🧬│ai-chat

[austere eagle]

it can be used

as long as it's good

[polar flax]

even 3060 is enough for valorant peeps

[austere eagle]

my budget is higher than 3060 i guess

I want to invest in a decent gpu

[polar flax]

4060 Ti 16 GB but worse perf per price than 3080

[austere eagle]

hmm

[earnest dragon]

Yeah the best value gpu is the 3080 20gb or 4090m desktop but you need to know someone in china to get one

[zealous zephyr]

hey guys! I actually needed some help, is there any way from which I can rephrase words from a movie scene by AI use?

[earnest dragon]

yes but it's more complicated to setup

[zealous zephyr]

can you please explain

[earnest dragon]

because you need rvc then a vocal splitter and then a lipsync ai and most lipsync ais are not that good

[zealous zephyr]

ok I see

[bronze robin]

ayy

[zealous zephyr]

I will look after this thanks!

[zealous zephyr]

actually I don't need lipsync AI

[austere eagle]

rtx 3080 at my location only cost 351

[earnest dragon]

a normal 3080 is not good for the price

[zealous zephyr]

and I have appolio RVC but I don't how I can get the emotion of the dialouges

[bronze robin]

ive been working on music for some time but i still dont know how to make beat or shi or even how do i remove backround noise from my mic could anyone help me wia tha??

[earnest dragon]

Yes you do if you want to change what they are saying if you are using the scene still because if not the speech will not match at all

[zealous zephyr]

ohh ok

do you know a website from which I can do that or anything that can help me?

[zealous zephyr]

hmm alr

[earnest dragon]

there is no website

[zealous zephyr]

.. ok

[bronze robin]

ive been trying make yeat type shi

[feral halo]

Ayo? @bronze robin level 1 !!!

[zealous zephyr]

so is there anything else you can help me with this lipsync AI thing?

[earnest dragon]

a okay quality lipsync software is facefusion but the good ones are paid or are hard to setup

[zealous zephyr]

hmm alr man thanks

[austere eagle]

is radeon card that bad for rvc tho

[earnest dragon]

Why not a more consistent video gen

[earnest dragon]

it's fine just some stuff like training and some other ais only work on nvidia but for training there is a janky fix for amd and there is w-okada and stuff that works on amd

[keen smelt]

Sup!

[earnest dragon]

#1159290752195633273 only

[austere eagle]

Technically nvidia is better for w-okada right?

[zealous zephyr]

ayoo is this parrot AI really worth it?

[earnest dragon]

yes

[earnest dragon]

Trash compared to rvc which is free

[covert lake]

#1159290752195633273

i replied to u in #✨│ai-help

[zealous zephyr]

hmmm alr

[barren compass]

hello, can someone help me with a bug in Voice Changer?

please

[potent cargo]

Hi

[feral halo]

Ayo? @potent cargo level 1 !!!

[lethal flax]

would there be any ai to remove this screenbug?

[earnest dragon]

is there not a better quality version

[lethal flax]

no

it was recorded to VHS and then broadcast

it's a Jack Stauber short

[ebon coyote]

why the graph doesn't appear on tensorboard?

[icy pendant]

ask in #🔍│help-w-okada

[icy pendant]

Make sure refreshing is enabled in settings. Sometimes its bugged, I experienced this with Applio before. I didnt have a fix except restarting the training (you can resume if you have already trained some), it will show everything after

[chilly lake]

make sure you start tensorboard only after sync graph done its run

or restart it

[lethal flax]

.

[inner badger]

@icy pendant @icy pendant

[icy pendant]

cool pings bro

[floral finch]

can i ask a question? where can i check what sample rate the w-okada software uses

also is it adjustable?

[icy pendant]

I think its limited to the voice models sample rate by default

If you use server mode, there is a function to adjust sample rate. Also, you can adjust the virtual cables sample rate in your sound device settings

If any of that helps

[lethal flax]

chat this server sux ngl

[earnest dragon]

It's good

[lethal flax]

NO IT ISN'T

I'm sick of users not answering simple Yes or No questions, and instead insisting that they're right.

like, the world doesn't revolve around you

just because you think your idea is good doesn't mean others will agree

[earnest dragon]

I answered I can but only will do it one way

[lethal flax]

why can't you just do it how I want and instead have to have it like you want it

last time I checked you haven't been looking for an official version without the watermark for months (as I have been)

I'm not trying to be rude, I just think you're purposefully being difficult

[earnest dragon]

No I was just looking for the best quality version which is the watermark one

[lethal flax]

and I'm looking for a version without the watermark

[earnest dragon]

there isn't

[lethal flax]

is there a way with artificial inteligence to remove the watermark?

