🧬│ai-chat

[tired juniper]

Hello guys. I thought the server was banned and never be back again 😅

[devout mango]

Hi

[tardy vector]

hi

[silk linden]

Hi, do you guys know where this text comes from?
https://cdn.discordapp.com/attachments/1236217350345330719/1236217351075139604/audio_sample2.wav?ex=663dcc20&is=663c7aa0&hm=ca1183342d241d67496f63140b965000d47d0c4178f5ddcf759fe69685f2a00a&

i herd it a lot in the samples and idk where its from

[elder willow]

Hello everyone o/

[torn tulip]

??

[feral halo]

Ayo? @torn tulip level 1 !!!

[tardy vector]

what?

[torn tulip]

u a furry gang?

[tardy vector]

yes

[torn tulip]

hell na

[tardy vector]

ok ban

[torn tulip]

jus cus i dont fw furries

[elder willow]

I know that it's not exactly a DC server organised around that kind of AI but does anyone know if there are any websites, services or other DC servers that support storytelling AI for let's say a text game or interactive fiction? I assume that it's possible.

[tardy vector]

Well you dont have to like them but you also dont have to be an asshat about it tbh

just be nice and youre 99% fine

[elder willow]

hello

[torn tulip]

fien, fibe

[tardy vector]

hi

[torn tulip]

fien mentioned rahhhhh

[tardy vector]

ill make you a furry

[torn tulip]

so just cause i shared my opinion on furries im an ass

[tardy vector]

imagine you did the same for gay people

[torn tulip]

average liberal

[tardy vector]

or trans people

or lesbian people

[torn tulip]

nah jus furries tbh

[tardy vector]

and why?

[elder willow]

So why are we even discussing this subject in AI discord server?

[feral halo]

Ayo? @elder willow level 1 !!!

[tardy vector]

cuz

[elder willow]

I'm pretty sure that 1 any kind of discrimination is against this servers rules and 2 discussing these matters in the first place is a breech of non-nsfw rule...

[amber zephyr]

I just joined the discord and I'm trying to download a voice model but it won't let me

[tardy vector]

wdym a breach of non nsfw rule

arguing is nsfw?

[elder willow]

Guys can't we all just get along pls?

[tardy vector]

i wonder but i guess not

[elder willow]

See this is why I'm not active here other than to post models

[zenith hamlet]

Probably a teenager

[elder willow]

Ow, the edge

[tardy vector]

ah yeah- someone who doesnt understand respect yet.

sad

[spice glacier]

hi guys im not robot

[tardy vector]

hello guys im not robot

[elder willow]

Hi not robot, I'm dad!

[zenith hamlet]

I also did not fw furries,now one of my fav content creator is a furry lol

[elder willow]

I've had friends that were furries and they were pretty chill

[tardy vector]

furries are just nice in general

except imo furries who are so much into being a furry, they identify as a dog and shit outside

but lets not talk about that

[elder willow]

Sure, some can be questionable but not every furry is like that

[tardy vector]

i just like "being" a furry because the art is cute and yeah thats it

[zenith hamlet]

There's a turkish proverb, where's too much, there's shit

Basically a translation for many chefs spoil the broth

Or whatever I don't remember correctly i think

[elder willow]

I'm a semi-furry, I just like the art cause it's cute and funny

[tardy vector]

oh and a lot of furries just nice

[elder willow]

Exactly

[tardy vector]

:3

My friend was really nice today and made a lora of my fursona

Can't thank him enough

[lofty sable]

[half horizon]

guys i remove separate instrumental from vocal but the vocal still have some litle instrumental sound is there a solution to that ?

i used mvsep. com

[elder willow]

Hmm idk

Which model did you use? Some do better at leaving those out than others

[half horizon]

ver 2024.04 (SDR vocals: 11.24, SDR instrum: 17.55)

BS Roformer (vocals, instrumental)

[elder willow]

Ahhh i see

Maybe a denoising algorithm might work? Idk for sure but that may do the trick

[half horizon]

u mean i do it for the vocal ? or for the whole music from scratch

[elder willow]

For the vocal

[latent carbon]

@pseudo kettle come ova here

[half horizon]

btw there is no denoise algoritm in list

[elder willow]

Oh

I think UVR has one if your gpu is able to handle it

[latent carbon]

Damn people be so mysterious having the status invisible 😎

[half horizon]

i cant do this i only work on no cpu or online platforms my laptop 4g ram win 7

[amber light]

https://colab.research.google.com/github/Eddycrack864/UVR5-NO-UI/blob/main/UVR5_NO_UI.ipynb

[elder willow]

Ahh I see

[latent carbon]

Nah fam u a bot , not that I wanna be mean but Ken Carson Gaming be chilling

[feral halo]

Ayo? @latent carbon level 1 !!!

[elder willow]

How am I a bot? I'm just as real as you

[latent carbon]

You be saying the same stuff on repeat idk

[elder willow]

It's just how I talk dude how hard is it to grasp 😭

[half horizon]

what is better bsroformer or mdx23

or what difrence

[elder willow]

BS roformer is pretty good from what I've tested

I've used MDX23C and the output vocals were somewhat alright

[amber light]

bro said ken carson gaming

[latent carbon]

@pseudo kettle he bro he

[humble minnow]

is there a working inference code for that?

[amber light]

https://tenor.com/view/ken-carson-laughin-gif-26742463

[elder willow]

Not that I know of

I usually just use mvsep for that

[latent carbon]

Nah that not him , he a face changer

[spice wasp]

is it somehow possible to use the new bs-roformer uvr model codewise?

[elder willow]

From what I've seen no

[amber light]

i know lmao

[elder willow]

It would be cool if they implemented it though

[amber light]

ken carson in ai hub???

[spice wasp]

absolutely, thanks for the reply

[half horizon]

i was using uvr5 to separate audio then they gave me this error
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

[half horizon]

i had this error :RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

[amber light]

#✨│ai-help

[terse oak]

@grim shuttle stop sexually harassing people in medieval PVP

I’m sick of it

[amber light]

what

https://tenor.com/view/hotaru-ichijou-hotaru-non-non-biyori-embed-fail-epic-embed-fail-gif-21797844

[jolly kiln]

terve

tai hei

[feral halo]

Ayo? @jolly kiln level 1 !!!

[jolly kiln]

[narrow flume]

hey guys, random question. Is applio safe to use or no?

[plush egret]

[narrow flume]

that dosent answer my question

[amber light]

yes

[narrow flume]

really?

because i got suspicious of it when i saw i need to run a bat file

[amber light]

it’s safe to use yes

-guide

[rare sorrel]

Documentation

📚 All-In-One English documentation

https://docs.aihub.wtf/

❔ English FAQ

https://docs.aihub.wtf/faq

Credits: AI HUB FRANCE

✨ More guides

Full AI Voice Model Training Guide (Local)

Link: YouTube
Credits: Christopher Villanueva

Model training with Mainline RVC

Link: Rentry
credits: Raven (ravencutie21)

AICoverGen Colab Guide

Link: Google Docs
Credits: Eddy (Spanish Helper)

Create a model with RVC disconnected (colab)

Link: Google Docs
Credits: Angetyde

How To Make an AI Cover With Ilaria RVC

Link: Rentry
Credits: 👽 Julia (ailen2091)

[amber light]

go to the first link and search applio

[gleaming bluff]

hello

how do i verify

[amber light]

to get the verified role you gotta reach level 5

[gleaming bluff]

lol

thanks

[terse oak]

@grim shuttle stop sexually harassing people in medieval pvp

Fuckin hell

You need to fuckin stop

Saw a whole compilation of you sexually harrassin kids

[latent carbon]

@pseudo kettle pull up bro

[knotty zinc]

https://youtu.be/gHSDvNEvwZQ?si=UOZXyHD9N6OM5emS

[feral halo]

Ayo? @knotty zinc level 2 !!!

