356 but i dont have a weight account

[lapis kiln]

What is the benefit of this?

I put it on

[full wadi]

Is there a way to prevent my voice changer client from having worse performance the longer it's opened

[hallow thistle]

Sure, now let's try Tg Develop's W-Okada fork. https://github.com/tg-develop/voice-changer/releases/download/b2397/voice-changer-windows-amd64-dml.zip Tg Develop's W-Okada fork has more recent features than the Deiteris one.

Or otherwise just go for an online website instead, as what I implied in #✨│ai-help message.

[full wadi]

that was the last fork I had in my client, I Probably had my settings wrong but I had less issues with deiteris compared to tg

[hallow thistle]

[full wadi]

I'll try to reinstall tg and ask what settings should I put but if that doesn't work I'll just use online cloud

[eternal shuttle]

why does it sound different when I import a voice like a booster?

[hallow thistle]

!howtoask

[patent trellis]

❓ How to Ask for Help

✅ Before You Ask!

1. Check Docs & Guides: Your answer may already be in the AI Hub Docs or the https://discord.com/channels/1159260121998827560/1159513888199540817 channel.
2. Search the https://discord.com/channels/1159260121998827560/1192011222023950368 : Look for existing posts that solve your issue. Do not invade someone else's post.

📝 How to ask?

Tell your:

• Full GPU Name: (e.g., NVIDIA RTX 4060 8gb vram desktop)
• Operating System: (e.g., Windows 11)
• Detailed Description: What were you trying to do and what went wrong?
• Tutorial Used: Link to the guide you were following.
• Screenshot: A picture of the full error message.

🚫 Prohibited Topics (We Will NOT Help With These)

To maintain a legal, safe & ethical community, we will NOT provide help for:

• ANY illegal activities.
• NSFW/Porn.
  Requests for these topics may be ignored, not helped and result in moderation action.

<:matsuripray:1159685390156967936> Community Expectations

• Be Polite & Patient: Our helpers are volunteers. You may ping the Helpers role once.
• English Only: Please keep all conversations in English.
• Don't Ask To Ask.

[hallow thistle]

Make sure to read help guidelines before start asking, or at least use a translator for this one.

[sonic lintel]

guys, help me find the effect of a walkie-talkie on my voice

transmitter

[hallow thistle]

Tg Develop's W-Okada fork has built-in effect (FX) feature. Make sure to read my message above your query to start asking anything.

[lone nest]

Can I load a model from my Google Drive on AICoverMaker Colab?

[tender kite]

any recommendations for running rvc locally on apple silicon? other than applio, i can't get it to work (can't find how to compile it myself while the compiled version is broken)

[lone nest]

I think apple silicon needs MLX?

[tender kite]

mps should work and it is supported in pytorch

[dry turret]

How do you download the latest version of W-okada?

[dry turret]

which version of wokada is this pleasee

[viral mason]

What gpu do u have

[dry turret]

rtx 3060 ^^

@viral mason

[viral mason]

Ooh nice

Should try out Vonovox it's the current best real-time ai voice changer ^^

-rt

[patent trellis]

💻 Local Realtime RVC

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

• Vonovox

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 GUIDE

• Wokada Tg-Develop Fork

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. GUIDE

• Wokada Deiteris Fork

Deiteris' fork (modified version) of wokada that doesn't get updates anymore. GUIDE

⚔️ Wokada Tg-Develop Fork vs Vonovox

For Windows Nvidia, Both Wokada Tg-Develop fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Tg-Develop Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[dry turret]

Interesting! Do you know a tutorial I could follow? I had searched through YouTube and cannot find anything yet? Is Vonovox something new?

[viral mason]

Vonovox is somewhat new yea! There's no tutorials on it other than just downloading the 1.6.9 version, extracting it, running setup, then finally running start

@dry turret sorry forgot to reply to ur message

[dry turret]

Pls don't apologize, you're all fine! Thank you so much for responding back so quickly however! I'm going to give it a try and will report back if I managed to do it 🫡

[viral mason]

Alrighty! Good luck :3

[tender kite]

mainline worked with some patches. gonna fork mainline and apply said patches, so others can get it to work too :)

[dry turret]

Ahhh here's a forehead kiss!! It's sooo much better than W-okada. I will play around with it more when I'm back home - I really appreciate your recommendation & kindness!!

[viral mason]

Yipeeee, glad I could help! Enjoy the new voice changer friend:3

[smoky dove]

I'm kind of stuck training a model in kaggle

The tensorboard just won't work at all

Gradio ❌
Local tunnel ❌
Horizon❌

But the others work

[twilit bridge]

Hey everyone, I'm reinstalling RVC but start_http doesn't seem to be installing anything when running for the first time. I have my antivirus off as well. It's probably an easy fix that I'm missing, but has anyone encountered the same issue and fixed it?

[twilit bridge]

Nevermind ** I ended up going with Vonovox and I already got it setup & working perfectly. Thanks to anyone in advance though who read it late & was going to help!

Didn't realize its been so long that there's new applications to try

[zealous steeple]

it says 0% when i yap, and works when i use passthru

any help?

[cosmic epoch]

Anyone knows any alternatives for rvc and weightd for making ai covers?

@viral mason excuse me, do you know any alternatives for RVC, Weights or hugginface for creating AI covers?

[viral mason]

really all u can use now is applio

[cosmic epoch]

What about replay? Does it work anymore?

[viral mason]

oh!

yea forgot about that

replay is said to be unaffected by weights shutting down

u can still use that

[cosmic epoch]

Do you have a link to the latest version of replay?

[hallow thistle]

Try delete your current RVC notebook and then import Applio RVC notebook to your Kaggle account again.

[hallow thistle]

I don't use Replay. I use Applio RVC and UVR5/MSST, different programs in separate environments. This might sure sound too hard for you, but that's how I done one AI cover.

[viral mason]

I don't, I would ask in the official weights discord server

[cosmic epoch]

Okay thanks

[hallow thistle]

Applio RVC never has a version that compiled for either Apple Silicon Mac or Intel Mac. Aside from Windows version, there's one for Linux, but I doubt if the Linux version even work on macOS since Apple macOS doesn't explicitly based on Linux, despite sharing similar "Unix-like" terminal. I think there's a workaround that you could get Applio RVC to work on Silicon system, but the Python executable or ".dylib/.so" must be the one compiled for macOS system, and then change Pytorch's Torch command to use "MPS" instead of "CUDA". For Applio RVC's repository, there's GitHub. https://github.com/IAHispano/Applio

By the way, in https://huggingface.co/IAHispano/Applio/tree/main/Compiled, the latest version for Applio RVC Linux is 3.3.0, somehow, while Applio RVC Windows is currently 3.6.1. To get the most recent version, trying to compile one from GitHub source is considered.

[full wadi]

can someone help me how to run tg develops w-okada fork on kaggle i feel like i did something wrong

do i put the token on 'TOKEN_HERE'

[full wadi]

why did i get an xhr poll error

also its taking too long to swap to other models

[nova snow]

hello can anyone help me with something so i have w-okada
and it's working but there is a tiny issue when i talk
the voice repeats like i'm teseting with another account
the voice sounds repeating and there is some sound of electristy in it

[hallow thistle]

What is your PC GPU? And did you follow any tutorial or guide?

[chilly furnace]

well

[subtle solstice]

Hello, could anyone help please?
I have an w-okada ai voice changer. I've watched tutorials, made everything like in it, and when I'm turning my voice on - it just doesn't turns on.
At first i thought that it doesn't work at all, but after some minutes - it turned on, and worked properly. I started to adjust another programm, and turned voice changer off. When i activated it back - it didn't work again. And now, i left it on for about 5-10 minutes, and still - it still didn't start to work, like last time.

[nova snow]

YES I DID
BUT I DON'T KNOW WHAT GPU AM I SO I CHOSE ANYTHING

AND NOT JUST THAT

WHEN

I SIT

IN A VC WITH SOMEONE OR THE GAME SOUND IS HIGH THE VOICE GO TO THE MIC

[hallow thistle]

Calm down with your caps lock. I didn't mean to sound angry or anything.

[hallow thistle]

Sure. What is your PC GPU? Because most tutorial videos from YouTube likely seem to tell you to use the older versions.

