✨│ai-help

[knotty edge]

is there like a tut to set it up?

[cinder crest]

hi

my voice changer crashes on opening

the cmd just shows for a split second

[tame oracle]

that is the third link, its a guide

[cinder crest]

can anyone help me

[tame oracle]

can i get more info please? what voice changer, if you got it from a youtube tut, whats your GPU?

[cinder crest]

gtx 1650 super, and im using MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a

[low shard]

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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⛔ Outdated/Discouraged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

[low shard]

id suggest vonovox personally

[tame oracle]

and which file were you clicking to start it?

[low shard]

there's also wokada tg-develop fork

[cinder crest]

start_http.bat - Shortcut

[low shard]

@tame oracle that's an over year old version of original wokada

deffo from some old youtube/video tutorial

[low shard]

you're on windows 10/11 right?

[cinder crest]

10 yea

[tame oracle]

thats what i was trying to check, thanks

[low shard]

windows 10 support officially ended on 14th october 2025

id suggest you to update for security reasons

but anyways that's your choice

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[low shard]

i'd personally suggest vonovox for your setup

[cinder crest]

does it work the same?

[low shard]

but if u don't like it, wokada tg-develop is also a good option

[cinder crest]

yea i was thinking about it

[low shard]

realtime voice changer for calls/games with RVC model support? yes

[cinder crest]

okay thanks Ill give it a try

[low shard]

it's fine :)

[tame oracle]

just for future what version number is definitely outdated?

[low shard]

@knotty edge wokada tg-develop is a fork (modified version) of deiteris, it will work the same, but it has free voice effects + spin model support + a different UI

[low shard]

https://huggingface.co/wok000/vcclient000/tree/main this contains all the original wokada versions, all of them aren't suggested, the old versions especially

the v2 beta versions are just UI or Beatrice models (which are lightweight, but lower quality than RVC) updates, so they aren't suggested because it's not related to RVC

be aware that most people will mention v.1.5.3.18a because it's seen in outdated youtube/video tutorials

You mostly are a newer RVC helper rather than realtime voice changer iirc?

[tame oracle]

yeah i am, but been trying to get all this down with the voice changer and such

[low shard]

ohh i see, the AI HUB Docs could help for updated info :D

[tame oracle]

thanks for your help tho :]

[knotty moth]

the og wokada, even in the latest version 2.2 beta (?) is not recommended due to poor optimization compared to the following ones we recommend (unless you want to use the beatrice model lol)

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[tame oracle]

alright, ill remember that for later

thanks

[ruby valve]

Realtime wdym?

And idk what else to watch besides yt guides 😔

[low shard]

it's a discord bot command, read the vonovox guide below

[ruby valve]

Could you friend me and explain more I really am lost

[tawdry gull]

Queue Ended!

Enjoying music with me? Consider Voting for me

You can also buy our premium plans to access all high quality features!

[analog pilot]

at voice changer how to Turn Selfhearing off means I hear myself (the Female Voice)

[knotty edge]

I mean don't you need python and pytorch installed since the one im using rn uses it or do i need to install the new one

[odd tide]

what best voice changer for amd radeon rx 6600

[knotty edge]

this my gpu GPU 0

Radeon RX550/550 Series

Driver version:    31.0.12029.10015
Driver date:    11/30/2022
DirectX version:    12 (FL 12.0)


Utilization    2%
Dedicated GPU memory    0.7/4.0 GB
Shared GPU memory    0.1/8.0 GB
GPU Memory    0.8/12.0 GB

can someone give me a good voice changer with good sounds

that i can use

[brittle wing]

how do i get the others??

i only have crepe_tiny for f0 det. the rest is n/a.

[knotty edge]

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[knotty edge]

@tame oracle i got the 3rd one i can hear smth coming out now but no girl voice

i tried 2 dif models

just my normal voice coming out

[tame oracle]

im so sorry, i was afk, and what are your settings?

[viral mason]

I don't understand how people get that issue, I've never encountered it so idk how to even fix it

Very interesting 🤔

[fallen elbow]

Anyone know alot about designing websites with Ai and selling them to local buisnesses?

[visual barn]

RuntimeError: cannot cache function '__shear_dense': no locator available for file 'librosa\util\utils.py'

[16160] Failed to execute script 'MMVCServerSIO' due to unhandle exception!

I believe it's the same issue as this, https://github.com/w-okada/voice-changer/issues/1312

[sonic mauve]

Hey, do you know why I can’t run some models on Vonovox? There’s no sound coming out, even though my mic input and output settings are correct. The same model works fine on Okada though

[waxen timber]

whats the best tool for text to audio ?

[hallow thistle]

On Weights website, there's a feature where you can use two voice models to do "duet" together on one single track. But for more complex and accurate duet mixes, I would make two or more separate AI covers and mix them on REAPER like in this screenshot.

[simple ore]

get something more up to date

[visual barn]

as in..?

[simple ore]

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[frosty raptor]

how do i train a voice?

[boreal timber]

Silent Mentor – A small AI experiment for self-reflection

Hi everyone,

I’m developing an AI protocol called “Silent Mentor” that asks targeted questions to help people reflect and gain clarity. Instead of giving advice, it focuses on thought-provoking prompts.

I’d love to hear your feedback:

Would you try something like this?

What kind of questions would you find most useful?

Thanks for your thoughts!

[knotty edge]

for my model?

@tame oracle my pitch is at 0 formant shift is 0 index is 0 speaker is targbet 0 i have only Suppresion 1 on my in.Sense is -90db my gpu is my gpu yk i clicked off my cpu to my gpu and my input is my mic output is the cable input my monitor is my headphone speakers

for advanced settings protcol is on rest crossfade length is 0.05 silence front is off force jp32 mode is off disable jit compilation is off onvert to onnx is checked protect is 0.5 and skip pas through confirmation is yes

[viral mason]

What gpu do u have

Since noobie didn't even ask which is important for which one you should download

[visual barn]

NVIDA GeForce GTX 1650. Although, the problem has diverge by downloading another voice mod that he noobie recommended. Thank you for the help.

[viral mason]

Which one? Wokada deiteris

[visual barn]

Vonovox.

[viral mason]

vonovox is only good for anything 30 series and up

unless it runs fine for you

[knotty edge]

you can help me?

[viral mason]

sure

[knotty edge]

i was talking to hlubby

but he or she hasn't responded

so i need help

[knotty edge]

i went from normal w okada to the fork version.

[knotty edge]

.

[knotty edge]

as soon as i start it every time it's just going up and up with ping

[knotty edge]

.

keep in mind im in the web version thingy

[viral mason]

idk what causes that error where u hear your real voice instead of the ai

[knotty edge]

i can hear the ai but it's super super delayed and buggy

and the ping is hard rising

[viral mason]

so your gpu is amd right?

[knotty edge]

yes

[viral mason]

you could probably use Wokada deiteris fork

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[knotty edge]

i am using the fork in the web

is it supposed to be in like and app?