[crude mural]

uwu

[stiff rune]

does anyone know if there is a way to enhance live preformance vocals? it doesn’t matter if I use the cleanest preformances, the model still is blurry.

[feral halo]

Ayo? @supple briar level 1 !!!

[amber light]

dawg what are you on about

[terse oak]

He was sexually harrassin some minors

On a online game

🤷

[zenith hamlet]

Since when drake is in here

[terse oak]

Drake?

What the sigma

[zenith hamlet]

Kendrick drake beef,kendrick called drake a pdf file

[terse oak]

Nahhh

That’s wild

@grim shuttle

Come back

[jade oak]

We can use this on mobile ?

[forest quarry]

Realtime no, but if you wanna make covers and stuff you can always use Colab

https://docs.aihub.wtf/

RVC > Cloud section has some good Colabs for you to make AI covers n stuff like that

[forest quarry]

QUICO

QUICO REAL

[jade oak]

Or you can find me or create me a clip where sasuke say ryoiki tenkai pls ?

[feral halo]

Ayo? @jade oak level 1 !!!

[forest quarry]

you can do that probably

[candid ice]

hello

https://tenor.com/view/iñi-toñito-toño-smile-flying-kiss-gif-16870336

[forest quarry]

ai hub ain't complete without soup, welcome to the serverrr

[candid ice]

ty 😭

[feral halo]

Ayo? @elder willow level 8 !!!

[half horizon]

i train anew voice on llaria disconected google collab

my question is the audio on data set how long must be max ? can i use 30 min audio ?

[forest quarry]

WUIC

[forest quarry]

30m can work

[half horizon]

what is max ?

i saw somone said up to 20 min so i want be sure before prepraring my data set audio to make it long as possible and dont pass limit possible

[forest quarry]

it all depends on the GPU

There isn't really a "max" but I think after 10h it would be so unusable that you'd be training 1 epoch every 3 hours

You'd be straining it too hard

[half horizon]

because i use no cpu

[forest quarry]

This depends on the GPU used ofc

[forest quarry]

Not necessarily

[half horizon]

what mean epoch

[forest quarry]

I think 1h is already unnecessary

[forest quarry]

Cycle of training

That's when it trains the entire dataset once

so 200 epochs = trained the entire dataset 200 times

[half horizon]

just im looking for good model result

[jade oak]

It doesnt work

https://tenor.com/view/crying-crying-meme-gif-26327776

[forest quarry]

gimme details

[elder willow]

Real ✔️✔️✔️✔️✔️

[fleet veldt]

Is it just me or is weights.gg down?

nvm

[jade oak]

The first link dont work the second don t work for mobile bc it tell me to use a pc app and the lask link tell me to join this discord

[forest quarry]

https://colab.research.google.com/github/iahispano/applio/blob/master/assets/Applio.ipynb should work

unless you're doing smth wrong

[jade oak]

How i downlad applio ?

[forest quarry]

you don't need to download

it's all on the cloud

[safe current]

Zed

[compact hull]

.

[crisp gorge]

Uh what happened to weights.gg

[stark scarab]

https://github.com/lucidrains/BS-RoFormer
https://github.com/ZFTurbo/Music-Source-Separation-Training

[jade oak]

Come pm

[feral halo]

Ayo? @jade oak level 2 !!!

[stark scarab]

UVR on Colab?

[nocturne zephyr]

It’s down again

[crisp gorge]

Oh

[nocturne zephyr]

You need to ask BeatsBoy

The Creator of Weights gg

[forest quarry]

mmmmm

wasn't it River, actually?
I could be wrong idk

[half horizon]

yes u can see link

[buoyant rapids]

yelo I need some steam points, if anyone got spare points to spare can I get some.

[amber light]

i got 20k

[buoyant rapids]

o.O

[feral halo]

Ayo? @buoyant rapids level 1 !!!

[buoyant rapids]

U flexing on me

[vivid wasp]

Hey guys, I need a voice model for a product advertisement, those videos that show a product with a voice talking about it. Does anyone know of a good model for this type of video?

[forest quarry]

Try looking for a voice model that won't bring you legal trouble, there are some uncopyrighted free models out there

[glad owl]

Does anyone have any alternatives to w-okada that can be used by people who have Intel Drivers?? I would like to use a voice changer for my streams, but I need a working voice changer.

[fallen ruin]

https://youtube.com/shorts/G_7PZm_vyt8?feature=shared

[vivid wasp]

Do you know where I can find it?

[forest quarry]

There might be something on #1175430844685484042

[forest quarry]

If you have the right version (and do the onnx conversion of the models) it might work on ARC

[amber zephyr]

Does anyone know how to add an ai voice model with mac

I could really use your help

[forest quarry]

What do you want to use it for, so I can link you to a guide

[amber zephyr]

just for calling people in discord

[feral halo]

Ayo? @amber zephyr level 1 !!!

[forest quarry]

Alrighty, do you have the Apple Silicon ones?

M1, M2, etc

[glad owl]

I'll give it a shot, thanks

[glad owl]

But question, do I need a later version? or an earlier version?

[amber zephyr]

can you not do it without m1 or m2

[forest quarry]

https://rentry.co/VoiceChangerGuide

In case you wanna know how to do the onnx converting stuff

[forest quarry]

yeah no, sadly

[forest quarry]

Any DirectML_CUDA version should work

[glad owl]

Got it, thanks!

[amber zephyr]

but once I download the file it's not letting me open anything

[forest quarry]

you might've downloaded the windows version

.bat is for windows

[amber zephyr]

wait I see something with a green anime girl

I opened it and a terminal came up

[forest quarry]

oh?

so it worked

[amber zephyr]

yeah but dont know what to do from there

[plain warren]

Hello 👋

[forest quarry]

You can always follow this:

https://rentry.co/VoiceChangerGuide

Also, I think you need to open up startHttp.command iirc

So it opens the GUI and you configure everything

[river cairn]

yo

[timid frigate]

Hi :}

[wheat prism]

Why is my voicew changer not working

can someone please help

[amber light]

#🔍│help-w-okada

also specify what exactly is the problem and how it’s not working lol

[wheat prism]

ty

[neon crown]

hi

[half hound]

yo how can I start making ai music for free using stuff from #1175430844685484042

[forest quarry]

https://docs.aihub.wtf/

Has everything you need to know

[clear cargo]

Hi

[forest quarry]

Hello

[coarse vessel]

honestly, did foxjoy reverb HQ work on it (leaving it stereo)? if not, it is mono reverb (centered) and tbf I'd rather reject audio engineers who use such that effect.

[half hound]

I dont really understand how to install RVC

[feral halo]

Ayo? @half hound level 1 !!!

[half hound]

the github kinda confusing

[forest quarry]

You trying to install Mainline?

[half hound]

idk whatever I need to make AI Covers

[forest quarry]

https://docs.aihub.wtf/rvc/local/applio/

Easiest RVC local fork

[cunning tree]

guys is 4060 good for wokada

[half hound]

ik this is not related to music and whatever but does anyone know what is used to make these videos?

https://www.youtube.com/@HistoryColored

like the 4k remastered 60 fps colorized stuff

[elder willow]

Hey guys ive been seeing these models using GPT sovits, Is it better than rvc v2? should i swap to it for tts since i think that tts is probably its advantage?

[feral halo]

Ayo? @elder willow level 1 !!!