[alpine lotus]

bro just click on task manager then performance

if you don't know what gpu you're using

[uncut haven]

Yo guys I've the problem that when I speak that my voice is so perfect and very clean which make others sus so how to fix it? And what does Exlusive Mode do abd F0 and pitch sommthing factor? Also I've index rate at 0.50 and format at 0 but what does Format do`?

[hallow thistle]

Perfect or robotic? Vonovox has been made to output better audio quality in mind. Well, that's too many questions.

• You can try reduce "extra time" value under 1.0 to make the audio quality sound less quality, similar to how W-Okada fork's extra value being left to 0.5 s default instead of 2.7 s.
• "Exclusive mode" is a mode in WASAPI/ASIO that makes that one program (like Vonovox) as the only program to use sound while mutes other programs at the time, if they all on the same audio system.
• "F0" refers to pretrain pitch extraction models that commonly used in RVC, especially RMVPE.
• "Pitch Smoothing Factor" indicates how curve the pitch of audio changes at the time.
• "Index" is a file that stores accent of an RVC voice model, alongside pth file. Generally works well in non-realtime RVC (like Applio RVC), but the exception is when you set "index rate" to any non-zero value, it can strain CPU usage, usually happens in W-Okada voice changer, although Vonovox might suffer similar issue as well.
• For what the term "formant" means, there's Google. I know how this setting works but I'm not sure what to explain about how this works.

[wanton valley]

Applio_noui inference is not working or something, is just this and the task is already "done"

[neat jackal]

Hi, I'm trying to retrain an RVC voice model from Applio AI Hispano.

I'm trying to retrain it from 100 to 200 epochs, and so on (up to 500 epochs), but when I open the link that takes me to the inference, I get this error: "No interface is running right now"

Does anyone know how I can fix this?

Thanks

[viral mason]

Did you re run the steps?

[patent gyro]

If i had the model maker role when I first joined the server and then rejoined, could I still get my role?

[neat jackal]

wdym ?, I only clicked on the steps: "Install Applio", then "Start Applio (Sync with Google Drive)" and finally "Start Server". When I try to access the public link, I get the error I already mentioned.

If you're referring to those steps...

[viral mason]

hmm not sure as I use applio on Kaggle

[neat jackal]

Which website do you recommend for training an RVC AI model with Applio?

[unkempt kraken]

Hello, i have 5070ti but RVC Dosnt work with it, how can i fix that?

With CPU Works, but with 5070TI Dont work

[unkempt kraken]

I did it 2 months ago with 4070, and worked perfectly but now with 5070 ti dosnt work

I dont need guideline

I know im doing it good

[tepid oyster]

Just use RMVPE (Not ONNX)

[unkempt kraken]

It worked! Thanks!

[hallow thistle]

Applio RVC or W-Okada voice changer?

[hallow thistle]

See #📤│model-maker-role to resubmit your voice model, or try contact an admin here for clarify.

[hallow thistle]

Kaggle can run "Applio RVC" notebook, the guy above your query implied, but you can't directly link your Google Drive account there. Google Colab however is known be able to link your Google Drive account.

[neat jackal]

Can you send me the link? Or where can I find it?

[hallow thistle]

On Kaggle, you'd have to import the Applio RVC (https://github.com/IAHispano/Applio/blob/main/assets/Applio_Kaggle.ipynb) notebook link to your Kaggle account by yourself. There are differences between Kaggle and Colab.

[stark plank]

Hey extremely new here can someone please tell me how to install the latest applio RVC

[hallow thistle]

What are your PC operating system and GPU?

[stark plank]

Windows Geforce RTX 2080 Ti

[hallow thistle]

Download Applio RVC from this link. https://huggingface.co/IAHispano/Applio/blob/main/Compiled/Windows/ApplioV3.6.1.zip

[stark plank]

got this step.

[stark plank]

is it ok if we can dm this?

[hallow thistle]

No.

-rvc

[patent trellis]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[icy vessel]

what does this graph represent in the rvc realtime client?

[hallow thistle]

Basically a performance statistic of a voice changer.

[naive pivot]

@hallow thistle my voice changer isnt working

[hallow thistle]

What is your PC GPU? And did you follow any tutorial?

!howtoask

[patent trellis]

❓ How to Ask for Help

✅ Before You Ask!

1. Check Docs & Guides: Your answer may already be in the AI Hub Docs or the https://discord.com/channels/1159260121998827560/1159513888199540817 channel.
2. Search the https://discord.com/channels/1159260121998827560/1192011222023950368 : Look for existing posts that solve your issue. Do not invade someone else's post.

📝 How to ask?

Tell your:

• Full GPU Name: (e.g., NVIDIA RTX 4060 8gb vram desktop)
• Operating System: (e.g., Windows 11)
• Detailed Description: What were you trying to do and what went wrong?
• Tutorial Used: Link to the guide you were following.
• Screenshot: A picture of the full error message.

🚫 Prohibited Topics (We Will NOT Help With These)

To maintain a legal, safe & ethical community, we will NOT provide help for:

• ANY illegal activities.
• NSFW/Porn.
  Requests for these topics may be ignored, not helped and result in moderation action.

<:matsuripray:1159685390156967936> Community Expectations

• Be Polite & Patient: Our helpers are volunteers. You may ping the Helpers role once.
• English Only: Please keep all conversations in English.
• Don't Ask To Ask.

[naive pivot]

Ive got a nvidia 4060
and i followed this guide https://www.youtube.com/watch?v=lzb09-j-7qU

Im on windows 11

[hallow thistle]

Try Tg Develop's W-Okada fork instead. https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/ https://github.com/tg-develop/voice-changer/releases/tag/b2397

[naive pivot]

My microphone connected and i could hear myself but whenever i clicked start it didnt work

[naive pivot]

Okay

[naive pivot]

Wait but the download it says amd

[naive kelp]

guys whats the best free voice changer for amd dedicated I tried okada and it doesnt work well for me

[hallow thistle]

So many people here been confusing "AMD64" for AMD Radeon GPU. AMD64 is another CPU architecture name for x86-64 or x64, which is used in most Intel and AMD CPUs.

If you see the one that says "CUDA", it's obviously for NVIDIA GPU; whereas DirectML (dml) is for Intel/AMD GPUs.

[hallow thistle]

Which AMD Radeon RX GPU?

[naive kelp]

rx6550m

[hallow thistle]

Try this W-Okada fork. https://github.com/tg-develop/voice-changer/releases/download/b2397/voice-changer-windows-amd64-dml.zip

[naive kelp]

it opens on chrome

is that normal

[hallow thistle]

It's completely normal for this specific W-Okada version.

[naive kelp]

is this better than the dml exe one

[hallow thistle]

I can focus on settings. Basically, I don't like answer questions that have less to do with the program itself.

[naive pivot]

Do u perchance have a diff one it doesnt let me open it

[hallow thistle]

Then I think you did wrong. Download both .zip.001 and .zip.002 files, these two files have to be in the same folder, use WinRAR or 7-Zip to open the .zip.001 one.

This is Tg Develop's W-Okada fork. https://cdn.discordapp.com/attachments/1159290139609137264/1463078807937482817/image.png?ex=69752312&is=6973d192&hm=159832e257a3ee50d98328b37af14636e99444fcdff479ec542a7574158eb5c0&

[naive pivot]

Alright

[naive kelp]

how do you merge models

[naive pivot]

@hallow thistle is the one u gave me only for the client cause it doesnt work on discord

[viral mason]

did u set you microphone up properly

[hallow thistle]

Try follow the guide doc about voice changer before you say something out here.

[naive pivot]

Yes

Its working like i hear it but it doesnt transfer to like my discord mic

I got the exact same mic connected

[hallow thistle]

On Discord, I usually do this. https://cdn.discordapp.com/attachments/1159290139609137264/1460231235514138759/image.png?ex=69755310&is=69740190&hm=1ce431b13904811f9f69b46935ee91700ec45d14690aacacdc977d2de929cb8e&

[naive pivot]

Ah okay lemme try that

[hallow thistle]

On W-Okada voice changer, with "Server" audio mode. https://cdn.discordapp.com/attachments/1159290139609137264/1460552908818087966/image.png?ex=69752d25&is=6973dba5&hm=72de3c1bb965af468dcc360b3141171c59d4315609da05b01179106cc2b0a1fd&

This and this might actually give you some ideas.