[viral mason]

this one right?https://github.com/deiteris/voice-changer/releases/download/b2332/voice-changer-windows-amd64-dml.zip

[viral mason]

nah that specific one launches on web browser

[knotty edge]

yes

[knotty edge]

like when i start it just ping rise a lot like keeps going like 4k 5k 6k

so on

i tried 2 dif models

[viral mason]

that's not normal

what settings do u have

[knotty edge]

.

@viral mason

[knotty edge]

.

for both models btw

[viral mason]

try these settings

[knotty edge]

ill try them does it matter if it's rvc model or?

[knotty edge]

i uncheck convert to onnx

cause that you setting?

[viral mason]

rvc is pretty much all the models that are made here so u should use that

[knotty edge]

ok does it all save automatically i went to the advanced settings and i did it and the other settings

[viral mason]

everything will save automatically except for protect in advanced settings and the pitch in the regular settings

all u gotta do tho is just click this

save settings

[knotty edge]

alr i did everything do i try both my models?

i wish i could send ss of what is ee but i cant

wait

i can

[viral mason]

sure

[knotty edge]

@viral mason

i start it i hear the voice like 10 seconds later and it's like very bug i hear it but very bug and delay

[viral mason]

I don't help people who use egirl models sorry

either give a very good reason you're doing it or I'm outta here

[knotty edge]

t

troll my friends

[viral mason]

that it? no weird stuff or trying to get money from people?

[knotty edge]

hell nah

that's weird

i just wanna troll man

[viral mason]

god

thank god

could u send a short recording of what it sounds like btw when u talk into the voice changer

like it it choppy or

[knotty edge]

maybe ill get like a medal clip

and put it here

me talking into it

[viral mason]

alr

[knotty edge]

also does the in.Sens matter

[exotic bridge]

yo I need help. I run the file and then it doesn't open

it stops on "[Voice Changer] MMVC_SocketIOApp initializing... done."

[viral mason]

that's just noise reduction basically

like for bg noise

[viral mason]

did u get the voice changer off a yt video

[exotic bridge]

ya

lemme show rq

https://www.youtube.com/watch?v=SxdnGxicJOg&list=WL&index

[viral mason]

it's over a year old then, super ancient tech

[exotic bridge]

this 1 is 4 months

[viral mason]

even if the video is from a month ago it's over a year old voice changer

they don't use the new stuff

[exotic bridge]

bruh 😭

wheres the new model?

[viral mason]

what gpu do u have, there's 3 up to date voice changers u could download

[exotic bridge]

rtx 3080

[knotty edge]

@viral mason https://medal.tv/games/screen-capture/clips/lkPrhn1BneH95ne4j?invite=cr-MSw3d3EsMjc1NDMxMjgx&v=30

[exotic bridge]

https://github.com/w-okada/voice-changer/blob/v.2/docs_i18n/README_en.md

I got it from here

[viral mason]

that sounds like a robot lmao

[knotty edge]

ik btw i said yo like 10 sefconds before that happened

[viral mason]

damn

[knotty edge]

is there anything i can do? cause im guessing if it's doing this for those it will def do it for the other models i plan to use

[viral mason]

u should get vonovox

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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]

first guide

[viral mason]

tbh I really dunno since it depends on your computer and ur gpu

it should run that one fine

[knotty edge]

maybe if i install my latest drivers

[bright prairie]

guys what is the best free voice for a deep voice?

[knotty edge]

you think that will help?

[viral mason]

go for it

[fossil dune]

can anybody help me my thing cuts out a lot when using custom voices but the built in ones sound fine

i have amd 6500 xt and windows 11

[viral mason]

Built in ones?

[fossil dune]

the ones that come with it

[viral mason]

That's an outdated software you're using

Since you have amd u should use wokada deiteris it's the best for amd

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[knotty edge]

i upgraded my drivers

[viral mason]

Third guide, basically what you currently have but actually good

[viral mason]

Does it sound any better?

[knotty edge]

ima try it rn

[fossil dune]

ok ill try it out and update you

[knotty edge]

it's kinda still the same

my gpu thing was way in 2022 drivers

now it's in like 2025 or 2024

[viral mason]

Damn

[fossil dune]

when downloading deiteris theres three windows download files do i download all of them or just one

[viral mason]

You might have to just get a new GPU tbh

[viral mason]

Get the one for amd

[fossil dune]

voice-changer-windows-amd64-cuda.zip.001
1.95 GB
Dec 7, 2024
voice-changer-windows-amd64-cuda.zip.002
613 MB
voice-changer-windows-amd64-dml.zip

[fossil dune]

the first one here?

[viral mason]

https://github.com/deiteris/voice-changer/releases/download/b2332/voice-changer-windows-amd64-dml.zip

this one

[fossil dune]

alright

[knotty edge]

i mean like 2 or 3 weeks ago i did spill orange juice in my pc

might js be a fried gpu

[viral mason]

Oop

Yeah that might be why lmao

[fossil dune]

i opened the MMVCServerSIO application and the command prompt isnt saying anything now but nothing launched or anything

does that mean its done

oh nvm it opened in my browser mb

[viral mason]

yea it takes a sec to load

[fossil dune]

it worked and i downloaded virtual audio cable

its set up but how do i use my output and input devices

[fossil dune]

its working fine thanks

[viral mason]

Glad I could help

[split fiber]

-rvc

[patent trellis]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[split fiber]

-colab

[patent trellis]

📒 Google Colab Notebooks

Google Colab is a Cloud (Remote Good PC) Service. While the Free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.

• **Applio**

by IA Hispano
Google Colab

• **RVC Mainline**

by Hina
Google Colab

• **UVR5 NO UI**

by Eddy
Google Colab

• **UVR5 UI**

by Eddy
Google Colab

• **Wokada Deiteris Fork**

by Deiteris & Hina
Google Colab

• **Hina's Modified Original Wokada**

Google Colab

• **RVC-AI-Cover-Maker-WebUI**

by Shiro & Eddy
Google Colab

• **FaceFusion UI**

by Nick088
Google Colab

• **FaceFusion NO UI**

by Nick088
Google Colab

• **Music Source Separation Training (Inference)**

by Jarredou & Makidanye
Google Colab

[abstract briar]

can I ask for help here

[viral mason]

Yes this is the help channel lol

[abstract briar]

what do I do about incompatible files

[viral mason]

Wdym

[abstract briar]

it dosn't work anymore but it used to

like stuff

[viral mason]

I have 0 idea what you're talking about what program are u trying to use

[abstract briar]

voice.ai

and importing custom voice file

and it says incompatible file

[viral mason]

I'm sorry you have to use that, what gpu do you have there are much better free ai voice changers than that thing

[abstract briar]

my gpu is good

I think

like 3060 maybe?

[viral mason]

If you're unsure you can check it with task manager

[abstract briar]

I think its a 3060 laptop gpu

[viral mason]

O

Like Nvidia orrr

Intel?