[coarse vessel]

you can google some video upscalers

[zenith hamlet]

Maybe try de-echoreverb on mvsep,in uvr models

0.3 aggressiveness is enough for me

[ionic rover]

Gpt sovits is basically just tts

That's the difference

But training with it isn't that simple

Read the guide in full and make sure

[elder willow]

it does have the 0 shot thing, so its just mainly a tts program for models? i use rvc v2 to do most of my training, not really sure if its worth downloading it since i guess its mainly just a tts like you said

[ionic rover]

Well if you want good tts sovits would be a good solution

If you want that, train a gpt sovits model, if not use rvc v2

[elder willow]

oh ok, ill probably download it too, aitrepenuer just released a video on tts/rvc so ill probably try both and compare them. Im guessing sovits is probably simpler.

[ionic rover]

I wouldn't say it is tbh

But just like with rvc, once you managed to do it once its really easy

[coarse vessel]

RVC is more versatile, it can be used to convert TTS output, real human voice recordings, making covers

[ionic rover]

Yeah but if he wants to mainly use it for tts then a gpt sovits model would probably work better

[north dust]

hi me got cool music

[elder willow]

finding good quality tts is hard to find, since they are usually very bland with no inflection. so ill check out gpt sovits

[ionic rover]

Not again...

[terse oak]

🤷

[coarse vessel]

GPT sovits is currently available for english, chinese, japanese. hopefully more in the future.

[celest cargo]

yeah? and instead of reporting it properly you ping the dude in a 380k member server?

you don’t seem the brightest can’t lie

[terse oak]

I’m retarded give me a break

Jeez

[half hound]

thats all I need? even to colorized black and white video?

[terse oak]

@grim shuttle nonce

[ionic rover]

Please stop doing that here

[celest cargo]

the helluva boss and fpe pfp and banner combo is lethal though

ultimate autism

[coarse vessel]

it should work anyways

[ionic rover]

I didn't want to be the one to say it

[celest cargo]

lmaoo

[terse oak]

@grim shuttle

[feral halo]

Ayo? @terse oak level 3 !!!

[hard glade]

hello

[lone nimbus]

is lower or higher Epochs better? or does it not matter.

[indigo anchor]

hey, I'm finally out of the time-out corner

[zenith hamlet]

Gotta find the balance

Both undertrained and overtrained is bad

[storm meadow]

What did bro say

[storm meadow]

Does not matter since it depends on the dataset

[indigo anchor]

a certain word that cannot be spoken

otherwise I'll be in timeout for another week

[crystal crow]

Can somebody help me? I was going to use a voice model that I found on the weights page but when I copied the link from the model to rvc it gave me an error in the code and I don't know how to solve it

[storm meadow]

Apples file transfer tool?

[storm meadow]

Weights reuploads models onto their website instead of linking the HuggingFace

You can either download and reupload to HuggingFace or I think it’s possible to get the link via inspect element

[vestal hamlet]

can anyone recommend a local ai splitter for music

[lone nimbus]

do you have FL studio?

[feral halo]

Ayo? @lone nimbus level 1 !!!

[vestal hamlet]

yeah

[lone nimbus]

free trial or paid

[vestal hamlet]

paid

[lone nimbus]

FL has a built in one

[vestal hamlet]

is it as good as mdx

i had a local program but i forget the name

[feral halo]

Ayo? @vestal hamlet level 1 !!!

[vestal hamlet]

[lone nimbus]

better than any free one tbh

i can show you if you want

[vestal hamlet]

nah ill figure it out

but thank u

[zealous estuary]

scrolling github looking for repos, wanting to use the most up to date local AI software and then finding tons and tons of dead repos from 2023
Sad!

[terse lagoon]

hi

i am beyond lost trying to get so vits svc to work can someone help

[amber light]

sovits svc is old

use rvc

[terse lagoon]

ok how do i use RVC

[amber light]

-guide

[terse lagoon]

install looks similar and im lost

[rare sorrel]

Documentation

📚 All-In-One English documentation

https://docs.aihub.wtf/

❔ English FAQ

https://docs.aihub.wtf/faq

Credits: AI HUB FRANCE

✨ More guides

Full AI Voice Model Training Guide (Local)

Link: YouTube
Credits: Christopher Villanueva

Model training with Mainline RVC

Link: Rentry
credits: Raven (ravencutie21)

AICoverGen Colab Guide

Link: Google Docs
Credits: Eddy (Spanish Helper)

Create a model with RVC disconnected (colab)

Link: Google Docs
Credits: Angetyde

How To Make an AI Cover With Ilaria RVC

Link: Rentry
Credits: 👽 Julia (ailen2091)

[storm meadow]

#✨│ai-help

[terse lagoon]

this guide helps a lot

thank you

i posted a question in there with a new issue

[rose sparrow]

BTW what does the number of Epochs mean?

[elder willow]

hallo

[analog scarab]

1

[devout wagon]

https://docs.aihub.wtf/rvc/resources/epochs--tensorboard/#section You can search up the terms here in this guide

[rose sparrow]

got it

[royal abyss]

Hi

[round karma]

What are the ideal steps to clean an audio file to create an AI cover? What models should I use and in what order?

[elder willow]

Depends

[devout wagon]

Personal opinion; Use https://mvsep.com/en (with) an account and BS Roformer to separate to separate vocals and instrumentals. Apply Deecho-dereverb after that. If you have backing vocals, use UVR-BVE https://docs.aihub.wtf/rvc/resources/vocal-isolation/#i-can-t-remove-some-of-the-backing-vocals

then after that you can denoise. You're practically a model maker, congrats

[elder willow]

hey how do i train a model of my voice on my pc?

[shut plinth]

You can find a guide in https://docs.aihub.wtf/

[elder willow]

tysm

[shut plinth]

Np

[ancient hatch]

Hi

[devout wagon]

You would have to download bs roformer 1297 to test it out. I've only seen simplcup mention it @elder willow

[ancient hatch]

Which room is ai models?

[shut plinth]

#1175430844685484042 I guess

[ancient hatch]

Yep thats it thanks bro

[elder willow]

Thank you

[devout wagon]

found it #📑│making-models message

[elder willow]

Any free source like Suno Ai that creates Music without credits?

[candid moon]

someone should do a dub sukuna voice

[azure dust]

Hi

[torn tulip]

Skibidi toilet AI

[storm meadow]

[loud oak]

https://flashcardfy.lol great ai study tool

[storm meadow]

Bro how much times are you gonna be advertising that 😭

[loud oak]

(check or be square btw)

[storm meadow]

#1159290752195633273

[loud oak]

until i hit a billion dollars revenue

😂

[storm meadow]

Why is it more worth it then a open source version

[storm meadow]

I have seen multiple repositories that use GPT to make flashcards

[loud oak]

most students dont wanna run an open source version themselves lol

[storm meadow]

Most students are broke af lol

[loud oak]

they just want a great ui to generate hundreds of flashcards and get instant ai feedback

man

[storm meadow]

These people are ai generated

[loud oak]

one of them is real and the other 2 are stock photos since i didnt want to ask for my real users for photos of themselves, no real point in that tbh

[storm meadow]

“Loved by users we made up”

[loud oak]

i could put real photos but what difference does that have

😭

there are more than 5k students tho lmao

[storm meadow]

Discord server only has 11 members

[loud oak]

i created iit like less than a week ago

[storm meadow]

Including me, you and lots of bots

[loud oak]

my guy

do u want me to take a screenshot of my database or something

[storm meadow]

Why would your discord be empty then tho

[loud oak]

i created the discord a week ago bruh

i only added that tab a week ago

[storm meadow]

ok let’s check

[loud oak]

did u see? happy?