[naive pivot]

My okada is different

[hallow thistle]

How so?

[naive pivot]

Can i dm you a screenshot?

[hallow thistle]

You can send an image to here now.

[hallow thistle]

What the hell??

[naive pivot]

Thats what i got from this guide https://www.youtube.com/watch?v=lzb09-j-7qU

[viral mason]

I sneezed

there was so much dust on that

[viral mason]

that's very outdated, what is your pc gpu

[naive pivot]

I downloaded the alpha one cause thats the one he said to download

[naive pivot]

4060

[viral mason]

don't follow yt tutorials they don't keep up with the up to date stuff

[naive pivot]

Ah do u have the recent download?

[hallow thistle]

This is what I told you. You didn't actually listen.

[naive pivot]

For the okada

[viral mason]

you should try out wokada tg fork or vonovox, either are good

namari is suggesting tg fork which I also use

download the 001 zip file and 002 zip

[hallow thistle]

Again, this is the program for W-Okada fork.

[naive pivot]

Yeah i need the link i went to the old one u sent but their is no release version

Ill download 001 and 002

[hallow thistle]

I checked my earlier message again, obviously it was tg-develop voice changer, not an alpha or original and outdated version.

When you still use the old and original W-Okada version, it's disappointing.

[young rover]

what do yall think is the best way to make voice models sound the most natural without making it obivous

for example sudden high pitch voice completely messes up the ai

is it the way u speak into the mic

if anyone got any tips lmk

[hallow thistle]

Vonovox or W-Okada voice changer (MMVCServerSIO) fork? Your PC GPU is NVIDIA GeForce RTX 3070.

Yes, I remember I was helping you about the voice changer before. How about your settings on your voice changer?

[young rover]

vonovox rn

[young rover]

its been great lol

ive noticed it pronounces most words good but sometime it jusdt like doesnt

[nova snow]

sorry i just it in some games and forgot it open

[hallow thistle]

While Vonovox is made to give better audio quality than other W-Okada versions in mind, when the audio quality still sounds low quality or robotic, it can happen with the quirk of a voice model itself.

[hallow thistle]

The term/initialism GPU means "graphic processing unit", it's a type of device that processes graphic tasks of a computer especially; modern GPUs have been designed to work with AI programs as well. To check your PC GPU, open Task Manager or press Ctrl + Shift + Esc, go to Performance tab, spot GPU 0 or GPU 1 in the left panel.

[nova snow]

it's nividia

[hallow thistle]

Click on that one GPU to reveal its full name on the right panel.

[nova snow]

can u come to a vc i feel lost ngl

CPU

Intel(R) Core(TM) i5-10400F CPU @ 2.90GHz

this ?

or this
GPU 0

NVIDIA GeForce GTX 1650

[hallow thistle]

Try follow "Tg Develop's W-Okada fork". https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork
I'm sorry, I don't have time to hop on voice chat for you. Download https://github.com/tg-develop/voice-changer/releases/download/b2397/voice-changer-windows-amd64-cuda.zip.001 and https://github.com/tg-develop/voice-changer/releases/download/b2397/voice-changer-windows-amd64-cuda.zip.002 to your download folder, use WinRAR or 7-Zip to open the .zip.001 one.

[elder coral]

How do I wait for Applio to finish installing on Dione because i keep getiing errors or lag

[hallow thistle]

Try run Applio RVC's bat file directly like mine, see if there any error in the terminal (CMD) part.

[tranquil python]

whats the best fork for amd cpu and nvidia gpu

[hallow thistle]

!howtoask

[patent trellis]

❓ How to Ask for Help

✅ Before You Ask!

1. Check Docs & Guides: Your answer may already be in the AI Hub Docs or the https://discord.com/channels/1159260121998827560/1159513888199540817 channel.
2. Search the https://discord.com/channels/1159260121998827560/1192011222023950368 : Look for existing posts that solve your issue. Do not invade someone else's post.

📝 How to ask?

Tell your:

• Full GPU Name: (e.g., NVIDIA RTX 4060 8gb vram desktop)
• Operating System: (e.g., Windows 11)
• Detailed Description: What were you trying to do and what went wrong?
• Tutorial Used: Link to the guide you were following.
• Screenshot: A picture of the full error message.

🚫 Prohibited Topics (We Will NOT Help With These)

To maintain a legal, safe & ethical community, we will NOT provide help for:

• ANY illegal activities.
• NSFW/Porn.
  Requests for these topics may be ignored, not helped and result in moderation action.

<:matsuripray:1159685390156967936> Community Expectations

• Be Polite & Patient: Our helpers are volunteers. You may ping the Helpers role once.
• English Only: Please keep all conversations in English.
• Don't Ask To Ask.

[tranquil python]

iam using NVIDIA RTX 4060 8gb vram pc
windows 11
i wanna know what vcclient or w okada fork is the best for a realistic sounding voice

[hallow thistle]

Tg Develop's W-Okada fork.

[tranquil python]

which one should i download

like from there

i have no knowledge about any of this

[hallow thistle]

If one says "CUDA", it obviously for NVIDIA GPU. Download both .zip.001 and .zip.002 files because these files are split zips of a single zip, use WinRAR or 7-Zip to open the .zip.001 one.

CPU is CPU-only, while dml (DirectML) is for Intel Arc/AMD Radeon RX GPUs.

[tranquil python]

why does cuda stand for nvidia

[hallow thistle]

Look up on Google about what does CUDA mean for NVIDIA, because while I know how CUDA is related to NVIDIA I just doubt if you wish to learn about it.

[tranquil python]

alr ill check it out rn

[formal ermine]

-realtime

[patent trellis]

💻 Local Realtime RVC

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

• Vonovox

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 GUIDE

• Wokada Tg-Develop Fork

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. GUIDE

• Wokada Deiteris Fork

Deiteris' fork (modified version) of wokada that doesn't get updates anymore. GUIDE

⚔️ Wokada Tg-Develop Fork vs Vonovox

For Windows Nvidia, Both Wokada Tg-Develop fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Tg-Develop Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[rich marsh]

how to train voice @bot or something like that

[tranquil python]

is local better or non local

[hallow thistle]

Local.

[hallow thistle]

Are you looking for RVC?

[unkempt kraken]

https://github.com/tg-develop/voice-changer/releases/tag/b2397 This!

[hallow thistle]

• RVC (retrieval-based voice conversion; e.g. Applio RVC) and W-Okada voice changer are two different programs.
• Well, the "ONNX" models only works in DirectML variant, which itself is for Intel Arc/AMD Radeon GPUs. Using ONNX models with NVIDIA GPU would cause the voice changer to fail, so original pth model files are preferred.
• While the pitch extraction model (F0) like rmvpe_onnx does work with any GPU, it's better to use regular rmvpe with NVIDIA GPU especially.

[patent trellis]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[fringe cave]

im want change voice of my streamer by using Desktop Voice changer
Can i somehow mute original stream and catch this stream on Program?

[fair prism]

is there other colabs for training aside from applio?

[fair flume]

can someone help me why no sound come from that voic changer?

[slender mantle]

Found any way out?

[viral mason]

This is old, what gpu do u have

Don't use any yt tutorials they are old, what gpu do u have

[slender mantle]

Same issue with me, by the way

Does the ability to scream or cry depends on the model character or it's type?

Or is it straightly about my GPU?

[fair flume]

what i need download if i have intel cpu? and nvidia GPU

[slender mantle]

Couldn't actually find the answer myself

[slender mantle]

What one exactly?

GPU

And how many ram GB

[viral mason]

It's completely dependent on the voice model

And what it was trained on

[slender mantle]

Do you have any examples?

[viral mason]

Not really but you can look through my models I've made in https://discord.com/channels/1159260121998827560/1175430844685484042

[fair flume]

32gb

[viral mason]

I'm on my phone so I can't get anything specific atm

[slender mantle]

But you have some that have the ability, right

[viral mason]

You should probably get Vonovox

[viral mason]

I do yes

[fair flume]

what is vonovox

[slender mantle]

Search it in internet

Why not

[viral mason]

It's a version of w-Okada that is made by a separate person from the original, it's very good

[slender mantle]

I mean, that's how I found some of good ones

[slender mantle]

Oh, well

[viral mason]

Anything that isn't from here is most likely very old or outdated

[slender mantle]

So RVC Voice Changer isn't the best anymore?