(as I asked hours ago)

[earnest dragon]

yes there is but I would only replace it with a older logo not a different one

[lethal flax]

https://tenor.com/view/shaking-crazy-feral-halloween-animatronic-gif-7415356108918299170

literally me if you even care

[lethal flax]

how would you go about doing it? (asking for a friend)

[earnest dragon]

they are all paid

[lethal flax]

oh okay

thanks for answering my question

sorry for being so rude

I will leave now

[maiden jacinth]

hi

[green stag]

guh

[strange hill]

There thick men's voices?

[earnest dragon]

what does that mean and that sounds weird

[strange hill]

I can send you an example to the DM

[earnest dragon]

yes

[inner brook]

Hi

[ionic pumice]

crazy pfp

[earnest dragon]

his is just a weeb pfp

[ionic pumice]

that's ...

not

[earnest dragon]

yes it is if not what is it

[timid musk]

Hello guys how do you ai cover

[tepid basin]

-colab

[rare sorrel]

Suggestions for @timid musk

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], by Eddy, Hina and Gdr Google Colab
• UVR5 NO UI, by Eddy Google Colab
• UVR5 UI, by Eddy Google Colab
• Modified W-Okada's Voice Changer, Google Colab
• 🆕 FaceFusion UI, by Nick088 Google Colab
• 🆕 FaceFusion NO UI, by Nick088 Google Colab
• 🆕 EasyGUI, by Rejekts Google Colab

ℹ️ Note

While the Colab free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[worldly stone]

Guy where is the ai picture generator room

[earnest dragon]

there isn't

[worldly stone]

Ok thank

[wary vapor]

Which one of these would be good for just doing a voice over?

[carmine ingot]

hello my name is dalton from indonesia iam ai enthusiast

[timid musk]

me too

[west escarp]

does anyone know what beatrice v2 voice number would be best to change for a jigsaw voice?

[earnest dragon]

beatricv2 you need to train the model for the specific voice

[static hazel]

hi, do you know how I can make ai covers of the weeknd for free?

[earnest dragon]

why the weekend

[static hazel]

posnomas

[tired bane]

[earnest dragon]

what

[iron patrol]

You need a rvc link to a huggingface page or Google colab, there's other links or software out there so pick whichever you like, and a ai model which you use the link to the model to put on top of a vocal track

[gloomy bane]

Hello

[solar compass]

whats a good way to do text to voice using your chosen model?

[wary vapor]

Applio

[green briar]

sometimes, i dream about cheese

[solar compass]

thanks!

[stone monolith]

Can someone please help me on how to use the RVC model I downloaded in text to speech form rather than real time voice conversion from okada software?
Thank you in advance.

[timid musk]

what

[stone monolith]

Suppose I downloaded a voice model.
Now I want that model to speak the texts I write instead of talking in real time

[frail pulsar]

Hey, want 1$

[ionic pumice]

what

[amber light]

https://tenor.com/view/clean-up-the-wallet-broke-no-money-gif-13587982

[tacit falcon]

sure i'd love one single dollar

[earnest dragon]

I would like 5 dollars

[polar flax]

https://tenor.com/view/broke-wallet-i-am-broke-no-money-no-cash-gif-24120042

[dawn temple]

This is so cute 🥺

[feral halo]

Ayo? @dawn temple level 1 !!!

[tepid mist]

rip discord in turkey😓
(they blocked the acces to discord in Türkiye, Turkey🥰)

[ionic pumice]

hello mod that I do not know of

[silk anvil]

hey im new to all of this ai stuff, what does the ov2 mean? is it like a setting i can change in the voice changer or is it something i dont have to worry about?

[covert lake]

its not any settings, its a pre train, used only for training

[tulip nest]

is there a guide for de-essing (for free)?

[silk anvil]

okay, thank you!