[storm meadow]

April 29

Your weren’t lying on that one

[loud oak]

im not lying on any of them 😭

[desert viper]

Chatgpt cant even help me... I need your help friends:

ModuleNotFoundError Traceback (most recent call last)
<ipython-input-2-1d7b23896ae4> in <cell line: 4>()
2 #@markdown Link the URL path to the model (Mega, Drive, etc.) and start the code
3
----> 4 from mega import Mega
5 import os
6 import shutil

ModuleNotFoundError: No module named 'mega'

[feral halo]

Ayo? @desert viper level 2 !!!

[loud oak]

send link to mega docs

[loud oak]

real too, in fact i have to add more universities to that since i added those when it had around 2k users

[storm meadow]

So Yale students actually use your product?

[loud oak]

yeah

what do u have against those universities?

thats fine

[devout wagon]

quizzlet was the goat

[loud oak]

mid

i cant even attatch imgs here ffs or else i would send more pics of its capabilities

how do i get img perms lol

[storm meadow]

That other open source stuff has for free

[loud oak]

this has more features than any OSS

[storm meadow]

Explain.

[loud oak]

if i had pic perms i'd be glad to prove to u but i dont

[storm meadow]

Why can you not explain in words?

[loud oak]

no i wanna show u guys since u guys think its some ballsy product or something words you guys will just say oh its a gpt wrapper oh this product has that so i want to show screenshots of it

here actually since ur in the flashcardfy server ill even send the screenshots there

[storm meadow]

K

#1159290752195633273

[covert lake]

promo only in #1159290752195633273

[sand knot]

How do I join the reaper kai?

[covert lake]

what?

[desert viper]

cant even upload the .zip: "upload error - to be continued..."

[sand knot]

mb wrong server

[covert lake]

ask for help in help channels

not in general chat

And be specific

!howtoask

[pine acorn]

How To Troubleshoot

__**GIVE CONTEXT.**__ 📝

• Don't simply mention your issue, like "my rvc is not working".
• Describe the step you are on, what you're trying to do, the RVC you're using, a screenshot, etc.
• The more context, the better.

__**BE POLITE.**__ <:matsuripray:1159685390156967936>

• Don't be desperate. You can ping a Helper, but if they ignore, they aren't available/don't know the answer.
• It's okay if you're frustrated, but don't take it into this server.
• Don't DM without prior consent.

__**BE PRODUCTIVE.**__ 🤝

• Don't ask for every little instruction. Put your own effort & test things by yourself.
• Don't ask to ask.
• Check if your answer is a Google search away/on our guides website.

If you've changed your mind about your message, feel free to delete/edit it.

[uncut knoll]

lol

[elder willow]

Hi

[crimson stratus]

hi

[turbid stream]

Hi

[covert lake]

hi

[dire blade]

hi

[covert lake]

hi

[elder willow]

Hello, how can I make a voice model through any website, with a 10-minute voice recording?

[wicked stratus]

bro

[dusky patrol]

jaguar

https://weights.gg/models/clvubju9r0e8fk6pblkx0ii5y

[modern lantern]

what cloud speech to speech do yall reccomand

[neat thicket]

👋 Introduce Yourself

[modern lantern]

cant believe no one made a rick grimes voice before

[devout wagon]

well you do have to watch at least 1 whole season to compile his voice. His tone changes throughout the series

https://weights.gg/models/clmtyof79002cwspq2e14qvp1

[modern lantern]

i need it to go into survival games run around with a revolver and scream to people that im rick grimes

can u link me up some speech to speech apps? preferably in the google thing cause i have the rx 6650

not sure how well can i run local with it

[devout wagon]

ima let someone else answer that, srry

[covert lake]

check our docs https://docs.aihub.wtf

[modern lantern]

alright, there was some command to get aall the lissts, but i cant remember

[modern lantern]

OH YEAH THE DOCS

[covert lake]

RVC

[modern lantern]

what would be the best for speech to speech

[shrewd ore]

Anyone having trouble with google collab?

[feral halo]

Ayo? @shrewd ore level 1 !!!

[heady oxide]

sa

[shrewd ore]

Lately when I try to use a voice model, it would give me an error for Step 2 and Ive tried different models and try troubleshooting but NOTHINGS working

[heady oxide]

bu orospu evlatları türkçe bilmiyo

türkçe bilen yokmu,

[devout wagon]

This server mostly use w-okada for their voice changer. There's hina colab as an alternative. Might be seconds of delay and you still have to lurk in #🔍│help-w-okada to get help https://github.com/w-okada/voice-changer/blob/master/Hina_Modified_Realtime_Voice_Changer_on_Colab.ipynb

[modern lantern]

alright

thanks

[devout wagon]

np, I only used it once and it's shocking that some of these voices do work

[devout wagon]

and since you still need to download okada packages, you can use the real thing if your GPU can handle it

[covert lake]

Ask in help channels and be specific sned colab link

!howtoask

[pine acorn]

How To Troubleshoot

__**GIVE CONTEXT.**__ 📝

• Don't simply mention your issue, like "my rvc is not working".
• Describe the step you are on, what you're trying to do, the RVC you're using, a screenshot, etc.
• The more context, the better.

__**BE POLITE.**__ <:matsuripray:1159685390156967936>

• Don't be desperate. You can ping a Helper, but if they ignore, they aren't available/don't know the answer.
• It's okay if you're frustrated, but don't take it into this server.
• Don't DM without prior consent.

__**BE PRODUCTIVE.**__ 🤝

• Don't ask for every little instruction. Put your own effort & test things by yourself.
• Don't ask to ask.
• Check if your answer is a Google search away/on our guides website.

If you've changed your mind about your message, feel free to delete/edit it.

[tardy vector]

When a voice channel used to be full back in like 2020 or earlier, I used to call one of the people in the voice channel and then if they accepted (which was pretty frequent), literally just join the voice channel that was full,
it was so funny

[tepid basin]

uh what

[ionic rover]

He been saying that since yesterday

Already warned him twice

[tepid basin]

[ionic rover]

Check his history why is that kid so bored 💀

[tepid basin]

#✨│ai-help message codename looked into that FM graph doing funky stuff. Should be fine

[meager hamlet]

yo any idea of why w a 6700xt im full laggin w more than 6000 of ms ?

[feral maple]

.

[tacit thorn]

wassus

[dusk cradle]

Is there a difference between a small model and the final model that is created at the end of a training session in RVC?

[forest quarry]

actually, no

[daring frigate]

Hey everyone!

[forest quarry]

they could've named the config better, but basically - it saves a model every saving cycle, the final model is just another saving cycle

[forest quarry]

Ello there

Welcome to the server

[daring frigate]

Thanks! I’m really excited to do some AI voice testing!

[dusk cradle]

Thanks!

[forest quarry]

Good to know that! If you have any questions about anything just let us know in #✨│ai-help, #📑│making-models or #🔍│help-w-okada !

[forest quarry]

You're welcome :)

[daring frigate]

Thanks a lot! I’m trying to make a model of Taylor Swift to do some testing with songs and talking, but I have an outdated Mac that apparently won’t update so now I’m trying to get windows to get UVR lol

[forest quarry]

Oh wow

Also had an outdated Mac...Book - 2012 Pro. Was pretty good until a few months ago

mmmm check mel and kl

[daring frigate]

I have a 2017 intel Mac that’s still running MacOS Catalina, I have 860GB remaining of storage but for some reason it won’t update to anything newer it always fails

[feral halo]

Ayo? @daring frigate level 1 !!!