[fair flume]

Operating System: Windows 11 Pro (64-bit)
CPU: Intel Core i3-14100F (4 cores, 8 threads, 3.5 GHz)
RAM: 32 GB
GPU nvidia 4060ti

[viral mason]

There's two that are good and one sorta ok one, there's Vonovox, Wokada tg fork, and wokada deiteris

All found here

-rt

[patent trellis]

💻 Local Realtime RVC

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

• Vonovox

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 GUIDE

• Wokada Tg-Develop Fork

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. GUIDE

• Wokada Deiteris Fork

Deiteris' fork (modified version) of wokada that doesn't get updates anymore. GUIDE

⚔️ Wokada Tg-Develop Fork vs Vonovox

For Windows Nvidia, Both Wokada Tg-Develop fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Tg-Develop Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[slender mantle]

I feel like vonovox is the best choice

[viral mason]

Vonovox would work fine with your system

[fair flume]

where i can download it safely

[slender mantle]

Look above

[slender mantle]

.

First paragraph

Click on guide

[fair flume]

and btw why this MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a dont work for me?

[slender mantle]

And find the installation part

[viral mason]

Made by the ancient greeks

[slender mantle]

Lmao

[viral mason]

https://github.com/dr87/Vonovox/archive/refs/tags/v1.6.9.zip

Here's the download in case u had trouble finding it

I got it from the official discord server

[slender mantle]

Is Vonovox fine for 2070 RTX, I7-6800K and 16 GB Ram?

[viral mason]

Yea it should also run fine with that too

[slender mantle]

Alright

[viral mason]

Maybe not as well at the 40 series but still should be fine

[slender mantle]

Thanks

[viral mason]

You're welcome, you have a virtual audio cable already right?

Like VB cable

[slender mantle]

VAC lite, yeah

I have been using W-Okada since 2024

But the thing is that it was REALLY outdated

So I decided to find something newer now

[viral mason]

Ooh nice, most don't have that one even tho it's the more reliable one

[slender mantle]

Yup, I feel like some people might even use VB cable (the other one) and don't even know about existance of this

[viral mason]

Truee

I've been using w-Okada for a while, don't remember when I started but I remember using the superrrrr old stuff until recently

I personally use wokada tg fork because Vonovox is currently making a new version after 1.6.9 that's going to be wayyyy better than the current one

1.6.9 has only like 8 slots i think and I need over a hundred because I play as different game and show characters on vrchat to have fun and just make friends

[slender mantle]

For real

[slender mantle]

That's actually the main reason why people use it (Besides catfish, lol)

[viral mason]

I use a complicated setup that has Fl studio, Voicemod, Wokada tg fork, VB cable, vac lite, and voice meeter all working together so I can use a soundboard and add voice effects to my existing voice changers

[slender mantle]

I see

[viral mason]

I hate those things, I give them 0 help and annoy them in this server because it's funny

[slender mantle]

This sounds actually nice

Lol

[viral mason]

It's pretty fun tbh, with a regular wokada setup, I wouldn't be able to make my voice glitchy or add echo and stuff

[slender mantle]

I am on my way to installing vonovox

And I have a problem

[viral mason]

Oh no

[slender mantle]

True

Expand-Archive : The path 'ffmpeg.zip' either does not exist or is not a valid file system path.
At line:1 char:1

• Expand-Archive -Force ffmpeg.zip temp_ffmpeg

•   + CategoryInfo          : InvalidArgument: (ffmpeg.zip:String) [Expand-Archive], InvalidOperationException
    + FullyQualifiedErrorId : ArchiveCmdletPathNotFound,Expand-Archive

The system cannot find specified path

Got any idea?

Might be incorrect installation

Nevermind, it's fine

Still got the issue though

[viral mason]

Hmm

What'd u press

After extraction

[slender mantle]

setup

Batch file

[viral mason]

Odd, that should have worked

[slender mantle]

Yeah man, I am not that cavemanish to press start right after extraction 🙏

[viral mason]

Smartest AI hub user

Anyways

[slender mantle]

'runtime\python.exe' is not recognized as an internal or external command,
operable program or batch file.

Found this also

[viral mason]

It's something to do with python

Do u have it installed?

[slender mantle]

I was resetting my pc so C++, python and haskell disappeared

Let me try

To install it

[viral mason]

Tell me how it goes 👍

[slender mantle]

Alr

@viral mason Installed, although, nothing changed

[fair flume]

oh nice its working

thanks

but idk whats is best settings 😄

[slender mantle]

You need to adjust it to yourself

[fair flume]

i cant find anything about vonovox

[slender mantle]

Just try to find the best for yourself

Ideal pitch and so on

Though, since we know your GPU

[viral mason]

All I'd say to change is the crossfade to 0.15

[slender mantle]

Got any idea how to fix this?

[viral mason]

Tbh I got no clue so I'd ask a helper here or I could send u the discord server for Vonovox so I could ask someone in there or ask the creator himself, he's pretty active

[slender mantle]

Can you send the vonovox server please

[viral mason]

I'd have to send it in dms or I'll be burned alive due to this server's rules

U good with dms?

[slender mantle]

For sure

[viral mason]

Bet

[fair flume]

i dont have anything this there

[viral mason]

?

Are you able to send screenshots

[slender mantle]

Waiting for you

[slender mantle]

Use imgur or any other app to send a link

With a photo

[wheat stone]

if it sounds a little too much robotic and cuts off what should I change in settings

[slender mantle]

Chunk size

It means that it's too small and that your pc not fully renders it

What's your GPU

[viral mason]

Just sent

[fair flume]

https://image2url.com/r2/default/images/1769279471198-03072dcd-4965-4e9c-958a-4c51a77121b6.png

[wheat stone]

geforce 5080

16gigs

[fair flume]

https://image2url.com/r2/default/images/1769279529218-7d782a8f-4fef-4404-b009-dc40bc853da2.png

[wheat stone]

how do i change chunk size

[viral mason]

The slider for chunk size

[wheat stone]

vonovox doesnt have that

o prolly named block size

[viral mason]

Oh yea

It's called block size in vonovox my bad

[wheat stone]

sorry so should I make it higher?

[viral mason]

It depends on your PC, if it's delayed lower it until it sounds good but not choppy

[wheat stone]

its not delayed but it has its robotic moment

[viral mason]

Hmmm

What model are you using rn?

It could be that

[patent trellis]

<:error:865860636084142150> Message error

The message template:sticky-ai-help could not be displayed. More details can be found in the error log.
Please report this to a server admin.

[viral mason]

@wheat stone

[wheat stone]

Sorry was eating

https://discord.com/channels/1159260121998827560/1403422896680079390

[viral mason]

oh

nvm I don't wanna help anymore, egirl stuff is gross

https://cdn.discordapp.com/attachments/1303206967308713985/1456732618577150124/HenryEmilysuicide.gif

[neat jackal]

Hi, how can I retrain a model with Kaggle? I trained a model to 100 epochs, and I want to train it to 200 epochs now. I hope you can help me, thanks 🙂

[viral mason]

as long as you had this enabled it will automatically have it saved, then u can rerun the steps in applio

in the training tab, have the same exact settings u had when u first trained it and same model name, skip step 2, then match the same settings u had on step 3 except up the epoch count

[neat jackal]

And where can I find the model files: pth and index, so I can download them?

[ashen patrol]

Idk where is the official weights discors

[lilac scarab]

hi I wanna try vonovox for the first time, is it using VB cables too like w-okada?

[chilly bear]

Bro does this shit work with intel arc gpus

I just switched gpus and it doesnt show up

[viral mason]

if u have intel u screwed yourself

[chilly bear]

Dw bro it was cheap upgrade

So i cant use it

With intel gpus

[modern sphinx]

can someone help me why when i talk it doesnt talk or even talk but like 2 minutes after

i talk but it delays and lags or sum

[viral mason]

u should switch to vac lite instead, it's just vb cable but better
https://software.muzychenko.net/freeware/vac470lite.zip

[lilac scarab]

why do my vonovox client keep screaming "trial trial trial" on my ears

[viral mason]

vonovox client??? where did u get it

that has to be a virus

[lilac scarab]

ooh I found it on youtube

[viral mason]

your computer is screwed man

[lilac scarab]

do you have the one without "trial" things

[viral mason]

delete that

[lilac scarab]

okayy

[viral mason]

https://github.com/dr87/Vonovox/archive/refs/tags/v1.6.9.zip

this is the actual download for vonovox

don't get anything from youtube it is outdated and could be a scam too

[lilac scarab]

I thought bcs mine not premium

so it ask me to upgrade

so no more trial

[viral mason]

you don't need premium, that's optional

[lilac scarab]

is this one dont have to be premium

okayy

thankyou bro @viral mason

[viral mason]

just run setup, then run start

and u should be good, and for vac lite run setup64

u won't have to restart ur pc btw like with vb cable

[brave axle]

what does formant shift and index ratio mean and what do they cause in your voice when being lower and higher? (wokada)

[hallow thistle]

If I explain too much of information, it would stress my brain more.