[tepid estuary]

I don't know if this is the right place to ask but, I want to make animated videos and want to use ai voices for the characters, record my own voice and then change into ai ones, but I was wondering if that could get me in trouble? it does not have to be anyone specific just different from my own voice

[feral halo]

Ayo? @tepid estuary level 1 !!!

[elder willow]

colab free plan 1 hours gpu In my experience

[covert lake]

the gpu time depends alot

[elder willow]

Thanks google for reducing it

[covert lake]

use kaggle

(still owned by google)

but gives 30 hours of better gpu weekly

[dapper drum]

Shimmy shimmy ya

[bitter moss]

yurr

[elder willow]

anybody got like- a list of ai tags to use for an ai cover im posting

[elder willow]

good idea

-kaggle

[rare sorrel]

📘 Kaggle Notebooks

• Applio Notebook, by Vidal Kaggle
• Applio Notebook, by Shirou Kaggle
• Music Source Separation, by Shirou Kaggle
• UVR5 NO UI, by Eddy Kaggle
• RVC Mainline, by Hina Kaggle
• Original W-Okada's Voice Changer, Kaggle
• Modified W-Okada's Voice Changer, Kaggle
• 🆕 UVR5 UI, by Eddy, ArisDev & Nick088 Kaggle
• 🆕 RVC AI Cover Maker UI, by Shirou & ArisDev Kaggle
• 📖 How to use RVC Mainline Kaggle by Cauthess

Note: Kaggle limits GPU usage to 30 hours per week.

[elder willow]

wait i can use ui in free ?

[covert lake]

technically not allowed just like colab, but still bypassed

• Mainline (UI)
• Applio by Vidal (UI)

those are the ones with guides

[elder willow]

oh lol so the i can use google drive on it ?

[covert lake]

cant do that

on kaggle u cant connect with google drive

But anyways its highly improbable u lose any work

30 hours weekly its more than enough

[elder willow]

yah its more fixble

[covert lake]

yup

[foggy cypress]

hi

[past smelt]

hey guys,
i have a rtx 2060 6gb of vram,
and i wanted to try flux models,
i downloaded flux [dev] model,
but the generation takes well over 40-60secs/it,
is there anyway to speed things up?
or is there any faster version of flux?

[tawdry ore]

i have rtx 3050 and ryzen 7
can someone come to vc and say is everything ok with my voice or no

[deft garnet]

i can’t react to anything

[tribal cipher]

Hi!

[floral finch]

well where can i see that?

it kind of fixed itself after switching to a different virtual cable, but might still be worth looking

dont use VB, its terrible

atleast for this

use VAC

nvm i found it

client info

tysm

so under client vs server info it has different sampling rates

[feral halo]

Ayo? @floral finch level 1 !!!

[floral finch]

if i open the client info for a model it says the sample rate is 48k, but finding that same model under server info says the sample rate is 40k, i even saw 32k in some

that cant be right, like are those even standard?

[dawn temple]

Hello there, How's it going? :D

[tawdry ore]

vijay u indian bro?

i am too

[dawn temple]

Yes

[tawdry ore]

bro can u help me up

[dawn temple]

With?

[tawdry ore]

okada

my voice somehow cracks

i followed the guide

[dawn temple]

Ahh, I would suggest you to talk to helpers here, I am not the best person to get help related to okada from 😅

[tawdry ore]

ohk

[elder willow]

good evening ai hub gen

[supple dew]

this is for up coming projects https://docs.google.com/forms/d/e/1FAIpQLSdGzt2lun5dr2NyUkEW4Q7ApS_ZJBQb-7vuRotUQMIRC0qUuQ/viewform?usp=sf_link

[brazen thicket]

then how did u became a mod o-o

[covert lake]

in this server its mostly about 2 programs, wokada & rvc

some helpers may know more of one than another

some helpers may know more of cloud than local or the other way around

for example i cant do local bc integrated graphics and never used wokada

[chilly lake]

comfyui, flux dev, ~1100x900 image takes under 5 minutes to generate in 20 steps on 6700xt

it is flux dev baby

you get quality

[past smelt]

What else can i use?

anything faster?