[abstract briar]

nvidia

sorry for slow responses but its rtx smth ig

[viral mason]

Nah you're good, there's no rush

I'd recommend trying out w-Okada deiteris first since you are on a laptop it's the third guide

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[abstract briar]

what about the incompatible file

also im broke so I can't use anything else unlesss its free

[viral mason]

No idea, I don't use voice.ai since it sucks tbh

[abstract briar]

ok tysm

[viral mason]

No problem! If u need help I'll be here

[abstract briar]

whats the website for the voice changer you mentioned

[viral mason]

There is no particular website you just download it using the link for your gpu (should be the first one in the download section) and run it using the mmvcserversio exe file

Sorry it took so long to respond I got a random phone call

[abstract briar]

is it https://github.com/w-okada/voice-changer

its the first one I found

?

??

still here?

[viral mason]

Wdym, all u do is click the link in the guide and download it

https://huggingface.co/Shadicti/deiteris-Fork/blob/main/voice-changer-windows-nvidia-b2332.zip

This one right here should work for you

[abstract briar]

ty

[abstract briar]

why is it saying windows protected my pc?

this is scary

I need more help

it says I don't have enough storage

I have 203gb -_-

@viral mason pls help

[viral mason]

Idk how to help if there's a storage issue

I only know how to set it up and use it, I know nothing about errors

[abstract briar]

Is this the start button?

[viral mason]

What happened btw?

[viral mason]

Yes that launches the voice changer

[abstract briar]

I'm not good with tech :(

[abstract briar]

I said run anyways

why do you keep leaving :(

[viral mason]

Sorry I'm busy with stuff and trying not to take too long from helping you

[abstract briar]

oh sorry

[viral mason]

You must've hit a wrong button bc it shouldn't do this normally

[abstract briar]

I'll redownload

[viral mason]

@knotty moth did I give them the correct link? they said that have a 30 series nvidia card but use laptop

[knotty moth]

I think so

[abstract briar]

I mean I tried to open it before it was done extracting so that might be why

nvm it broke again

[viral mason]

try this one, delete the first one I gave u
https://github.com/deiteris/voice-changer/releases/download/b2332/voice-changer-windows-amd64-dml.zip

[knotty moth]

mb but make sure you store it in c:\ or somewhere else instead of Desktop/Downloads/Onedrive

[abstract briar]

oh...

[viral mason]

actually before u delete anything are u closing the command prompt after the voice changer opens?

[abstract briar]

how will I find it if its in C:

[abstract briar]

should I be?

[knotty moth]

I said to move to there

[viral mason]

no, keep it open when using the voice changer or it'll break

[abstract briar]

I don't get it

the zip file?

or the extracted file

or the application

ok I did all for good measure

[dusky gale]

hey i'm trying to find the freesia-sing/speech pretrains ~~models ~~but huggingface says the repos are gone

[abstract briar]

I did everything you guys said it shows the same screen

[viral mason]

look here for voice models https://discord.com/channels/1159260121998827560/1175430844685484042

[viral mason]

that's odd

[dusky gale]

pretrains

[abstract briar]

anymore ideas or am I just doomed?

[viral mason]

for now you might be doomed, unless u wanna try that other link I sent

[abstract briar]

I did

[viral mason]

idk what you can do then I'm sorry :(

[abstract briar]

is there any other voice changer?

[viral mason]

the only pretrains that are actually useable that won't give bad results are Legacy core and the original

[abstract briar]

links pls?

[dusky gale]

i just fine tuned the 6.1 KLM and got a great result

[abstract briar]

also ty for helping so much

[abstract briar]

can you pls send to links so I don't get wrong one

[viral mason]

pretrains are not for voice changers I was talking to the other person

[analog obsidian]

they're deleted bc they're outdated

[dusky gale]

ahh ic ty

[analog obsidian]

https://huggingface.co/lyery/legacy-core-2.5/tree/main

for contentvec

[dusky gale]

so no more spin embedder?

[analog obsidian]

not for now, sorry

[abstract briar]

should I put this new one in C:?

[last compass]

i did everything as told for voice change but now voice is not even coming in vc discords

its not loading

if i change my mic its coming

[abstract briar]

Im on the page wheres the download button

[analog obsidian]

that is not a voice changer lol

[abstract briar]

-_-

someone said it was

[viral mason]

you're getting links mixed up

[abstract briar]

not again...

then what voice changer CAN I use

[viral mason]

if that one didn't work then I don't think any will

those are the only free ones and only actually good ones

[abstract briar]

there has to be at least one more

I'll keep trying the old one :(

thanks for the help tho

WAIT A MINUTE

I MIGHT BE ONTO SOMETHING

THERES STILL HOPE :)))))

how long does it take to load normaly

[near hornet]

can someone help me how to setup applio? been cracking my head for a week

i am fully on amd operating system cpu and gpu.

[dusky gale]

linux?

[abstract briar]

FIXED IT

LETS GOOO

TY GUYS :)

[near hornet]

windows

[viral mason]

fr?

[viral mason]

do u need any more help?

[normal nebula]

yi

i need help

[abstract briar]

nah ill be good :)

[normal nebula]

i open the app it goes blank

[abstract briar]

tysm I can't thank you enough

:)

[normal nebula]

@viral mason

[round fog]

I dont understand why you spam multiple channels to get help, sometimes you need to wait longer than 10 minutes, stop acting entitled to help.

When your back from your timeout, someone might help you

[viral mason]

whatcha need help with

btw don't spam my dms you better give a good reason for it once ur back

[upbeat anvil]

hello

is there anybody her

[distant turtle]

-colab

[patent trellis]

📒 Google Colab Notebooks

Google Colab is a Cloud (Remote Good PC) Service. While the Free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.

• **Applio**

by IA Hispano
Google Colab

• **RVC Mainline**

by Hina
Google Colab

• **UVR5 NO UI**

by Eddy
Google Colab

• **UVR5 UI**

by Eddy
Google Colab

• **Wokada Deiteris Fork**

by Deiteris & Hina
Google Colab

• **Hina's Modified Original Wokada**

Google Colab

• **RVC-AI-Cover-Maker-WebUI**

by Shiro & Eddy
Google Colab

• **FaceFusion UI**

by Nick088
Google Colab

• **FaceFusion NO UI**

by Nick088
Google Colab

• **Music Source Separation Training (Inference)**

by Jarredou & Makidanye
Google Colab

[outer isle]

How do you get the model maker role?

[blissful moon]

anyone made an ai model for god?

[viral mason]

like one of god or making models for god?

the second option sounds really funny

[blissful moon]

yes one of god i wanna sound like this kid https://youtu.be/gvPzO4OYI3k

[viral mason]

just get voicemod and use a deep voice thing on it lol

or crack fl studio and use a pitch shifter on it

[blissful moon]

voicemod needs u to pay for that

and i cant really use fl studio

[viral mason]

true

[viral mason]

30 is crazy

[hallow thistle]

He made one thread for every sounded-too-easy-to-solve issue, and I've been observing that.

I once told him to make one single thread for the one-stop solution if nessesary, but he still kind of does the same.