Nah, what I said in my screenshot is about Vonovox, though few lines there are similar to ones in W-Okada voice changer.

[hallow thistle]

You have called W-Okada a shit. Is your PC GPU Intel Arc dedicated GPU or an integrated GPU?

[hallow thistle]

Neither Vonovox or W-Okada forces you to use a virtual line program, both Virtual Audio Cable and VB-Cable will work with both voice changers.

[hallow thistle]

I've never been to Weights Discord server, but because the server went away from "Discover" feature on Discord so it likely be turning uh as well.

[hallow thistle]

Better not. Try use an anime voice model instead.

[mystic tangle]

with vonovox I can't stop hearing myself

[hallow thistle]

If I were in the US, I could've help these people if they ask in a different timezone from mine; otherwise that one non-helper member who lives in America world might hop on instead of an actual helper.

[hallow thistle]

Try set "output" to "Line 1 (Virtual Audio Cable)".

[elder coral]

how do i fix this because applio makes a black appearance on Terminal after i installed python 3.14 because it's above 3.10

[chilly furnace]

dude

thats the command line

it's probably still downloading dependencies

just let it be 😭

[worthy glacier]

Heyy quick question 😅
I’m testing a voice changer but it’s kinda broken either my mic doesn’t pick up my voice at all, or there’s a huge delay when I talk (like 10 seconds later)
Anyone know how to fix this?

[hallow thistle]

!howtoask

[patent trellis]

❓ How to Ask for Help

✅ Before You Ask!

1. Check Docs & Guides: Your answer may already be in the AI Hub Docs or the https://discord.com/channels/1159260121998827560/1159513888199540817 channel.
2. Search the https://discord.com/channels/1159260121998827560/1192011222023950368 : Look for existing posts that solve your issue. Do not invade someone else's post.

📝 How to ask?

Tell your:

• Full GPU Name: (e.g., NVIDIA RTX 4060 8gb vram desktop)
• Operating System: (e.g., Windows 11)
• Detailed Description: What were you trying to do and what went wrong?
• Tutorial Used: Link to the guide you were following.
• Screenshot: A picture of the full error message.

🚫 Prohibited Topics (We Will NOT Help With These)

To maintain a legal, safe & ethical community, we will NOT provide help for:

• ANY illegal activities.
• NSFW/Porn.
  Requests for these topics may be ignored, not helped and result in moderation action.

<:matsuripray:1159685390156967936> Community Expectations

• Be Polite & Patient: Our helpers are volunteers. You may ping the Helpers role once.
• English Only: Please keep all conversations in English.
• Don't Ask To Ask.

[worthy glacier]

ahh thx

[hallow thistle]

If you keep hoping into random people's direct message for your query, I won't be helping you.

[worthy glacier]

yes sorry i'm just new in the serv so idk how the server works

[dense verge]

hello sry for bad eng.

I have rtx 3080 10gb n r7 7500f 5.0hz 16 gb ddr5 5200mhz. my res time 20.000 in game cs 2

chunk 112 n extra 65k

When playing CS:GO, there is almost no lag, but after five minutes or connecting to a certain map or game mode, the res increases to 20,000 and there is a 20-second delay. How can I fix this? A good video card, a good processor, and a large lag. I have a Virtual Cable client.

[hallow thistle]

Try Tg Develop's W-Okada voice changer fork instead. https://github.com/tg-develop/voice-changer/releases/tag/b2397 https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/

[craggy bough]

• gets sent the "how to ask" guide
• "ahh thx"
• refuses to ask

[long obsidian]

Guys for a 50xx card is vonovox or the deteris fork better?

[hallow thistle]

Deiteris' W-Okada fork wasn't compiled to work with NVIDIA GeForce RTX 50 series; Tg Develop's W-Okada fork does. Vonovox an alternative voice changer also does support RTX 50 as well.

[hallow thistle]

[hardy yew]

Hey! Asking here, because I haven't managed to find an answer in the search box. Is there a way to select a particular GPU in Vonovox (in case of multi-GPU setup)? Doesn't seem like there's any dropdown in the app or an entry in the config. Using the most recent version (1.6.9). Maybe it's possible in beta?

[lavish rover]

it seems the find command in #🔍│find-models is not working

[long obsidian]

Ye but my question is like quality wise the compatible version of w okada for 50xx series or vonovox gives better performance

[thorn mantle]

erm guys how do i make an ai cover i made one like 2 weeks ago with sum called gradio and i forgot how to do it

i looked at the ai docs but i cant find it like the collab im slow

actually nvm i managed to find it

[hallow thistle]

The quality difference between Vonovox and Tg Develop's W-Okada fork is inconclusive, both softwares seem very competitive, because while Vonovox is made to give better audio quality in mind, the W-Okada fork can also give comparable audio quality as well if set chunk and extra values properly but then most of the time the audio sounding robotic with specific RVC voice models on W-Okada.

Oh, because nobody here makes a benchmark between Vonovox and W-Okada fork on exact system, as of now, as much as I've been asking for it, when anyone gets Vonovox they simply just going away, and that's why I can't decide which one absolutely gives better quality everyone cared about.

[hardy yew]

yeah, a direct comparison would be great! It would be easier to make if there was a way to play recordings directly in w-okada just like in vonovox. But this can be done indirectly by using a VAC as input and playing recordings like that.
I might do it some day, because i'm very curious about the actual difference in sound between w-okada and vonovox. Still wondering if any of the two is superior

If i do some tests like that, i'll definitely share the results

[hallow thistle]

There are #1175430844685484042 and voice-models.com (only model list section).

For voice models, I'd use both #1175430844685484042, the websites like voice-models.com and weights.com (about to close down soon) to download voice models. The #1175430844685484042 and website voice-models.com do provide direct links to Hugging face repositories of voice models respectively, whereas Weights likely just borrows models from #1175430844685484042 and stores them on their own server.

[fleet trail]

@hallow thistle

[hardy yew]

did you download the .zip.002 too? you need both of them, it's basically one .zip file split into two

[hallow thistle]

https://cdn.discordapp.com/attachments/1159290139609137264/1456293998196035707/image.png?ex=6976d239&is=697580b9&hm=42b565d648085efd969f5329367a125f65bac87856e2225fef1e672c8387200b&

[fleet trail]

ohhh

okay

soo what should i do now

feel like im lost lol

like idk what to do

[hallow thistle]

Basically wait? That's an awkward question.

[fleet trail]

what?

[hallow thistle]

I use Tg Develop's W-Okada fork.

When you ask me "what should you do", I can't think of anything to say.

[fleet trail]

ohh cool my graphics card :rtx 4070\

i7-13700k

sooo what do u think is the best

[hallow thistle]

-realtime

[patent trellis]

💻 Local Realtime RVC

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

• Vonovox

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 GUIDE

• Wokada Tg-Develop Fork

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. GUIDE

• Wokada Deiteris Fork

Deiteris' fork (modified version) of wokada that doesn't get updates anymore. GUIDE

⚔️ Wokada Tg-Develop Fork vs Vonovox

For Windows Nvidia, Both Wokada Tg-Develop fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Tg-Develop Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[fleet trail]

😂

okay thx

[hallow thistle]

I'm not sure what to say, so please stop asking something too little.

[fleet trail]

im sorry mb

[lavish rover]

Does anyone use Vonovox? Would like to see your settings

it sounds muffled and with a lot of noises for me but I have no issues with RVC GUI

[simple ore]

Applio comes with its own conda environment which is 3.11

whatver you install separately does not matter

for the 'black termina' you just need a wait a minute

[wanton swan]

I keep getting this when trying to generate a song from Heartmula Suno UI. How do I fix it?