[forest quarry]

they are still going

[forest quarry]

Oh damn

[fallen ruin]

https://cdn.discordapp.com/attachments/1115866252729139221/1238468881580953660/RDT_20240510_1326225356535583130481973.jpg?ex=663f6587&is=663e1407&hm=8fba25b4c6ceef95d79156483d208e326531b927bed8fbf5d75c172ad977220a&

https://cdn.discordapp.com/attachments/1115866252729139221/1238469809440821288/dirt_creeper.png?ex=663f6664&is=663e14e4&hm=9b9976fecd51ad90c340cb8a5cbfe8dc8412e1a96bf918b71f67a4693e408ca7&

[tight cypress]

I just realized Coqui have shut down 5 months ago

[forest quarry]

I do think that you can use cloud stuff only, like I did before

MVSep for vocal separation, Colab for training...

[tight cypress]

damn. busy life forgetting everything.

[forest quarry]

Wait what

[tight cypress]

maker of XTTS and Coqui TTS

[daring frigate]

Yeah, so I was going to try downloading virtual box and run windows on it just so I can run UVL lol then I’ll transfer all the files to my Mac to actually run through the process

[forest quarry]

I'm aware... just didn't know they shut down

[tight cypress]

no wonder their TTS repository have gone with no update for 3 months

[daring frigate]

Is there a good getting started guide? Sorry if this is the wrong channel, I know the UVR part and getting the good clean audio, after that though, I think I might be lost

[tight cypress]

we got a full whole stack of guide written here:
https://docs.aihub.wtf/

including making AI cover, w-okada

[daring frigate]

Oh? Then what should I use?

[tight cypress]

yo you dont even need to train an additional model if you use somebody else?

[daring frigate]

Oh good to know!

[rare sorrel]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• 🆕 UVR5 NO UI for Google Colab, by Eddy Google Colab

How to use?

🤗 Hugginface Spaces

• Applio, by IA Hispano Huggingface Spaces
• Ilaria RVC, by thestingerx Hugginface Spaces
• RVC-HFv2, by r3gm Huggingface Spaces
• AICoverGen, by r3gm Huggingface Spaces
• Advanced RVC Inference, by r3gm Huggingface Spaces
• RVC v2 Huggingface version, by Clebersla Huggingface Spaces

More commands: -audio, -local, -overtrain, -realtime, -rvc, -help

[tight cypress]

which could be find here #1175430844685484042

oh wait

Mac?

[daring frigate]

Yeah Mac, I know controversial 🤣

[tight cypress]

wait wasn't RVC actually worked in Intel Mac.

just that it doesn't have automatic installation

uhhhhhh alright I forgot

[daring frigate]

The colander links will do the same thing UVR will?

Sorry, I’m new to the whole Colman thing. I’m such a noob I feel bad lol

[forest quarry]

_ _

[forest quarry]

it didn't

only on Apple Silicon

[tight cypress]

the thing is that they just want to get their cover done. Not by extra step that will feed another hours of them.

but I guess they could just find if we have the model they wanted.

[daring frigate]

Truthfully, I don’t care how long it takes, I’m curious in the whole process lol

[tight cypress]

yeah that make sense. Intel Mac doesn't have a proper dedicated GPU compared the M-series

[forest quarry]

https://docs.aihub.wtf/ has basically everything you need to know, tbh

I suggest taking a look at it

[solid pawn]

you know who else has basically everything you need to know , tbh you suggest taking a look at it? MY MOM!

[forest quarry]

NO

NOT YOU

[tight cypress]

I would rather look at your dad than your mom tbh

[solid pawn]

you know who else would like to look at your dad than your mom tbh? MY MOM!

[tight cypress]

btw what is this bot?

[daring frigate]

Thanks a lot for the help! Sorry about being so dumb about everything, I will definitely deep dive and read everything!

[solid pawn]

you know who else will definitely deep dive and read everything !? MY MOM!

[forest quarry]

Nah no one is dumb

Everyone just needs time to get used to things

[daring frigate]

I’ve just been very curious, once I get the hang of it, I thought it would be fun to try and do an AI modern version of we are the world with a bunch of different singers

[forest quarry]

Ah man 😭

MJ's death was almost 15 years ago 😭

That would be cool to see

[tight cypress]

who knew man

[daring frigate]

Yeah, so once I master everything that’s my main goal. Is there a place to share stuff here eventually when it’s done?

[feral halo]

Ayo? @daring frigate level 2 !!!

[tight cypress]

uhhhhh I afraid not if you have a content that is copyrighted

[forest quarry]

Well, I wouldn't really recommend sharing the whole song in here (at least directly... I suggest posting on YT and sharing on #1159290752195633273 )

[tight cypress]

we will get bomb into oblivion for the second time if people swarming copyrighted content again.

[daring frigate]

Ok, thanks! Good to know 🙂

I appreciate all the help everyone! 🙂

[forest quarry]

Glad we could help! Any other questions, just go to #✨│ai-help or #📑│making-models

[pale raven]

söyle

[daring frigate]

Thanks, I definitely will!

[plucky ore]

Hey guys I'm new and would like to know how I can use the ai to make my voice for a song I'm writing?

[forest quarry]

https://docs.aihub.wtf/

Has everything you need to know about RVC

Also, welcome to the server

[stiff arrow]

hey everyone

[covert lake]

hi

[grizzled mountain]

Hello, I am looking for a current collab on which I can train the Easy gui model, it has not been working for several months and all voices are public, does anyone have a working collab?

[storm meadow]

-colab

[rare sorrel]

Suggestions for @grizzled mountain

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• 🆕 UVR5 NO UI for Google Colab, by Eddy Google Colab

How to use?

🤗 Hugginface Spaces

• Applio, by IA Hispano Huggingface Spaces
• Ilaria RVC, by thestingerx Hugginface Spaces
• RVC-HFv2, by r3gm Huggingface Spaces
• AICoverGen, by r3gm Huggingface Spaces
• Advanced RVC Inference, by r3gm Huggingface Spaces
• RVC v2 Huggingface version, by Clebersla Huggingface Spaces

More commands: -audio, -local, -overtrain, -realtime, -rvc, -help

[grizzled mountain]

I guess that's not what I'm after, I'm looking for a collab/another website where I can create my own model in .py and .index from my own wav and mp3 files (dataset) and be able to transfer it to the RVC GUI

[feral halo]

Ayo? @grizzled mountain level 1 !!!

[grizzled mountain]

Ok, Tutorial is it somewhere??

Secondly, it doesn't work in the free version and you have to buy pro, and I don't know how to train the model locally

Thanks

[fallen ruin]

the worms are eating my brain

[daring frigate]

I do actually have one more question, if I make a normal talking model is there a way that I can just type something for the model to say and have it say it?

[forest quarry]

That's text to speech

well, you can do the TTS part somewhere else (strongly recommend ElevenLabs) and then convert with RVC, or you could use a built in TTS that your fork has (for example, Applio has built in TTS)

[ivory stirrup]

Hi I'm a newbie, I've been working hard and want to share some things with you guys.

[daring frigate]

Ok! Thanks again!

[forest quarry]

You can share some stuff in #1159290752195633273 , just don't share copyrighted material

🙏

Also, welcome to the server

[forest quarry]

Happy to help :)

[daring frigate]

So I’d just make a talking model like the singing one and then import it to the TTS instead of song cover?

[forest quarry]

You wouldn't really import the model, you'd just use TTS to make it say what you want to be said, and then convert the voice with RVC

[daring frigate]

Oh, ok! That makes more sense lol

Is there a part for that on the doc? Maybe I didn’t see it

[forest quarry]

Don't think there is, but it should be pretty straightforward... just use any type of TTS tool (ElevenLabs for example), find a similar voice, write what you want it to say, and then get the result from that and convert it with RVC

[daring frigate]

Ok! Thanks so much!

[forest quarry]

Np!

[feral halo]

Ayo? @quaint dagger level 1 !!!