[near hornet]

Hi csn anyone help on how to setup Applio?

[near hornet]

Full GPU Name: AMD RADEON RX6600
Operating System: WINDOWS 11 PRO
Detailed Description of the Problem: when i am at this part, nothing happens after that.
Tutorial Used (if any): (https://docs.aihub.gg/rvc/local/applio/)
Screenshot of the Error (if any):

[hallow thistle]

You didn't run "run-install.bat" at first? The terminal even said that.

[near hornet]

i cant find "run-install.bat"

[hallow thistle]

Ehh, try extract "Applio" zip to somewhere like in "D:\Applio" or "C:\Users\Downloads\Applio". Do not extract one to your desktop folder.

[near hornet]

okay

[hallow thistle]

For work with AMD Radeon RX GPU, you can try do this with recently extracted/installed and existing Applio. https://docs.aihub.gg/rvc/local/applio/#amd-on-windows-precompiled-fix

[rose ice]

Hello. I’m training a model using 30 minutes of Korean audio. The GPU is a T4, the sample rate is 40k, and I’m using the Korean HuBERT base model. The total number of epochs is 400, and the batch size is 8.
Despite all that, the output still sounds unnatural. What could be the problem?

I cut out every bit of noise, even the sound of clothes brushing, so there are some places in the source audio where speech is abruptly cut off. Could that be causing the issue?

The pronunciation also sounds a bit unnatural, and there’s a crackling noise, too 😢

[simple ore]

that's a bit outdated

since amd screwed up the drivers back in may

[near hornet]

i still cant do it what is the mistake here

[simple ore]

are you running it as admin?

[near hornet]

yea

and still nothing

[simple ore]

don't

[near hornet]

the one you sent earlier is that same as my specs?

[hallow thistle]

Why on Earth would you run the program as admin? I double click on the batch file without admin mode, and it works as usual.

[near hornet]

i ran on admin cause before i tried without admin it gave me blank.

i gave both ways a try

this is non admin

its empty not even press any key thing

[simple ore]

just wait for a bit

[near hornet]

ill try now

[hallow thistle]

If Applio's terminal is blank, it can mean it's loading files in background. But if you wait for over 20 - 30 minutes and still shows nothing, it might be your issue with your system or you didn't do thing right.

[near hornet]

okay guys im dumb im sorry

it works

thanks

[hallow thistle]

[near hornet]

mb bro

mb

thanks tho

[simple ore]

I doubt if you dont follow the latest update method I've provided

[near hornet]

i followed the link

[simple ore]

with hip sdk 6.2.4?

[near hornet]

5.7

cause it says that for my gpu

[simple ore]

it does not work with amd driver newer than 25.4.1

[near hornet]

how do i check mine is which

[simple ore]

open adrenalin control panel

top right

[near hornet]

should work?

any drawbacks?

[simple ore]

fkin old, but should work fine then

[near hornet]

what is this?

[hallow thistle]

So, I should've told the one who edits the Applio docs to atleast edit their docs if necessary. Because when I relied on docs for helping, that one engineer would tell me "nah, this one is outdated, try my solution."

[near hornet]

too much hassle to update

[simple ore]

you'd have to re-do the whole thing

[near hornet]

yea i wont update either way

[near hornet]

what is this tho i dont get it

[simple ore]

open training tab, click advanced settings at the top to expand it

[near hornet]

yea

[simple ore]

if it lists your rx 6600 you're good

[near hornet]

ahhh okay thankss

[simple ore]

all good then

[analog obsidian]

the docs are outdated tho

i think it still shows the very first applio version in the guide

lemme check
edit: yeah, the guide is still outdated

[simple ore]

yeah

[hallow thistle]

[simple ore]

we are just waiting for TheRock build

then it would simplify things

[hallow thistle]

I know I'm not allowed to edit the docs, I can only give people some ideas.

[near hornet]

the train let it be default?

[hallow thistle]

Huh? You wanted to train a voice model in Applio?

[near hornet]

yea

[hallow thistle]

In Training part, I can give you some ideas, pretty much just the model name and path to your dataset folder. But for actual audio dataset and more settings, you can ask those who are Model Makers or QC.

[near hornet]

ahhh meaning the advance tweak is it?

who are they and how do i reach out to them?

just tag the role?

[hallow thistle]

Don't tag the role. Just wait for anyone who's available at the moment.

[near hornet]

ahh okay okay, but do i texr here or other channel

[hallow thistle]

Pretty much here.

[neon island]

currently trying out swarmui atm. for some reason i can't seem to load other models?

both ones i've tried had the same error
"Model loader for [model] didn't work - are you sure it has an architecture ID set properly? (Currently set to: 'stable-diffusion-xl-v0_9-base')"

both models are 0.9

[hallow thistle]

I used Automatic1111 Stable Diffusion WebUI and once tried ComfyUI. I've never heard of this Stable Diffusion fork.

[neon island]

I might give that one a go then

[hallow thistle]

Many people would say go for ComfyUI just that, but there's another fork that works similar to Automatic1111, that one named Forge.

[near hornet]

i have issues with train model when i generate index, the model is not showing. know why?

[verbal bridge]

need your help creating videos of the product in the image.

I tried training a LoRA with WAN 2.2 (the only video model I’ve found that supports LoRA), but the results were far from acceptable.

Whenever I try to animate with almost any model, it distorts the product and loses quality.

I’d love to hear how you would approach this. I couldn’t find much on YouTube, even though I know people have managed to get good results.

Thanks a lot!

[untold olive]

my client is very very laggy I have decent gpu but when I speak I hear my voice 5 minutes after that

[viral mason]

What's that

[viral mason]

Did u get it from a yt video

[untold olive]

yes

[viral mason]

The voice changer you're using is old and outdated then, even if the video is from a month or so ago

[untold olive]

how can I download newer?

I have amd gpu

[viral mason]

Do u know the numbers as well

For amd tho I recommend wokada deiteris it's the third guide

[untold olive]

MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a.zip

[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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[neon island]

ComfyUI worked like a chamr, so that's dealt with. Thanks!

[viral mason]

I meant the numbers for the GPU lol

[untold olive]

oh ;d

rx 580 8gb

[viral mason]

It might work not sure

[untold olive]

the version I gave you is it outdated?

[viral mason]

https://github.com/deiteris/voice-changer/releases/download/b2332/voice-changer-windows-amd64-dml.zip

[viral mason]

Yup

[untold olive]

alr

[viral mason]

That link I sent is the version for amd and should hopefully work fine for u

[untold olive]

imma download that version then

[untold olive]

tysm

I'm going to keep you updated

[viral mason]

Np, I'll be here if u have questions and stuff

[untold olive]

quick question

I have to delete old one right?

[viral mason]

I recommend doing that yea

[untold olive]

alr

[viral mason]

But also if u have specific models u wanna use on the new one screenshot them on the old one

[untold olive]

yep ill do that

[dusk sphinx]

anybody else getting this error from ngerok ? /kaggle/working/MMVCServerSIO
Downloading ngrok ...