?

@tame oracle

[tall depot]

I'm still waiting, Ilaria

[balmy rain]

Where was it

Does this site work anymore? And if not is there a newer one I can find

If there are any youtube videos I could follow i'd really like that

[unborn oracle]

My RVC Doesn't launch anymore, I cant send images but it keeps saying

[VCClient] Access http://127.0.0.18888/
[VCClient] wait web server...0 http://127.0.0.18888/
[VCClient] wait web server...10 http://127.0.0.18888/
[VCClient] wait web server...20 http://127.0.0.18888/
[VCClient] wait web server...30 http://127.0.0.18888/
[VCClient] wait web server...40 http://127.0.0.18888/
[VCClient] wait web server...50 http://127.0.0.18888/
[VCClient] wait web server...60 http://127.0.0.18888/
[VCClient] wait web server...70 http://127.0.0.8888/
[VCClient] wait web server...80 http://127.0.0.18888/
[VCClient] wait web server...90 http://127.0.0.18888/
[VCClient] wait web server...100 http://127.0.0.18888/
[VCClient] wait web server...110 http://127.0.0.8888/
[VCClient] wait web server...120 http://127.0.0.18888/
[VCClient] wait web server...130 http://127.0.0.18888/
[VCClient] wait web server...140 http://127.0.0.18888/
[VCClient] wait web server...150 http://127.0.0.18888/
[VCClient] wait web server...160 http://127.0.0.18888/
[VCClient] wait web server...170 http://127.0.0.18888/

I removed the dots so it doesnt get detected as a link obviously, but yeah, it just goes on forever, it was working fine an hour ago and I re opened it thats when it just died, I tried rebooting and using a VPN but neither worked, what should I do?

[viral mason]

you're using something old

what gpu do u have

[quasi condor]

since is weights is shutting down

how the hell do i use my flux model

[cinder crown]

Where do I download it and where is the guide?

RX 6800 GPU

[quasi condor]

@viral mason

thus fellow person here needs dem voice changer

[viral mason]

what is a flux?

[viral mason]

-rt

[patent trellis]

💻 Local Realtime RVC

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

• Vonovox

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 GUIDE

• Wokada Tg-Develop Fork

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. GUIDE

• Wokada Deiteris Fork

Deiteris' fork (modified version) of wokada that doesn't get updates anymore. GUIDE

⚔️ Wokada Tg-Develop Fork vs Vonovox

For Windows Nvidia, Both Wokada Tg-Develop fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Tg-Develop Fork Pros&Cons & Vonovox Pros&Cons

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

[viral mason]

use wokada tg fork

[quasi condor]

idk cuz i don't use flux

what is flux hmmmmmmm.....

[viral mason]

no clue lol

@cinder crown https://github.com/tg-develop/voice-changer/releases/download/b2397/voice-changer-windows-amd64-dml.zip
here's the download

[unborn oracle]

nvidia rtx 4080 super, ryzen 7 7800x3d cpu

[light swift]

Please explain why if I put a tag on the language model search, it takes a very, very long time to load.

[lavish rover]

Is there a definitive answer to using Okada/ RVC GUI/ vonovox for rtx5000series cards for the best accuracy?

Running a 5090

[light swift]

infinity money glitch?

[viral mason]

u should use vonovox

[lavish rover]

Haha bro you should ask the fellas who own the RTX6000

[unborn oracle]

I honestly have no idea what that is if you could help :)

[viral mason]

rvc gui is separate from realtime wokada

[light swift]

If i have amd card better use Rvc gui?(found this info on reddit)

[viral mason]

what are you referring to when u mean rvc gui

realtime voice changer or trianing a voice model

[light swift]

voice changer

[viral mason]

you mean w-okada then

for amd I would use wokada tg fork

[light swift]

Can I have a direct download link? I was a little confused when I was reading the guide.

[viral mason]

here's the downloadhttps://github.com/tg-develop/voice-changer/releases/download/b2397/voice-changer-windows-amd64-dml.zip

do u have a virtual audio cable like vb cable?

[light swift]

i think... yes?

[viral mason]

uh....

idk what that is so

[light swift]

i download something like vb cable

[viral mason]

use this one instead since I am not sure what that is https://software.muzychenko.net/freeware/vac470lite.zip

it's more recommended

[viral mason]

I can send the download if u need it

https://huggingface.co/dr87/vonovox/resolve/main/Vonovox169.zip?download=true

[lethal basin]

hi

[lavish rover]

which one is recommended for rtx5000series?

I mean the voice changer

[light swift]

I have a question, should I open this server? It's just that all the guides on this point direct you to the website. https://huggingface.co/wok000/vcclient000/tree/main And They say download vcclient_win_cuda_2.0.78-beta or alpha

[vale carbon]

hello

vb audio cable isnt working for me

anyone know why?

it doesnt work when i put output device to discord

or sorry input device

<@&1159293204038955078>

[viral mason]

hi

[viral mason]

this link you have sent is old

[lethal basin]

how are you

[viral mason]

what are you using it for?

[vale carbon]

to route from the vc to discord

[viral mason]

have it like this in the voice changer

[light swift]

soo i need server.exe?

[vale carbon]

oh my UI looks different. im using the one from like 3yrs ago

[viral mason]

u should get a new one

[light swift]

Are there any major downsides to using it online rather than on a PC? And when did you manage to update? The guides are only a month old and already outdated.

[viral mason]

the online versions are eh, I only use the local version

depends on what model you're using, if it wasn't trained on much variety not really

[viral mason]

if you're using wokada tg fork

[light swift]

what mean wokada tg

[void flume]

People are using this version: https://github.com/tg-develop/voice-changer/releases/tag/b2397
It's slightly different but works more or less the same.

Put VB cable into the output and your microphone into the input
and in discord, put the VB cable into the input there.

I recommend reading the readme.md https://github.com/tg-develop/voice-changer/blob/master-custom/README.md. It should help a bit explain how to do stuff and what bugs to look out for.

[viral mason]

are you training a model? I'd say maybe 15-30 minutes but if you know what you're doing you can make a good model out of basically anything

[lethal basin]

kk thank you so much

[dusty sparrow]

TTS on applio colab seems to be bugged is there any fix or workaround?

[lavish rover]

any answer to this question please?

[void flume]

You can't really expect people to afford a 5090 rtx card. Due to nvidia's absurd pricing, you're basically on your own with this. I recommend trying out the two and see what options work for you.
You also need a headphone with 192k sampling support and some audio modulation tools to be sure with audio accuracy.

the output tends to be soft, so you do need to enhance the audio at least and you also need cables with proper isolation; some usb cables can be noisy.

(audiophiles recommend XLR connections instead of usb or standard audio jack)

[karmic vessel]

Sorry this May be a dumb question I have been using a Yozora onnx voice file with version v.1.5.3.18a however recently I upgraded to the newest version 2.0 version which I like a lot its faster and less strain on the gpu however the new version I don't think is compatible with my old voice files and the old voice files dont exist on this discord anymore so I can do a fresh download of the voice is there a way to update a voice file from an older version to work with the new version?

[viral mason]

what link did you get this from?

[karmic vessel]

The Yozora voice pack?

[viral mason]

also that model you have is really old as nobody makes onnx models anymore

[viral mason]

is it a voice changer you're using?

[karmic vessel]

this is what I am using

[viral mason]

yeah.. that is ollld

with your gpu u should be running vonovox, it's currntly the best

[karmic vessel]

yeah... The issue is I have a speech impediment so I use the voice changer to help me sound better but I have also grown very attached to this voice since I have been using it for the last 3 years

[karmic vessel]

I switched to this one because the load isnt to high on the gpu but again I cant use the Yozora voice model

[viral mason]

the voice changer is quite outdated tho, I'm pretty sure you can use that onnx file with vonovox, it'll make it sound better

[karmic vessel]

Alright I will look into Vonovox is the load on the gpu high I just use RVCC for performance

[viral mason]

I'll get u the link

[karmic vessel]

alright should I keep my request open in https://discord.com/channels/1159260121998827560/1465151191678718034 because even if I can use the old version would it still run better if there was a newer trained voice?