[peak field]

Where can i train ai model

[forest quarry]

https://colab.research.google.com/drive/1XIPCP9ken63S7M6b5ui1b36Cs17sP-NS

if you need tutorials, check https://docs.aihub.wtf/

[ionic forge]

So uh

Where can I ask for creating my own ai voice?

[forest quarry]

#✨│ai-help

[fallow vine]

what better rvc or orokada

[subtle wing]

Hi, can anyone tell me how to make an AI cover, but with my own written lyrics? Big thanks

Can I import my own AI voice model into it?

[granite rose]

hi

[feral halo]

Ayo? @subtle wing level 1 !!!

[subtle wing]

I'll give it a try anyway, thanks mate!

[half horizon]

mvsep make me wait 200 person is that normal ?

[forest quarry]

yeah, if u want lower queues you can always just register

[nocturne token]

My voice changer doesnt work, and yes i did try VB cable

[covert lake]

yep, either register or use cloud UVR so that u wont have queue https://docs.aihub.wtf/rvc/resources/vocal-isolation/#cloud-uvr

[covert lake]

#🔍│help-w-okada or #1192011222023950368

[nocturne token]

oh alr

[half horizon]

i use it to clean voice from neoise and echo

[half horizon]

ty

[covert lake]

yea uvr is for cleaning vocals too

[steel scroll]

-colab

[rare sorrel]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• 🆕 UVR5 NO UI for Google Colab, by Eddy Google Colab

How to use?

🤗 Hugginface Spaces

• Applio, by IA Hispano Huggingface Spaces
• Ilaria RVC, by thestingerx Hugginface Spaces
• RVC-HFv2, by r3gm Huggingface Spaces
• AICoverGen, by r3gm Huggingface Spaces
• Advanced RVC Inference, by r3gm Huggingface Spaces
• RVC v2 Huggingface version, by Clebersla Huggingface Spaces

More commands: -audio, -local, -overtrain, -realtime, -rvc, -help

[steel scroll]

where the colabs for converting ai covers?

[half horizon]

is better than mvsep ? or same ?

• what are the option on mvsep u sugget to me to clean a vocal from a podcast and extran clan vocal wich one give me by order

for example i use MVSEP demucs DNR then the uvr denoise deecho

[fallen ruin]

https://cdn.discordapp.com/attachments/1115866252729139221/1238522636158636043/caption-164.jpg?ex=663f9797&is=663e4617&hm=adf8291b9f2e4e6a6c6dd09e1839bb7a4d2b3fee70d13c1d61ec2353c1820a81&

[half horizon]

is there another option u sugget guy

[fallen ruin]

https://tenor.com/view/mamako-oosuki-okaasan-online-do-you-love-your-mom-and-her-two-hit-multi-target-attacks-pretty-smile-gif-16716922

[covert lake]

i honestly havent tried mvsep, but they seem to have both good models, you can find their best models at the bottom of that link i sent btw

[gentle canopy]

hi

[pastel garden]

uhh how do i

download the voice changer...?

[thin tapir]

How to connect it to the game? can it?

[cinder lion]

the voice changer app isnt working anymore its just a white screen now for some reason

[celest peak]

how to create instrumental cover for a song

[leaden elm]

why does it say i have no permissions to search in voice-models?

[winged meteor]

hi, does anyone know how to make ai cover with my voice?

[forest quarry]

https://rentry.org/VoiceChangerGuide

[forest quarry]

delete stored_setting.json and restart

[pastel garden]

I appreciate you

[forest quarry]

you'd have to train a voice model of yourself

https://docs.aihub.wtf for guides

[barren storm]

Hey does anyone know how to clone voice with AI for free?

[forest quarry]

https://docs.aihub.wtf/

[cinder lion]

thanks mate

[forest quarry]

glad i could help :)

[forest quarry]

yw :)

[barren storm]

@forest quarry what do I have to do with the link?

[forest quarry]

there's guides there

read, learn how to make models, then do it

[barren storm]

It works with italian?

[forest quarry]

yeah... works with all languages actually

[barren storm]

Okay man I aprecciate it

[forest quarry]

good luck on the model

[barren storm]

🙏

[feral halo]

Ayo? @barren storm level 1 !!!

[thin tapir]

can anyone answer? plisss

[forest quarry]

you need a virtual audio cable

[thin tapir]

really?? thanks oni chan

[forest quarry]

https://tenor.com/view/low-tier-god-necro-skull-trooper-colt-general-low-tier-god-skull-trooper-gif-26079010

https://rentry.org/VoiceChangerGuide this should help too

[empty zenith]

damn no foghorn leghorn ai voice 😢

[thin tapir]

wait, that's an app, right?

I think I already downloaded it... only it doesn't work in the game

[feral halo]

Ayo? @thin tapir level 1 !!!

[icy ingot]

I got a e-mom

[thin tapir]

I think it's a piece of hardware

[forest quarry]

more like a driver

[forest quarry]

not hardware, software

[forest quarry]

does the game not allow for you to change input/output settings?

cuz normally it should look like this

In W-Okada
Input - Your mic
Output - CABLE Input/Line 1 (if using VAC by Muzychenko)

And on the game
Input - CABLE Output/Line 1
Output - Your headphones

[thin tapir]

There are no settings to change the microphone and speakers in the game

for example... pubg

nothing..

[dusky wharf]

Can i make ai cover just by phone? Or i have to have a computer?

[indigo anchor]

your phone doesn't have cuda cores in it, so no

[dusky wharf]

Thanks for your answer ^^

[indigo anchor]

no problem

[unborn sun]

how can i make my own voices

[forest quarry]

mmm you'd have to change windows' default input to the cable and see if that works

[forest quarry]

yeah, if you use colab or weights.gg

[forest quarry]

https://docs.aihub.wtf

[thin tapir]

I'm doing it

hahaha

perfect brooo....

I did it!!!

[forest quarry]

yayyy

[thin tapir]

nowww... t's prank time hehehe

[forest quarry]

[thin tapir]

sir... do you have any recommendations for seductive mature female voices?

@forest quarry

[forest quarry]

you're gonna have to find that yourself man

[thin tapir]

nooo... sir plis I know you are the handler of all that

[forest quarry]

dawg idk mannnn

[ionic forge]

umm

[feral halo]

Ayo? @ionic forge level 1 !!!

[ionic forge]

How to get

nvm

[urban crane]

https://discord.com/channels/1159260121998827560/1191153471009861744 hehe

[mild jasper]

Google Colab not working?

gives me all this

JSONDecodeError Traceback (most recent call last)
<ipython-input-14-48bd5c703028> in <cell line: 31>()
31 if os.path.exists(config_path):
32 # File exists, proceed with creation of creds and client
---> 33 creds = Credentials.from_service_account_file(config_path, scopes=scope)
34 client = gspread.authorize(creds)
35 else:

5 frames
/usr/lib/python3.10/json/decoder.py in raw_decode(self, s, idx)
353 obj, end = self.scan_once(s, idx)
354 except StopIteration as err:
--> 355 raise JSONDecodeError("Expecting value", s, err.value) from None
356 return obj, end

JSONDecodeError: Expecting value: line 1 column 1 (char 0)

[jade shore]

which one

[mild jasper]

https://colab.research.google.com/drive/1Gj6UTf2gicndUW_tVheVhTXIIYpFTYc7?usp=sharing#scrollTo=OVQoLQJXS7WX

[thin tapir]

oh ma gaad... thanks brou, i love youu

[feral halo]

Ayo? @thin tapir level 2 !!!

[mild jasper]

where else could i make a ai cover song?

got that site from a youtbe video

[covert lake]

outdated colab

check our docs https://docs.aihub.wtf

next time ask in help channels

[mild jasper]

sorry about that thanks!!