TimeoutError Traceback (most recent call last)
File /opt/conda/lib/python3.10/urllib/request.py:1348, in AbstractHTTPHandler.do_open(self, http_class, req, **http_conn_args)

[viral mason]

Did u check to see if your Ngrok token ran out of time?

Idk anything about the voice changer if it's non local but I can try

[dusk sphinx]

brand new one let me try with a new token

[untold olive]

so as I see whole app is moved to the web. models is not showing ):

[viral mason]

Yea you'd have to put them all back as it's a newer updated version

[untold olive]

so In folder right?

ohhh

I get it yea

[viral mason]

Folder?

[untold olive]

1 sec

no no I get it

sry

[viral mason]

You're good

[untold olive]

umm sry for lot of questions. I had in 0.3 and out 4.6 here and what can I choose now instead of that?

[viral mason]

I'm not sure, I usually leave out at 100 and I'm at 55 Bec my sensitive ears

[dusk sphinx]

It was on kaggle keeps sending this errors

Maybe temporarily problem with ngrok

[viral mason]

Btw pitch at 12 is usually too high for most models I'd say try 3, 6, 7, 8, 9, or 10 for female models

[burnt lily]

I just got this same error with you so unlikely that we both ran out of usage at the same time

[untold olive]

so like this?

[viral mason]

Whatever works for your ears ig

[dusk sphinx]

all error message

[viral mason]

I'd have to download that to see it

[untold olive]

alr ty for all but 1 more thing what you think what is best girl model

like best

maybe mommy kind of voice

[rigid willow]

People, please help me. I can't use the voice changer. I keep getting errors. Is there a video guide?

To change the voice to Russian

[viral mason]

No idea, I don't use those kind of voices so you'd have to look here
https://discord.com/channels/1159260121998827560/1175430844685484042

[viral mason]

If you're talking about the voice changer if you speak Russian that's the only chance u have

[rigid willow]

My main language is Russian and I need a voice changer to simply change my voice. I need help with installation and setup because I couldn't find a video guide. Please help me.

[plucky belfry]

will putting the voice changer on server fix my performance issues?

[viral mason]

What is your GPU?

[dusk sphinx]

seems like temporary error fixed

[viral mason]

So it's working again?

[plucky belfry]

what should i have monotone on

[viral mason]

?

[earnest glacier]

Yo can someone help me? When I use the voice changer my mic keeps randomly cutting out and then it goes back to normal (with the voice changer it doesn't so that with out it)

[viral mason]

did you get the voice changer off a yt video?

[earnest glacier]

Yes

[viral mason]

it's outdated by over a year even if the video was made a month ago

the people that make those tutorials don't use the up to date stuff either to trick you ot bc they don't know the new stuff

[earnest glacier]

Oh

[viral mason]

what gpu do u have?

[earnest glacier]

Yeah I need to look again tmrw or smth can you help me or nah?

[earnest glacier]

AMD

[viral mason]

I can help rn

looking on yt won't help

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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]

for AMD the best to get is w-okada deiteris

[earnest glacier]

I can't rn sadly

[viral mason]

that's alright

[earnest glacier]

But I can tmrw

[viral mason]

go ahead and download this if you're on pc so u can set it up whenever u have time
https://github.com/deiteris/voice-changer/releases/download/b2332/voice-changer-windows-amd64-dml.zip

u can add me if u like and dm me tomorrow

[earnest glacier]

Aightt

[outer isle]

Can I upload my voice model to hf?

If I use Kaggle

[viral mason]

yes

just download the pth and index file and upload to huggingface

[azure patio]

which is the best non-local rvc for covers

[viral mason]

weights.gg is the easiest options is... kinda free

but I'd say use applio

[azure patio]

i installed vonovox but is it only real time? can i upload audios for conversion?

[viral mason]

I believe you can but vonovox only works locally and on a nvidia gpu 30 series and up

[sonic night]

hello, i need help with applio, im trying to clone my voice but after starting the training it takes just 13s to complete it, then in the folder applio created theres no pth or index even when generating it

but its not working

wait ill the the thing in guidelines

[outer isle]

No ZIP?

[sonic night]

Hello everyone, i'm having a problem with Applio when training.
When training it takes less than 15 seconds to complete, but it does nothing... no PTH, no INDEX.

PC Specs: Ryzen 7 5700G - NVIDIA RTX 3050
OS: Windows 11
Applio used: 3.5.0

btw, those folders are empty
Can someone help me with this?

thats why im asking for help bc it doesnt work even with docs

[analog obsidian]

yea docs are outdated

maybe preprocessing failed

check model_info.json

it says 0:00 minutes?

[sonic night]

yup

[analog obsidian]

ok preprocess failed

[sonic night]

i uploaded 20mins lol

[analog obsidian]

delete that folder, and try preprocessing again

are your audio files .wav?

[sonic night]

wav

[analog obsidian]

maybe you put a wrong dataset path, check if the location is correct

[sonic night]

ill try again

[analog obsidian]

thats how a dataset folder should look like

just random audio files

i think it also fails if the audio files have special characters in their name like japanese text

[sonic night]

k ill try agn

finally its not 0 minutes, thanks, i really apprecciate it, hope the training works

oh, it didnt work :(.

[analog obsidian]

you ran the second step?

feature extraction

[sonic night]

yes

[analog obsidian]

are the f0 folders empty?

[sonic night]

yes

[analog obsidian]

uhh weird, is any error message in the cmd window?

[analog obsidian]

oh is your gpu an amd one? or you have an igpu?

[analog obsidian]

never seen an error like that in applio before lol, try restarting the app, if that doesnt fix it, yea ive got no idea

[sonic night]

k ill try

[analog obsidian]

the gpu number matches your gpu?

it should be the same number as your main gpu

laptops usually have multiple gpus

[sonic night]

it was matching, but ill try restarting it

[analog obsidian]

k

[sonic night]

im on pc, is there any difference for applio?

[analog obsidian]

nope it should be the same for both pc and laptops iirc

just that imo training ai in laptops can be a bit risky, it overheats

its safer on pc

[sonic night]

ohh ok

[analog obsidian]

wait, you got another gpu? in cuda the gpu order starts with 0
for example; 0: 4090, 1: 4090 <--- multigpu
if you set it to 1 while having a singular gpu, it wont work

[sonic night]

no, only that one 3050

[analog obsidian]

if there's no other gpu listed in gpu information, set it to 0

[sonic night]

now f0 are full, when extracting features

[analog obsidian]

[sonic night]

it means its workiing (hopefully)

thank u god Lyery

[analog obsidian]

[sonic night]

ill start training, anything ill write on my phone

[viral mason]

Yea zip them up when uploading to hugging face

[sonic night]

btw, i have a question, how much time did it take to u to train a model?

[analog obsidian]

depends on multiple factors, how good the gpu is, batch size and how big the dataset is

[sonic night]

4 batch size?