[lethal basin]

hi ur amazing

[viral mason]

https://huggingface.co/dr87/vonovox/resolve/main/Vonovox169.zip?download=true here u go

[viral mason]

it would be better to make a new one

[karmic vessel]

thank you yeah.... hopefully somebody can help me with that part I have 0 training when it comes to making models

it doesn't help I choose a really obscure voice from 3 years ago just because I like how it sounded

[viral mason]

do they have voice lines on youtube?

if they do u could get the audio, use a website like mvsep to clean it since it's free, and then use applio to train it all for free

-rvc

[patent trellis]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[karmic vessel]

oh really? so sadly the voice is literally from the senko english dub its not on youtube like thats how rare it is!

[oblique briar]

yo

can someone help me

my voice]

won t work

sopmeone help pelase

[void flume]

Explain what's up, then people can tell whether or not they can help you.

!howtoask

[patent trellis]

❓ How to Ask for Help

✅ Before You Ask!

1. Check Docs & Guides: Your answer may already be in the AI Hub Docs or the https://discord.com/channels/1159260121998827560/1159513888199540817 channel.
2. Search the https://discord.com/channels/1159260121998827560/1192011222023950368 : Look for existing posts that solve your issue. Do not invade someone else's post.

📝 How to ask?

Tell your:

• Full GPU Name: (e.g., NVIDIA RTX 4060 8gb vram desktop)
• Operating System: (e.g., Windows 11)
• Detailed Description: What were you trying to do and what went wrong?
• Tutorial Used: Link to the guide you were following.
• Screenshot: A picture of the full error message.

🚫 Prohibited Topics (We Will NOT Help With These)

To maintain a legal, safe & ethical community, we will NOT provide help for:

• ANY illegal activities.
• NSFW/Porn.
  Requests for these topics may be ignored, not helped and result in moderation action.

<:matsuripray:1159685390156967936> Community Expectations

• Be Polite & Patient: Our helpers are volunteers. You may ping the Helpers role once.
• English Only: Please keep all conversations in English.
• Don't Ask To Ask.

[oblique briar]

HELlo someone help my mic is not connecting like

i can hear the voice for myself but no one elsee can

[void flume]

Can you show your settings?

[oblique briar]

okay

i dont haev

pic perms

[void flume]

So, input needs to have your microphone
and output needs VB Cable (or Line 1 in case you use VAC)

and then in discord, Input needs the VB Cable (or Line 1 in case you use VAC)

[oblique briar]

can i show u?

via dms?

[void flume]

Maybe you can post now since you went level up. You can probably also post pic in #1192011222023950368

[oblique briar]

im trying to use it on google voice on telgram

and its just not

is it the way im inputting it?

[viral mason]

that's old

[oblique briar]

oh.

[void flume]

No, that looks fine actually

[oblique briar]

wheres the new one>?

[viral mason]

is that wokada deiteris?

[oblique briar]

why is my voice delayed too is that my mic?

@viral mason

[viral mason]

oh

yea that is ollld

[oblique briar]

wheres the new

what the heck

[viral mason]

u should swtich to vonovox

https://huggingface.co/dr87/vonovox/resolve/main/Vonovox169.zip?download=true

[oblique briar]

where?

is this legit?

[viral mason]

yeah it is

[void flume]

It's a different voice changer, and it is legit

[oblique briar]

isi t better

and does it have delay?

[viral mason]

it's the current best

[void flume]

but your settings in this looked okay, so I think the problem is with the settings on telegram or google voice

[viral mason]

all voice changers have some delay

but you can minimize the delay with settings

[oblique briar]

which voice changer

do u use?

can u show me

whatit looks like

[void flume]

I use tg-develop's fork of w-okada .... also I don't think my settings will help you lol

[viral mason]

I also use that one

[oblique briar]

why?

im trying to sound like a girl

wit hthe less delay

[viral mason]

nvm don't help this user

[oblique briar]

dude

im transganger

gender*

sorry

[viral mason]

oh nvm then

I thought u wanted to scam ppl

[oblique briar]

no

im trans

[viral mason]

that's cool :D

[oblique briar]

are you able to help please 🙂

[oblique briar]

why so big]

[dapper imp]

Lol

[dapper imp]

Cause more advance software

[void flume]

see, this is why it wouldn't be helpful. I have some customized settings that are unusual on my system

[oblique briar]

is this ?

[dapper imp]

ASIO is supported?

[void flume]

Yes

[dapper imp]

Haha, time to go that route

[viral mason]

the default settings on vonovox is fine you won't need to change them tbh

[dapper imp]

I already got asio setup anyways

[void flume]

no, vonovox is better for your GPU, I think. It's not the same

[oblique briar]

is their inputs?

or can someone give me girl settings please and thank you!

[viral mason]

you would need a girl model from here https://discord.com/channels/1159260121998827560/1175430844685484042

[dapper imp]

First, get the right girl model that is trained on your language. Pitch and formant to fine tune it

[viral mason]

formant honestly doesn't help me ever, tho I don't use normal human voices ever

[oblique briar]

can you link it please?

[dapper imp]

If I had it I would😭

I'm also search for model

[oblique briar]

r

im trying right now too

[dapper imp]

I use hindi as my main and english for secondary. The english model are fine but the Hindi ones are nowhere to be found

[oblique briar]

can i hear?

[viral mason]

you have to look here I don't know the best girl voice since I dislike those kind of models since scammers usually use them https://discord.com/channels/1159260121998827560/1175430844685484042

[dapper imp]

Mine is no special. I'm using the smartflace provided one

https://discord.com/channels/1159260121998827560/1464921135920713835

This one

[oblique briar]

oh i would like to hear if can

[oblique briar]

that's just sad

[dapper imp]

It's only 200mb or less. You can download it. There's also some voice samples in the chat itself

[oblique briar]

thank you

[dapper imp]

Np

[oblique briar]

is this the download?

[void flume]

I should note that I am only doing basic stuff with ASIO, as I have no VSTs or special drivers to route audio through. I don't have hardware for it either.
Basically I use ASIO as a replacement for WASAPI, connecting two drivers together.
the reason why I did so is because I am routing audio to OBS, using OBS's build in filters to enchance the audio coming from my mic. (Equalizer, compressor, limiter, etc.), so that the voice changer hears me better. I also route it back to OBS and enchance it afterward, since I noticed the build in Audio Effects (for tg-develop's fork) behave a bit unexpectedly). This allows me to set the gain up and make the output near equal to the input in volume.
After that I route the audio to discord or where-ever.

but ASIO works, and you should be able to use studio tools.

[viral mason]

I heard wasapi is better, how come u use asio?

[void flume]

I can tune the buffer size and delay to more exact numbers, basically forcing things down.
A second reason is because tg-develop doesn't save settings properly and with ASIO only one device shows up.

[oblique briar]

the floder

gave me a virus normal?

[void flume]

There isn't a noticable improvement between wasapi and asio on my hardware at least, but the delay is kinda large anyway

[oblique briar]

=normal?

@viral mason @void flume

normal?

it gave virus

[viral mason]

what???

did you download

[oblique briar]

,

[viral mason]

that's not a virus dummy

/nm

[void flume]

First time downloading stuff?

you always get a risk warning, when downloading, because microsoft, discord, etc doesn't want to be held liable

[viral mason]

yeah it always says there is a risk for downloading literally anything

on discord

[oblique briar]

WAIT

im going to try now

[void flume]

I use speakers, normally at least. Sometimes I use a headphone, and microphone. I don't always have the voice changer running.
and I also turn my monitor off, which has a HDMI interface, so .. there is a whole list of devices that gets added, changed and removed, especially when I test software like Wave Link which also adds audio drivers.
The list is unsorted, and changes. ASIO makes it easier to select input and output; that's the primary reason for me currently.

[oblique briar]

how to work iut

@viral mason

[void flume]

You need to decompress the zip file first

You're currently glancing inside of the whole zip file, which is a compressed file.

[oblique briar]

extract?