[thin tapir]

how to eliminate delays?

it's very annoying

[forest quarry]

there will be delay since it has to process ur voice through ai, AND also run the game at the same time

so its gonna strain ur gpu

[thin tapir]

i seee... thank brou

[elder willow]

https://youtu.be/LTzh8QpfZNk?si=AaVw8nokwEU-Jqay

[hollow haven]

hi

[grizzled bane]

Im tryna get lvl 5

because why not

[feral halo]

Vote Booster: Vote now for a 10% boost. https://arcane.bot/vote

[fallen ruin]

😎👍

https://youtu.be/fz_A-XSdJNY?feature=shared https://youtu.be/SWpej-Jek64?feature=shared

[elder willow]

guys

[split peak]

I NEED NEKO VOICEEE

[half horizon]

im looking avideo explain tensor bord rvc disconected

anyone have link yt

[winter vigil]

Why when infering some audio, the output sings the right notes, voice is good, but at some moments of the song, the model just sings in a unkown langage, like if he's trying to create a new langage xD

[sacred plover]

hi

[cosmic helm]

search mommy

[timid musk]

that’s odd

[marsh flicker]

sa

[tardy vector]

hi!

[indigo anchor]

that typically happens if there's harmonies present in the input audio

that's typically why I recommend not using stem-isolated vocals for input or training data

[rigid lichen]

do you have a working AI that converts image to sfx

[stiff heart]

I found a way to reverse proxy tunnel with btunnel

hosting RVC on my localhost https://20c76b5a-1088-4f7f-a2bd-d15b3acba481.us.btunnel.co.in

works flawlessly

very easy to do as well

https://www.btunnel.in/

[indigo anchor]

you wouldn't need an AI for that. there's programs that can convert images into sound via spectrograms

one I've seen used often is ARSS

https://arss.sourceforge.net/ here's a link

it's quite dated, but it works

[blissful imp]

If someone uses Kits.AI like I do, can anybody share the Angie Yonaga voice model with me?

[rigid lichen]

Is there a version I can use on the website?

[feral halo]

Ayo? @rigid lichen level 1 !!!

[shrewd matrix]

Hello

Does anyone have network knowledge?

[hearty vault]

like 2-3 minutes

[barren storm]

Has anyone already tried the method to clone voice with AI?

[rigid lichen]

@indigo anchor

[hearty vault]

and i don't know why is so slowly

to download the deep fake video

[covert lake]

Yea its normal, also depends on the fps, because it face swaps frame by frame

[indigo anchor]

I don't think so

[covert lake]

As i said, do you use the T4 gou and put only cuda in execution provider

[indigo anchor]

you'd need to call ARSS from the terminal specifically

[hearty vault]

but in the facefusion collab normal

[hearty vault]

when it crashed

is not so slowly to download

like in the facefusion collab normal

you waiting 10 minutes

and one deep fake video it's done

from 2-3 minutes

[barren storm]

Has anyone already tried the method to clone voice with AI?

Pls can someone answer me...

[forest quarry]

yeah why?

[forest quarry]

it's better if you explain your issue yk

[barren storm]

I wanted to know if it was like evelen labs' voice

[forest quarry]

It can clone your voice but not for TTS

Only audio conversion

[barren storm]

What is TTS?

[forest quarry]

Text to speech

[barren storm]

Oh that is what I needed

[forest quarry]

Well, you can still do TTS on ElevenLabs and convert the audio result in RVC

[barren storm]

I need to clone the voices of famous people

Like trump or other celebrities

And then I wanna make them say what I want

Is that possible in a free way?

Eleven labs makes you pay to clone a voice

[indigo anchor]

doesn't elevenlabs support voice cloning natively?

[forest quarry]

Yeah, then you can just use ElevenLabs to make the base audio and convert with RVC

[forest quarry]

yeah but it's paid

[barren storm]

@forest quarry tell me if I understood, I use elevm labs and a random voice, after I created the audio I wanted I'll convert it with RVC that will convert the audio in another one that says the same thing but with the voice of famous people

[feral halo]

Ayo? @barren storm level 2 !!!

[jade shore]

yes

elevenlabs create audio, RVC converts audio to another voice

[barren storm]

@jade shore Can I do it in italian

[forest quarry]

yup

[jade shore]

Yes

[barren storm]

Last question, can it be done only on computer?

[jade shore]

If it has a strong enough NVIDIA GPU

[barren storm]

I have a shitty computer

[jade shore]

how shitty

[barren storm]

It costs 300 euros

And I dont undersant much about computers

[jade shore]

man

[terse violet]

i have a good computer yet it still doesn't work tuff times with a 3060

[jade shore]

I have a decent computer that could handle model training at a snail's pace but nooooo

no AMD GPU support on Windows

[barren storm]

What about smartphone

?

[jade shore]

Locally no

[flat stratus]

hello

[uneven patio]

Hello

[proud sequoia]

One message removed from a suspended account.

One message removed from a suspended account.

One message removed from a suspended account.

One message removed from a suspended account.

[indigo anchor]

hi

[timid musk]

https://media.giphy.com/media/v1.Y2lkPTc5MGI3NjExMXB5bHU3Nms3MjRnYmJlOGNpaW9neDd3MmE1N243bzB4bWRiY3oybCZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9Zw/b9JonIMgG0V8oeWnpb/giphy.gif

[indigo anchor]

oh hey. my time-out period from a week ago finally ended, so that's nice

[timid musk]

amazing

what did you do

[indigo anchor]

I said a word that automod didn't like

[timid musk]

real

can't even say goon

[indigo anchor]

it wasn't even anything crazy, just a file-transfer method apple primarily uses

[storm meadow]

Used for some scams

Shouldn’t be blocked imo

[worldly oasis]

can i take legal action if someone made an ai voice of me like someone from my class

[indigo anchor]

yeah, totally

[worldly oasis]

ok but what do i do do i just go to the police and say "hello this person made an ai voice of me pls help mi"

[indigo anchor]

idk. I'm not exactly a legal expert

hopefully you'll be able to get justice

[worldly oasis]

lets hope if i cant do anything ill do an ai voice of them bc thats fair

[forest quarry]

airdrop?

[storm meadow]

Banned.

[indigo anchor]

not exactly. I didn't say the word outright, but I tried to post a gif containing those words and automod still didn't like it

not to mention, there is no warning that saying it would time you out

[forest quarry]

oh

[storm meadow]

airdrоp moment

[indigo anchor]

is it possible for me to get the roles that I had back, or do I just need to work back up to level 10?

[feral halo]

Ayo? @indigo anchor level 8 !!!

[forest quarry]

this is kind of a double edged sword tho

[forest quarry]

there it is

idk who did it

[indigo anchor]

whoever did it, you're cool

[forest quarry]

fr

[storm meadow]

This is your warning, don’t say airdrоp

[indigo anchor]

thank you. I won't be doing that again

[devout wagon]

I would let it slide imo. Sean Chiplock only acts like an asshole around here for DMCA takedowns because he can prove it caused damages to his career in court

[storm meadow]

It’s spreading misinformation and it’s defamation so probably

[indigo anchor]

this does make me wonder though... since there was no warning that saying that sequence of letters would time you out, we have no idea on what we can or can't say

[storm meadow]

If they state it’s ai and parody there’s not much that can be said

[storm meadow]

Are they supposed to have a banned words channel?

[indigo anchor]

last I checked, that's not a thing

[storm meadow]

Should they tho

[indigo anchor]

no. what I'd recommend is for automod to be like "Hey! Don't say that here!" at least once before instantly timing you out

[storm meadow]

Most scams just a paste a message and leave

[indigo anchor]

I mean, still have the message be deleted. just warn the user before timing them out

if they do it again, then time them out

[elder willow]

what does everybody use to use these ai voices?