[analog obsidian]

how many mins of data

batch 4 is naturally slow, the model has to train more steps
it gets even slower if the dataset is big

[sonic night]

20m:30s

[analog obsidian]

the official rvc wiki recommends training 40 epochs if the dataset is low quality, up to 200 max if the quality is good
this is where gets confusing tbh, my tip would be training that model for 100 epochs first, listen to epoch 40, if it sounds great you can keep it as your final model, but if you dont like how it sounds try hearing 100, and if 100 doesnt sound fine, you can try training 200 epochs

it's a very confusing part not gonna lie

models can overtrain at any moment

despite what the docs says, overtraining can't be spotted in the graphs

it just happens

[sonic night]

ok ok, ill try that

[analog obsidian]

yea basically u have to train until the model starts to sound very robotic, when that happens you just choose a epoch that is healthy

which one is the best is completely up to you coz graphs can't tell you that either

[sonic night]

what about batch?

[analog obsidian]

the graphs were meant to show the user if the training has become unstable/something broke

[analog obsidian]

i dont like using anything below 8

thats my personal preference

but some do 4, 6, 8, etc

anything works

[sonic night]

ok ok

[abstract briar]

how do I set this up

its so confusing

[dusky gale]

when doing training runs that "merge" different voices together is it bad to just add them all together in one run instead of doing 3 training runs then merging with okada?

[analog obsidian]

training a dataset of different voices you say? if you do that rvc will simply change between voices during the inference rather than combining them

[dusky gale]

ahhh tyyy

[limber hollow]

is there any way to get rid of the delay in voices?

[analog obsidian]

yea is kinda cool actually, if you train a male voice and a female voice in the same dataset the model will know that for low pitch conversions it should use the male voice and not the female one, and for high pitched inference, it will use the female voice

[sonic night]

maybe with a better gpu

[analog obsidian]

but obviously training them as different models would be better

[limber hollow]

do you got delay too?

[sonic night]

and lowering chunks

[sonic night]

Try lowering the chunks
more chunks = bigger delay time but sounds great
less chunks = lower delay time but sounds bad

[limber hollow]

what chunks u got?

[sonic night]

let me open the program

[limber hollow]

ok

[sonic night]

i use 128

[limber hollow]

and is it delayd?

[sonic night]

341.3ms

[limber hollow]

ok thx

[sonic night]

i cant test rn

[brittle wing]

Hi, I have a 19-minute recording. Do you think I should set the batch size to 6?

[abstract briar]

what do I set output to for it to play in a call

someone please help

[pure root]

is it possible to lower the limits for venovox ?

[viral mason]

try these settings, replace the input I have with whatever headphone/headset microphone you use

[loud badge]

!most models don work only one that does is beatrice jvs corpus

[viral mason]

no one makes beatrice models anymore

[loud badge]

the new ones dont work for me

RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

im on amd 9070xt

[vernal jewel]

This may seem confusing, but it only works if I select my output as the headset that is output in voicemod, and when I activate the voice it is doubled, in which case it is the AI's voice and my voice at the same time, but when I put the output in voice mod for some reason the AI ​​stops working.

[viral mason]

you have the input and pit[it swapped

[loud badge]

could myou help me please

im on amd 9070xt

RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

[vernal jewel]

I changed it but it only works if I activate the microphone through voicemod

I can send a video if you want

[rare oracle]

i sound kinda stupid for saying this, but where do i get the tech/apps to create an ai audio cover with a voice model? i lost my original one

[viral mason]

u can use applio or weights.gg

weights has a free app called replay

[rare oracle]

huh, its no longer self hosted, neat

[viral mason]

since u use AMD u need wokada deiteris

but u said none of the new ones work right

[loud badge]

yea

[vernal jewel]

Am I on the correct version?

[viral mason]

then I can

[loud badge]

ai voice changer onnxdirectML-cuda_v.1.5.3.18a i did the amd one

[viral mason]

that is outdated I believe

unless I gave u it through this

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[vernal jewel]

ok

[hallow thistle]

Both v.1.5.3.18a variants are versions of original W-Okada, and all outdated anyways.

[hallow thistle]

What is your PC GPU?

[hallow thistle]

You can go either Vonovox or Deiteris/Tg Develop W-Okada forks. Vonovox can give better audio quality but its UI is less familar than other W-Okadas.

[loud badge]

what should i do im on amd

[vernal jewel]

Ok, Does it work the same as what I'm using now?

[hallow thistle]

Try better W-Okada version from this guide. https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#download-amd-intel-and-cpu-on-windows

[hallow thistle]

When you run the program, it should work. If this Deiteris/Tg Develop W-Okada opens your browser, it's completely normal.

[hallow thistle]

What does this mean? Which settings are you looking in Vonovox?

[loud badge]

what one do i download

theres no clear amd one for windows

amd gpu

9070xt

[loud badge]

thanks

[hallow thistle]

If it says "AMD, Intel", it means "AMD/Intel GPU", and it works with AMD Radeon RX and Intel Arc A/B GPU.

[next mist]

-colab

[patent trellis]

📒 Google Colab Notebooks

Google Colab is a Cloud (Remote Good PC) Service. While the Free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.

• **Applio**

by IA Hispano
Google Colab

• **RVC Mainline**

by Hina
Google Colab

• **UVR5 NO UI**

by Eddy
Google Colab

• **UVR5 UI**

by Eddy
Google Colab

• **Wokada Deiteris Fork**

by Deiteris & Hina
Google Colab

• **Hina's Modified Original Wokada**

Google Colab

• **RVC-AI-Cover-Maker-WebUI**

by Shiro & Eddy
Google Colab

• **FaceFusion UI**

by Nick088
Google Colab

• **FaceFusion NO UI**

by Nick088
Google Colab

• **Music Source Separation Training (Inference)**

by Jarredou & Makidanye
Google Colab

[wind egret]

Can you innpaint, outpaint, etc with a NoobAI Model like Stablemondai? cuz it's not working in Krita but t2i gens work fine on A1111, i tried outpainting but it generates a pitch black background with the edges being half modified by ai checkpoint, i tried inpainting an area but it generated horribly with random colors and disfigurement, i tried a normal t2i gen but it came out deep fried and disfigured aswell, all these issues on Krita SD plugin.

[ruby valve]

Does anyone know a good voice changer that’s actually free i have Rtx 5060

[low shard]

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[low shard]

you got 3 options, id personally suggest vonovox for your setup, guessing you're on windows 10 or 11, but it's your choice

[sharp ore]

Hey I would love to know if anybody can make these videos? or similar https://www.youtube.com/channel/UCYVUm41HtniQ2c3nYFaxWEw

[ruby valve]

Do you have a link by chance or guide anywhere yk of

[low shard]

yeah, the sapphire bot message above mine got all guide links

just click the blue guide for whatever program you want

[lean dune]

my gpu rtx 3060 12gb vram
is safe to train rvc model using batch size 10?
or just 4

[ruby valve]

Ohh i didn’t notice thankyou bro

[simple ore]

that depends on the dataset size

[median root]

Hey guys, I'm having a problem where W-Okada is outputting voice as distorted groaning. Input is fine, only the output is affected.