[void flume]

Yeah

https://support.microsoft.com/en-us/windows/zip-and-unzip-files-8d28fa72-f2f9-712f-67df-f80cf89fd4e5

in case you need microsoft's guide for it

[oblique briar]

@void flume wh

how to use

[void flume]

https://docs.aihub.gg/realtime-voice-changer/local/vonovox/

I think even the bot has docs for vonovox, but I forgot how to call those

You add your model to the top list, and then change settings below it, and then run the server basically

[oblique briar]

how i fix delay

[void flume]

there is always about a second of delay

[oblique briar]

do u have settings

[void flume]

just set your audio backend on wasapi
and then set your devices, set the F0 to rmvpe

sample rate 48000

input device needs to be your microphone, output device whatever you want to use as input in discord

the software takes care of the rest automatically, and attempts to keep the delay the lowest

[oblique briar]

what abouit this

@void flume

[void flume]

crossfade 1.5
extra 2.7 (though I guess with vonovox you can go higher)
block size, not sure

[oblique briar]

any ofthese?

or these to sound more like a women

[void flume]

pitch should probably be at 12, formant 0

[void flume]

if you just read the guide / website link I gave earlier you'd be able to tweak it yourself better

[oblique briar]

okay

and the lagis normal cus this delay is lowkey bad

[void flume]

you can optimize a bit on the delay, but it will always be somewhat around over half second at the very minimum

someone said 700+ms once

AI simply isn't that fast

so if you're closer to a second of delay, then you can go slightly faster by lowering crossfade, extra, block

it will cost the quality

[candid urchin]

Is there a way to make the voice stay exactly like the one I trained, without having to adjust any settings

I’m using this software, but when I adjust the parameters, the results are often not what I expect

[void flume]

sorry, formant should be 0, put it at 1 if you want to make it sound more like an anime character (unless the voice is already an anime character voice)
formant is basically how tight the throat gets squeezed. Negative makes it, more open which can help offset anime voices to a bit more realistic ones. As for the other settings, experiment.
Make sure you actually speak like a woman to sound like one, Pitch can only do so much.

[oblique briar]

@void flume

my friend can hear himself

thru the mic

how do i fix

[void flume]

Is that friend's audio playing through your speakers?

[oblique briar]

yes

he can hear myself

he can hear h9imself

[void flume]

then it's no wonder, since your mic picks up the same audio

[oblique briar]

thry my micwhen i call

[void flume]

You wil need to lower the sensitivity of your microphone and lower the volume of your speakers..... and also use a Noise Gate setting.

so that low volumes are ignored by the software

[oblique briar]

THIS?

[void flume]

Most people use a closed ear headset to avoid getting cross talk between the mic and speaker

[void flume]

Your settings are fine, but the problem is that in real life, your speakers are making sound, and your microphone catches that sound

[oblique briar]

yeah i dont know how to fix that

[void flume]

so you need to work around that

[oblique briar]

is their a way i can cancel noise?

or i need a better headset

[void flume]

Like I said, you can try a noise gate, and then lowering the volume of your speakers, and then setting the noise gate to a level where only your louder voice gets accepted

you'd need to tweak it till it matches things

Your mic's volume setting has no influence on this I think

but if it has a hardware volume switch, then it may

[void flume]

This is a recommendation for this anyway.
and also a directional microphone

it doesn't need to be expensive, cheap mics work, but like.. a cardioid dynamic mic is probably optimal

[oblique briar]

dxo you have any

[void flume]

'cardioid' means that it is trying to only get sound from 1 direction.

[oblique briar]

i should buy?

[oblique briar]

okay

[void flume]

dynamic mics have an inherit sound filtering property, so it catches less noise

they are less popular than compression mics though, which most headsets have

[void flume]

I don't think I am allowed to advertise brands lol

but there is a chinese microphone brand that is affordable yet has nice quality that I would recommend I guess.

[oblique briar]

send in dms please

rumi

should i mess with this or no

im trying to fix more settings rn

[void flume]

You don't need to mess with the EQ band settings

the noise gate should be all that matters, and higher is better especially when you have speakers, but don't set it too high or your own voice will be filtered.

I send a mic I am positive about in your dms as you requested

I'm gonna go afk for a while

[lavish rover]

is it true that higher block size = better quality and accuracy but with more delay?

[lavish rover]

fifine?

[kind moss]

can p;z

can someone help plz

i installed vcc and vc cable

did everything as tutorial said

it just doesnt work

i choosed my micro as input and vb calbe as output

but nothing happens plz

<@&1159293204038955078>

<@&1159293204038955078>

<@&1347412420498292827>

<@&1159293140440723499>

[hallow thistle]

Please no mass pinging. Just one "@ helper" is enough.

[hallow thistle]

[tame oracle]

[kind moss]

ok

but

plz

help me

with vcc

[hallow thistle]

!howtoask

[patent trellis]

❓ How to Ask for Help

✅ Before You Ask!

1. Check Docs & Guides: Your answer may already be in the AI Hub Docs or the https://discord.com/channels/1159260121998827560/1159513888199540817 channel.
2. Search the https://discord.com/channels/1159260121998827560/1192011222023950368 : Look for existing posts that solve your issue. Do not invade someone else's post.

📝 How to ask?

Tell your:

• Full GPU Name: (e.g., NVIDIA RTX 4060 8gb vram desktop)
• Operating System: (e.g., Windows 11)
• Detailed Description: What were you trying to do and what went wrong?
• Tutorial Used: Link to the guide you were following.
• Screenshot: A picture of the full error message.

🚫 Prohibited Topics (We Will NOT Help With These)

To maintain a legal, safe & ethical community, we will NOT provide help for:

• ANY illegal activities.
• NSFW/Porn.
  Requests for these topics may be ignored, not helped and result in moderation action.

<:matsuripray:1159685390156967936> Community Expectations

• Be Polite & Patient: Our helpers are volunteers. You may ping the Helpers role once.
• English Only: Please keep all conversations in English.
• Don't Ask To Ask.

[hallow thistle]

And also please, please stop spamming each words into different messages separately. Don't be too hurry. You basically just type out a single message like mine; otherwise I won't be able to identify the issue from your "words".

[viral mason]

Some people are very desperate to get it working, it's not life or death 🤷‍♀️

[kind moss]

Full GPU Name: RTX 5070 ti 16 gb
Operating System: Win 11
Detailed Description: I installed VCC (voice changer) but it doesn't work, i mean i choosed and setted up everything, i put my microphone as input and vb cable as output. but i dont hear any sounds, i choosed vb cable as input in discord and other apps
Tutorial Used: https://youtu.be/O3Q2urZSgM8?si=F88EtoONVRezuz9y
Screenshot: theres no an error but i will just send some screenshot

<@&1159293204038955078> pls sir check this out

[kind moss]

maybe for u for me it is important.

[hallow thistle]

Unless if you want to enhance your audio artificially using FX.

[kind moss]

sir pls check my message

[tame oracle]

No need to ping the helper role when someone is already helping you, be patient please

[hallow thistle]

[kind moss]

okay

[hallow thistle]

Right now I am providing an answer for you, so if you could stop forcing me it would be great.

[viral mason]

This software you're currently using is quite old

I'll let namari help you as I've got to get my beauty rest

[hallow thistle]

Try Tg Develop's W-Okada fork instead. https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/ https://github.com/tg-develop/voice-changer/releases/tag/b2397

[kind moss]

wait i tried other one before also

it was same

[kind moss]

ok ty

is it like same voice chanfger

[hallow thistle]

Try newer W-Okada fork instead.

No, different one.

[kind moss]

son

will models work from this server

on w okad

[hallow thistle]

[kind moss]

?

[viral mason]

Yes

[kind moss]

ok thanks for answer unlike this namari guy

he answers so long and not understable

[hallow thistle]

That's an insult.

[kind moss]

its not

just saying facts

[hallow thistle]

It is.

[kind moss]

son

can u help what to choose

[hallow thistle]

I'm sorry, I'm more of detailing my own answer before I send anything, that's why I'm so fucking slow. But if you're looking for a fast-answering person, just look for someone else that's not me.

[hallow thistle]

Download voice-changer-windows-amd64-cuda.zip.001 and voice-changer-windows-amd64-cuda.zip.002.

[kind moss]

what do u mean, i need to download 2 files?

[kind moss]

what if i have nvidia not amd

[hallow thistle]

I still can't keep up with your typing speed, by the way, I've been saying about this so many times. These two files are split zip files of a single file, AMD64 doesn't refer to an AMD Radeon GPU, AMD64 is a CPU arch name used in most Intel and AMD CPUs.

[kind moss]

i mean, should i download 001 or 002?

whats the difference

[hallow thistle]

Don't ask more. Just download them.

[kind moss]

ok

i hope it not virus brotha