[chilly lake]

6GB VRAM is not a good thing, if you got less than 32GB ram just forget it and move on

[past smelt]

Nah, tried that already

[past smelt]

So let flux go?🥲

Vram is the ram of the graphics card

Ram is the sticks that you can upgrade

Meant Noobie, not you

[chilly lake]

6700xt has 12GB, not enough for FLUX without a model offload, but it works... 6GB on 20xx gonna suck a big d

with 6GB best you can do is SD1.5

XL gonna be a struggle even with 32GB RAM

need to use fixed vae, tiling, etc

[covert lake]

Lucky

are u german

anyways watch it

[compact coral]

bist du crazy

aint no way

[elder willow]

suoiresnu fn

[floral finch]

the client and server info say different things about the sample rate of a wokada model

idk which one im supposed to listen to

client says 48k server says 44.1k

is it like the client is input server is output? does it do resampling live? bc my microphone is fixed at 48k (built in laptop mic)

[compact coral]

do the 48k one

[feral halo]

Ayo? @compact coral level 4 !!!

[compact coral]

shut up

[floral finch]

so its input and output at 48k?

i should set vac to 48k then

[compact coral]

try

[floral finch]

its like it works fine natively but when i try to use it with discord its really choppy probably sampling mismatch

[safe dove]

Anyone know why I get errors when i click start training when im on local with a 3060TI Nevdia ?

[compact coral]

turn of noise surpression

and the other weird shi

[compact coral]

yes

[safe dove]

Can u help me rq

If possible

I’ll show u the error whenever I get it

[compact coral]

sure

[safe dove]

Bet ima dm u thank u 🙏🏼

[floral finch]

"echo cancellation"

i already had noise supression off

yea discord is being an asshole, i figured bc like a different mic check was fine

tyty

[compact coral]

np

[mortal bough]

anyone know where i can get audio files to convert into voice model

[covert lake]

so you mean the dataset to train the voice model?

youtube, or whever u can get good samples

[sand vigil]

I can still acces tho

[feral halo]

Ayo? @sand vigil level 1 !!!

[tepid mist]

I need use vpn to acces

[compact coral]

why do yall even want ai voices

[mortal bough]

so id have to seperate the voie from any music or effeccts?

[covert lake]

yes ofc

[mortal bough]

was hoping there was a premade datset library somewhere

[compact coral]

yall the biggest robots, starting to think yall ai as well

[covert lake]

check https://docs.ai-hub.wtf/rvc/resources/datasets/ And https://docs.ai-hub.wtf/rvc/resources/vocal-isolation/

[mortal bough]

im not convinced im not ai

[covert lake]

no, people make their own datasets to train rvc models

[mortal bough]

yeah i know, but i thought just maybe i could get those files somewhere

the program im using doesnt take models

[covert lake]

what are u using?

[mortal bough]

makebestmusic

[covert lake]

That seems like an ai music generator

[mortal bough]

yeah

[covert lake]

we are talking about RVC for ai voices

[mortal bough]

i know

[covert lake]

what does that program have to do with rvc then

[mortal bough]

i cant upload models but i can upload audio to make a model aka datasets. my original question was if theres a place to find datasets but you answered my question. Thank you!

[covert lake]

you can find the voice samples somewhere yourself, then use the tools to clean the dataset

we domt allow sharing datasets for copyright issues

[mortal bough]

Im so used to AI spoiling me and doing all the work already

[mortal bough]

understood

[feral halo]

Ayo? @mortal bough level 3 !!!

[covert lake]

atleast for rvc models not really, u just can find like a samples of the voice such from yt or pirate sites and clean it to make a dataset

if the site u are talking about is used for making voices

its better to use rvc

[mortal bough]

a friend told me to try twitch archives i think

[covert lake]

rvc is free and open source

[mortal bough]

it makes music, but currently id have to use their models or make my own. i cant simply upload a model for some reason which doesnt make sense to me, but what do i know

[covert lake]

Models for voices? Its better to use rvc

rvc is not like thos ai sites paywalled

its open source

[mortal bough]

yeah but im using it to create the music

[covert lake]

its speech to speech

for music generation then yea

[mortal bough]

could i just use their models and then do voice conversion on rvc?

or would there be distortion?

[covert lake]

Are their models about voices?

like is ur goal to make a voice model sing

[mortal bough]

yes

[covert lake]

with rvc u can just sing the song yourself, then use the rvc model over your voice

its better to use rvc tbh, its all free and open source

[mortal bough]

i cant even whistle a tune -___-

[covert lake]

rip then

we dont really use those sites here

[mortal bough]

but i write lyrics and this site is pretty cool

[covert lake]

we mostly use open source ais instead of closed sourced like those sites

[mortal bough]

i know

i've used okada

[covert lake]

ohh lol

ig for datasets u just have to rely on urself

unfortunately there isnt a way to find already made ones for rvc

[mortal bough]

seems like it. thanks for your patience. where do you usually begin to look? Youtube?