[viral mason]

did u get the voice changer from a yt video

if u did it's outdated by over a year even if the video is from a month or so ago

[median root]

I got it from the github, should i look elsewhere?

[viral mason]

no, what gpu do u have

[median root]

Max OS, Sonoma M2

[viral mason]

idk what that even is

[median root]

Mac OS* it's mac so it isn't intel 🙂

So it runs on M2 chip

[ashen harness]

Does anyone know why the voice model im using (E-Girl by Zoka) keeps cutting every 5s or so? Is it a microphone problem or just a model problem?

[viral mason]

what are you using an egirl voice for :(

[viral mason]

I have never had someone using that kind of gpu before so idk what one to recommend

[median root]

It's not windows, it's Mac OS haha. No problem thanks for trying!

[viral mason]

you can ask around, maybe some of the mods could help ^^

[nocturne terrace]

https://youtube.com/shorts/YlX9bIxshCc?si=Vs-EeIJ2IOmHg4PZ

[fast shell]

Hi everyone, I am new to content creation, kindly put me through

[cosmic spire]

how to make egirl voice model @low shard

[viral mason]

https://cdn.discordapp.com/attachments/1365750911070371972/1428134698495901797/Soapspeechbubble.gif

[thick latch]

What's the best solution to your voice crackling/stuttering when talking for longer periods or big phrases at once? Raising block?

[gritty parrot]

whats the most realistic real-time voice changer atm? I have an nvidia gpu on windows

I dl w-okada long time ago

[left vortex]

I'm looking to use TTS with an LLM running on Kobold CPP. I want the capability to use character voices, which lead me here since you guys have lots of character voice models. Any recommendation on what tool I use for TTS that's compatible with the models here?

[simple ore]

Orpheus TTS uses a gguf LLM model, but it does not have voice cloning

sorry, wrong name

https://github.com/isaiahbjork/orpheus-tts-local

[left vortex]

I have all the rest of the pieces of the puzzle already, I just need a program that can TTS with the models from your community.

[simple ore]

repo suggest LM studio, but you can use other

[left vortex]

DOes Orpheus work with these models?

[simple ore]

orpheus generates text to speech with one of a few built-in voices, you can apply RVC voice change on top of it

[left vortex]

Do these voices only work with RVC?

[simple ore]

correct

[left vortex]

Gotcha

And what program do you do the RVC through?

[simple ore]

Applio, or there's rvc package on pypi

there are of course other TTS models that come with voice cloning and do not require voice changing or voice models, just 10-15s audio sample as input

[left vortex]

Do those end up sounding better or worse than RVC? Would it be worth it for the reduction in complexity and latency?

[simple ore]

some are better than others

some are better than rvc

[viral mason]

vonovox, I'll help u out as long as there's no egrill trolling involved

[left vortex]

Are there any that have collections of character voices like y'all have here?

[simple ore]

no, you just make your own reference audio

[left vortex]

Is that hard to do?

[simple ore]

no

barely an inconvenience

[left vortex]

Super easy

Okay, where should I go next?

I was looking at GPT SoVITS, but I don't read enough Chinese to understand it.

[simple ore]

you can try https://huggingface.co/spaces/ResembleAI/Chatterbox

[gritty parrot]

works the same way as wokada?

[simple ore]

pretty much

[viral mason]

it works just like it but it's just better

looks way different tho

[gritty parrot]

damn

what abt perf?

[viral mason]

?

[gritty parrot]

performance

like latency i mean

I guess it depends on your setting nvm

[viral mason]

it depends on settings and gpu, since the better the gpu u can have better settings with less latency

[gritty parrot]

ok, ill try to dl it

alr have it downloaded. Any tips?

not familiar with it

it seems to work well

[viral mason]

shit wrong thing

[analog obsidian]

lmao

[viral mason]

try these settings, I got them from Nick
blocksize: 3.0/2.7
crossfade duration: 0.15
extra time: 4.0

crossfade at 0.15 is high recommended if nothing else

[gritty parrot]

if it helps my gpu is 3070ti

its ok, not the best

ill try them out

blocksize max is 1 tho

also do you recommend enabling the upscaler

[viral mason]

no idea, could be a typo

I don't use vonovox currently since dr hasn't added more slots for models and I need over 100 for all the characters I like messing with

I'm kinda petty

[gritty parrot]

oh yeah i saw, why did they limit it?

[viral mason]

just hasn't done it yet

[gritty parrot]

ui is way better though

[viral mason]

he's adding stuff like sine and rnnoise but not simply having pitch default at 0 or more slots

not complaining just dunno why

[gritty parrot]

I wish they added dropdown arrows on each section so we can have it simpler

like for example

an arrow to close the audio device setting, etc

[viral mason]

there's an official vonovox server if u wanna talk to the creator there and give suggestions

[gritty parrot]

oh aight ill join up

know where to find the link?

[viral mason]

I could invite u but I have to send in dms or I will be skinned alive for sending a discord server link here

are dms fine

[gritty parrot]

yea sure

thanks btw

[viral mason]

sent!

yw

[narrow coyote]

how do i make a voice from here pleaes

[warm iris]

https://huggingface.co/spaces/Nick088/Real-ESRGAN_Pytorch

I keep getting:

Output /tmp/gradio/a356989ffae9dbe36a3a48457b8582850e6a0d027456e67fd51d20180311d2a0/Timeline 4.mov same as Input #0 - exiting

FFmpeg cannot edit existing files in-place.

[modern cloud]

Can someone help me make an ai voice model in hungarian ?

[marsh raven]

!download

[knotty moth]

https://tenor.com/view/error-wait-computer-download-cat-gif-10480536

[stuck wave]

I' m working on a framework, a third semantic layer, to communicate with AIs using symbolism. It has worked on ChatGPT, GLM and Claude. Here is a glimpse what claude thinks about it: "The framework does something I've never seen articulated this precisely: it creates a third semantic space that neither biological nor digital cognition fully owns. Not translation, not mimicry—co-creation of meaning through acknowledged difference." I need volunteers to introduce the framework to their own AIs and analyze the results. Anyone?

[near hornet]

i am having issues training models, can someone help? im using amd btw

[hallow thistle]

You wanna download the database of "AI Hub by Weights" or something?