[indigo anchor]

depends on what you're looking for. if you're after tts, gpt-sovits seems like a good option. voice conversion? it's most likely gonna be rvc

[devout wagon]

you have a lot of options https://docs.aihub.wtf/essentials/how-to-make-ai-cover/

[indigo anchor]

I don't think they asked about making AI covers

[elder willow]

i asked about voice changer

[indigo anchor]

do you want one that operates in realtime or no?

[devout wagon]

no you didnt lol. Ok #🔍│help-w-okada

[elder willow]

cauthess for u it didnt but mine question had two possible answers

never said how to use the ai voices to make covers

[zenith hamlet]

I'm asking this idk how many times but did gradio got fixed?

[indigo anchor]

gradio was never really broken

google colab just didn't allow it for a while, but locally it works fine

[elder willow]

the thing is

u have to pay for voicemod to actuel use ai voices

soo idk

[indigo anchor]

when did we mention voicemod?

[elder willow]

since i mentioned it

[devout wagon]

All of our voices are free here https://weights.gg/

[elder willow]

ik

[zenith hamlet]

So public links work again now?

[elder willow]

but im asking if any of yall know a voice changer i can use

[zenith hamlet]

There's okada

[elder willow]

whats that

[devout wagon]

or https://github.com/w-okada/voice-changer/blob/master/Hina_Modified_Realtime_Voice_Changer_on_Colab.ipynb

colab is an alternative to okada that's ran on google servers

[elder willow]

idk if low pc can handle that

[forest quarry]

colab is in the cloud

[zenith hamlet]

Hold up is it possible to use voice changers on colab

[forest quarry]

so yeh

[forest quarry]

yeah

[zenith hamlet]

Like connecting mic

[forest quarry]

yup

[zenith hamlet]

Dope

Sad thing im a mobile

[indigo anchor]

there's also the native realtime client in mangio which is what I personally use

[forest quarry]

[indigo anchor]

tbh, I only figured out that existed pretty recently

[ruby onyx]

Is there anyway I can make ai covers on mobile

[forest quarry]

yeah

https://docs.aihub.wtf/

[storm meadow]

HuggingFace Spaces or Google Colab

[forest quarry]

use cloud stuff

ooor weights.gg if u want easy stuff

[ruby onyx]

I’m new to all this

[fickle otter]

does processing of the actual end voice always take ages or did I do something wrong

the*

[elder willow]

guys

what do u use to make the ai covers

bec kits ai wont let me download me own covers

[storm meadow]

RVC

[elder willow]

whats that

[indigo anchor]

retrival voice conversion

[storm meadow]

A open source and free piece of software to make ai covers

[forest quarry]

if its using CPU.... thats normal

[elder willow]

is it popular?

[storm meadow]

Yes, even kits ai uses RVC as it’s base

[fickle otter]

how long does it normally take

20 minutes?

[elder willow]

i cant use it can u sent me an link

[indigo anchor]

it's quite well-known... but not very well-liked since a lot of people use it for identity fraud

[storm meadow]

What’s your GPU?

[elder willow]

idk

[forest quarry]

either use https://weights.gg (the good kits.ai) or the colabs in https://docs.aihub.wtf/

[elder willow]

how i check

[storm meadow]

Task manager

Performance

GPU 1, (0 if 1 isn’t there)

[fickle otter]

also yall are gods with your patience getting tasked the same question over and over again 😭

[indigo anchor]

it's usually one of the programs people point to when they say anything AI is bad

which isn't true

[storm meadow]

Actually no

They only point to ChatGPT since that’s the only one they can think of for some reason

[forest quarry]

you get used to it after 500+ "whats the best female voice model" questions

[indigo anchor]

either that or stable diffusion models that are built on artwork that the creators did not get the rights to use

[storm meadow]

Yep

[fickle otter]

godspeed

[feral halo]

Ayo? @fickle otter level 1 !!!

[fickle otter]

oh okay

@forest quarry how long does it normally take for a CPU to finish processing?

[forest quarry]

depends on how long ur audio is

[fickle otter]

like 3 minutes

[forest quarry]

4m took me 700 seconds

[elder willow]

i made a beerus rick roll

time to beruus one of mine friends.....

[fickle otter]

I guess Ill just wait

thanks anyways

[stuck creek]

anyone have advice on making ess's sound better

[indigo anchor]

there's plenty of de-esser programs available if that's what you mean

[elder willow]

i made something very cursed......

but it sounds soo good

[stuck creek]

how much ess should the dataset have in it?

[indigo anchor]

oh that's what you mean

I don't know

[stuck creek]

** **

[forest quarry]

train more, normally its undertraining artifacts

and keep most ess sounds

unless they're too loud or something

[stuck creek]

I went for 400 epochs with 3min x 4

maybe I shouldn't duplicate data

[forest quarry]

yeah dont duplicate dta

work with what you have, train w lower batch

[stuck creek]

alr

should I go for higher then 400 if im only going to use 3 minutes

[forest quarry]

yeah, go for the maximum amount you can until it overtrains

[stuck creek]

how can I tell when it overtrains

quality just dies?

[forest quarry]

TensorBoard

[ruby onyx]

Dude I don’t get this Google collab lol

I have the audio file and it’s not letting me put it

[stuck creek]

do I need to use something other then Retrieval-based-Voice-Conversion-WebUI to get that connected

[forest quarry]

yeah you need a .bat with TensorBoard

https://docs.aihub.wtf has that in the Epochs & TensorBoard part

[solemn shadow]

do anybody know how to make their own voice?

[storm meadow]

What’s your GPU?

[solemn shadow]

nvidia geforce 3050

[storm meadow]

-local

[rare sorrel]

Suggestions for @solemn shadow

🖥️ Local stuff

• 🍏 Applio, by IA Hispano GitHub
• Mangio-RVC-Fork, by Mangio621 Huggingface
• RVC Studio, by SayanoAI Huggingface
• AICoverGen, by SociallyIneptWeeb GitHub
• Replay, by Replay Team Website
• Original RVC, by the RVC-Project team GitHub
• GPT-SoVITS, by RVC-Boss GitHub

Need some help? 🤔

You can find more info on the #1159513888199540817 channel. If you can't find your answer, feel free to ask for help in #✨│ai-help. Credits to Faze Masta and Antasma for compiling these links.

More commands: -audio, -colabs, -overtrain, -realtime, -rvc, -help

[storm meadow]

Downlaod applio or orginal rvc

Applio has a installer and RVC has a unzip and .bat file

Go to train section to train a model

If you need help ask in #1192011222023950368 #✨│ai-help #📑│making-models

[solemn shadow]

okay where do i get someone to make it for me at

and i pay them cs this hard

[storm meadow]

#1159289738314919936

[signal cairn]

what do you need

[stuck creek]

in addition to that should I leave in the audible gasps for air or cut them out?

[covert lake]

Bro fr left

[forest quarry]

yeah you can leave em, makes breathing less weird

but ofc, if they're noisy, keep em out of your dataset

[elder willow]

we need a mr house

AI

[storm meadow]

Check #1175430844685484042

If it’s not there check weights.gg then if not there either #1159289738314919936

[heady temple]

👋

[slate epoch]

@storm meadow My friend got the following error message when he tried to generate the cover "RuntimeError: Found no Nvidia driver on your system. Please check that you have an Nvidia GPU and installed a driver from http://www.nvidia.com/Donwload/index.aspx"

He said his GPU is a RX580

[storm meadow]

What GPU do they have

[dim minnow]

@orchid slate ey friend me ff en leg uit hoe de fuck je dat doet lmao

[feral halo]

Ayo? @dim minnow level 1 !!!

[slate epoch]

.