[covert lake]

personally youtube, but you can also look into piracy sites yk

Also its better to get a .wav which is lossless

[mortal bough]

thanks for that protip!

[covert lake]

your welcome

[elder willow]

-colab

[rare sorrel]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], by Eddy, Hina and Gdr Google Colab
• UVR5 NO UI, by Eddy Google Colab
• UVR5 UI, by Eddy Google Colab
• Modified W-Okada's Voice Changer, Google Colab
• 🆕 FaceFusion UI, by Nick088 Google Colab
• 🆕 FaceFusion NO UI, by Nick088 Google Colab
• 🆕 EasyGUI, by Rejekts Google Colab

ℹ️ Note

While the Colab free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[elder willow]

-kaggle

[rare sorrel]

📘 Kaggle Notebooks

• Applio Notebook, by Vidal Kaggle
• Applio Notebook, by Shirou Kaggle
• Music Source Separation, by Shirou Kaggle
• UVR5 NO UI, by Eddy Kaggle
• RVC Mainline, by Hina Kaggle
• Original W-Okada's Voice Changer, Kaggle
• Modified W-Okada's Voice Changer, Kaggle
• 🆕 UVR5 UI, by Eddy, ArisDev & Nick088 Kaggle
• 🆕 RVC AI Cover Maker UI, by Shirou & ArisDev Kaggle
• 📖 How to use RVC Mainline Kaggle by Cauthess

Note: Kaggle limits GPU usage to 30 hours per week.

[bleak sand]

e

[covert lake]

Rip then

[ionic pumice]

what

[minor blade]

what

[gilded eagle]

is the ai voice changer server down?

https://gyazo.com/41c7e5aab885241a626bbe71294af7c6
https://gyazo.com/5e66c6fdc4c7f568ad9ea52e7c652ca2
was workign fine then closed and now doing this

[tacit oxide]

Why cant i see voice models?

[hardy fox]

hi

[ionic pumice]

what

[icy pendant]

Not down
Delete stored_settings.json and try again

[covert lake]

#1175430844685484042

U have the perms to see not post them

[tacit oxide]

Nvm, i just had alot free space for some reason

[covert lake]

If they don't show, clear any researches in the forum search bar or refresh the site, might be a bad wifi moment too

[tacit oxide]

I had to scroll down like for 2 seconds

[covert lake]

Oh alr

[quaint jewel]

sup

[zealous zephyr]

hey nick man can you help me with one thing please?

[drifting gate]

hello

[zealous zephyr]

@covert lake I want to send a youtube link can I send it to your dm please?

[potent oracle]

@errant plaza ew

[covert lake]

if you are following a yt tut

its bad

dont follow it

most of thise are outdated

[zealous zephyr]

no no it's not a yt tut

[covert lake]

if u are tryna promo your video use #1159290752195633273 instead

[zealous zephyr]

nooo I am not self promo

I just want to make vids like that

[covert lake]

?

u can send the video here

[zealous zephyr]

I don't have perms

[covert lake]

its a yt video, u can send the link

[zealous zephyr]

oh ok wait

https://www.youtube.com/shorts/fZtJ9YlSXgE

[feral halo]

Ayo? @zealous zephyr level 4 !!!

[zealous zephyr]

@covert lake I want to make videos like this

how can I please help

[covert lake]

either TTS or u say the word and convert it with rvc

[zealous zephyr]

oh

[covert lake]

There are different Text To Speech (TTS) AIs:

GPT So Vits: RVC isn't as good as GPT So Vits for tts, but gpt so vits (few shot tts, which means needs just a lil training for models) can't use rvc models (and viceversa), and its only limited to: english, chinese & japanese, if you wanna check gpt so vits instead, read https://docs.aihub.wtf/tts/gpt-sovits/

Freemium 11labs: A easy way to do TTS is https://elevenlabs.io/, you can't use RVC model on this but its a mostly premium easy way for good quality TTS