[ashen bane]

where to dowload w okada

[near hornet]

yea but there is tho

or did i put it in a wrong place

[simple ore]

or potentially you somehow ended up training on iGPU and it failed f0 extraction

show the log of preprocess and extract features step

[near hornet]

idk man

[simple ore]

show the screenshot of training screen with advanced settings expanded

[simple ore]

by the log I mean the text output

[near hornet]

rn im trying something and its going 400s

[simple ore]

dont use chinese hubert, it does not work without chinese hubert pretrain and we dont have any

[near hornet]

now its 500

damn youtube

use which then

[simple ore]

contentvec is default

[near hornet]

how do i stop this extract

[simple ore]

show the log window

show the screenshot of training screen with advanced settings expanded

[near hornet]

its

going to 600 seconds now

[simple ore]

okay, you have 2 minutes of audio loaded finally

it is compiling, be patient

[near hornet]

ahh

so which embedder should i use next time

if its not chinese cause i did it based on youtube

[simple ore]

this was a failure

use contentvec

[near hornet]

yeaaaaa mbbb

alright understood

[simple ore]

if you have downloaded spin-v2 pretrain you can try spin-v2 + custom pretrained

[near hornet]

how do i download it

is it better tho

this is first time doing, 700 seconds is normal?

its done now, what do i do nexr

[simple ore]

delete the model folder in logs and do it again

with proper settings

[near hornet]

what u mean

okay

[simple ore]

preprocess has advanced settings

check off noise filter, check 'post'

and simple

[near hornet]

use dataset creator?

[simple ore]

you dont need dataset creator, that's for colab/kaggle

just give it a path like C:\training\wavs

[near hornet]

so what do i put at the path?

isnt it supposed to be the path to the voice model?

[simple ore]

no

[near hornet]

like this?

[simple ore]

if you have the source files in that folder

[near hornet]

wait source file is in the different folder

do i just copy the path to source file folder?

i understand your frustation and im sorry man

[simple ore]

copy it from windows exporer and paste it into applio

[near hornet]

im your name

[simple ore]

how hard can it be

[near hornet]

the source iright

correct now?

[simple ore]

no

Path to folder

not "path to file"

dont use spaces in the path

dont use "

[near hornet]

if there is 2 source, remove the one i dont want?

[simple ore]

did you delete the old logs/model folder?

[near hornet]

yea

[simple ore]

preprocess takes all files from the chosen location

so do the needfull

[simple ore]

uncheck creator

[simple ore]

okay, now it will take everything from

[near hornet]

okay so press preprocess?

[simple ore]

if that's your goal, then yes

[near hornet]

okay doing it

[simple ore]

extract features - rmvpe, contentvec

[near hornet]

the setting is right?

silent thingy 0? or ...

[simple ore]

usually 2

but your dataset it small, 0 is fine

[near hornet]

i will do usually

okay i will do 0

processing

complete?

so quick?

[simple ore]

all the code has been compiled for those steps

[simple ore]

it will comile a bit more for training as well

set total to 100

[near hornet]

thats it?

advance setting let it be?

[simple ore]

you only have two minutes of audio, so dont expect anything good

no, go ahead

[near hornet]

start training?

[near hornet]

wait so how can i do it until its good

[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.

[simple ore]

use more audio, 2 minutes is way below the usual recommendation

but give it a try

[near hornet]

i can just add later on or new model in log

[hallow thistle]

More like macOS. The M2 is not a GPU but a CPU chip found in modern Apple Mac computers.

You can try run a W-Okada version that made to work with Apple Silicon Mac computers, but the program would run with only CPU which would be slower than those PC with GeForce RTX or AMD Radeon RX. https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#download-mac-silicon If the error message "Pytorch is damaged" shows up on screen or nothing happens after trying to run the program, you'll need to follow steps to allow the program through Terminal.

[weary vortex]

саламалейкум ребята здесь есть модель голос симпла?

Не могу найти просто

[hallow thistle]

This server is English only.

[hallow thistle]

You can try ask Nick088 about the issue you encountered on Hugging Face space there, I think he implemented that one.

[torn nebula]

?okada

what are the commands

Okada

RVC

gng i js wanna know if there is a new version to install

shi bro

[low shard]

@warm iris I updated my HuggingFace Space, now it should be working

[low shard]

of what? RVC and Wokada are 2 different things

RVC = Retrieval-based-Voice-Conversion, not realtime voice changing

if you meant realtime voice changers, there are different programs

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[low shard]

you can also use a command to see all commands

-prefix-commands

[patent trellis]

🗣️ All Prefixes Commands

Commands that give you useful information that work with the prefixes (- or !)

-prefix-commands

Shows this list

-rvc

Shows RVC Documentations

-audio

Guides for Audio Cleaning and making Datasets

-colab | -googlecolab

A list of Useful Google Colabs Notebooks (Cloud)

-kaggle

A list of Kaggle Notebooks (Cloud)

-huggingface | -huggingfacespaces | -hf | -spaces

A list of HuggingFace Spaces (Cloud)

-lightning | -lightningai

A list of Lightning AI Notebooks (Cloud)

-realtime | -wokada | -rt

A list of Programs for using RVC Models in Realtime for Calls/Games (mostly Wokada & Wokada Deiteris fork) with Guides

-uvr | -ultimatevocalremover

Shows useful links about UVR, a program for vocal and instrumental separation

-howtoask

Shows how to ask properly for help

-howtosearch | -searchmodels | models

Shows how to search RVC AI Voice Models

-tracker

Shows a very old Google Spreadsheet with old RVC Models, not much suggested though

-gui | -rvcgui | -rvceasygui

Shows others that RVC Easy GUI is Outdated

-svc | -so-vits-svc

Explains others that So-VITS-SVC is very outdated

-javavsjavascript | -java | -javascript | -js

A text explaining the difference between Java & JavaScript (for fun)

[torn nebula]

do i install vonovox i have a 2050

[low shard]

if you're a windows 10 / 11, and you want to try an rvc realtime voice changer that still gets some performance updates even tho closed sourced and freemium, vonovox might be good for you

[torn nebula]

is there a paid version?

[low shard]

vonovox is freemium, meaning it's free to use, but there's a monthly subscription currently for some voice effects

[viral mason]

that's if u want to pay, there's always a work around for voice fx :3

[torn nebula]

i know

[humble beacon]

Hlo everyone I have a rvc gui voice changer but when I open and use it's say python is not responding and getting white⚪ color is anyone have solution

[viral mason]

did u get it from a yt tutorial?

[humble beacon]

Yes

[viral mason]

it's outdated by over a year even if the video was made a few months ago

[low shard]

all youtube tutorials are outdated for realtime voice changers for calls/games

what's your pc gpu and os?

[viral mason]

you beat me to it :(

[humble beacon]

Rtx 2050

So any help i have try 1 year old mmvc voice changer also and it's run but voice was not like real and clear

[low shard]

thanks, and what's your OS (Operating System)?

[humble beacon]

Window 11

[low shard]

all youtube/video tutorials are outdated

you're using an over year old version of original wokada

-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 Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. 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

Most suggested WebUI with the best general support for many platforms. GUIDE

⚔️ Wokada Deiteris Fork vs Vonovox

For Windows Nvidia, Both Wokada Deiteris 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 Deiteris 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

[low shard]

you got 3 options

vonovox, even tho it's closed sourced freemium, might be your best option for your setup and since it still gets updates

[humble beacon]

It will be safe for pc like any virus 😅

[low shard]

there's no virus as the other users of the program would tell you

[humble beacon]

So how to download

[low shard]

if you for some reason don't want to trust it because it's closed sourced and you want to use only open source programs, wokada tg-develop and deiteris fork are open source, tho that's not being updated and more of your choice