FishSpeech: FishSpeech is a 0 shot (no explicit training needed) TTS, if you got a good pc you can use it locally else use their site

With RVC Models:

RVC is natively for Speech To Speech, but forks such as ilaria rvc mainline & applio have built in tts (using Microsoft Edge TTS to make a generated tts audio, which i suggest you to choose a tts model that is the same gender and language of the rvc model you wanna use, and then convert it with rvc)

If you wanna do tts locally with RVC Voice Models (if you got a good pc):

• You can get Applio in our docs
• While Ilaria RVC Mainline here (no guide as of right now)

If you don't got a good pc you can do tts with RVC Voice Models on cloud:

• Ilaria RVC Zero (Running on A100 GPU, free fasted rvc on cloud) and the guide

• Use Applio UI Colab (with google colab T4 free daily limit gpu)

• if you don't wanna use edge tts, you could try another tts ai from our tts index and use the output as an input in rvc

[zealous zephyr]

how about I use parrot AI for this? cuz the creator himself told that he uses parrot AI

[primal needle]

https://tenor.com/view/jajka-nxcp-gif-2871455001541256807

[covert lake]

its paywalled

just use rvc with tts or an actual tts program just like gpt so vits

[quartz flame]

[elder willow]

!colab

/colab

[vital oracle]

https://www.tiktok.com/@ai_fusionmaster/video/7417673102479396139

how do u make ai images like that

like for motion

[dusky tendon]

can anyone help me in help forum

[chilly lake]

img2vid, plenty of models

or txt2vid

[zealous zephyr]

is there a way by which I can just change the wording of the dialouge and don't need to clone the whole voice then apply it by rvc like that takes a lotta time?

[covert lake]

It's -colab

[covert lake]

I mean u could just convert the specific parts u need and cut it inside the dialogue ig

[leaden haven]

My name is khairul

[ionic pumice]

hello khairul

[leaden haven]

Hi

[elder willow]

Reading AI research papers while hunkering down for Milton😎

Anyone here checked out "Situational awareness" by Leopold Aschenbrenner?

[hollow forum]

hello everyone

[leaden haven]

Hi

[night lake]

https://tenor.com/view/zenless-zone-zero-zzz-belle-alleyway-gif-3239349112686033801

third

[crimson silo]

Hi

[elder willow]

Definitely not the chat I was looking for good lord son

[safe mantle]

hey, what's the best ai to generate images in your opinion?

[weak kraken]

This link is not working anymore Can you give me another link please

[agile wren]

#📰│dev-updates message

[weak kraken]

When I click on On the cloud It doesnt work

[agile wren]

Not sure what u mean, but I think I'd just paste the link here
https://docs.ai-hub.wtf/

[weak kraken]

I cant sent photo here Go to the site > how to make Ai cover > Convert the vocals Theres a link On the cloud I used to use that to convert vocals and it's not working

[weak kraken]

I come back after a while and the whole thing changes Lmao

[agile wren]

Hyperlinks on the page are broken, u have to manually add '-' in the link

like, "https://docs.aihub.wtf/" -> "https://docs.ai-hub.wtf/"

[weak matrix]

👋

i am sigma ligma

https://tenor.com/view/cat-meme-gif-24404127

NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

😭

[agile wren]

https://tenor.com/view/epic-embed-fail-ryan-gosling-cereal-embed-failure-laugh-at-this-user-gif-20627924

wtf lol

[weak matrix]

let me cry let me cry

um

[feral halo]

Ayo? @weak matrix level 1 !!!

[woven ether]

i need help about rvc can you help me any one

[feral halo]

Ayo? @woven ether level 1 !!!

[weak matrix]

WHY DID I TURN INTO ANDRIOD

[weak matrix]

im sigma

[woven ether]

Will you help me

[weak matrix]

i dont know anything about ai

[woven ether]

ok

[weak matrix]

🤗

[elder lark]

Hello community! Could a mature ask how would one go about making there own voice model

[tawny sage]

Yo

[fresh coyote]

how would i make a ai voice?

[weak kraken]

Why the download so slowwwww Im using apollo

[feral halo]

Ayo? @weak kraken level 2 !!!

[weak kraken]

Nah Y'all dont have a model for Ty dolla sign?

[weak matrix]

ELLO

[tepid basin]

Bruh

If there isnt a good Ty model then I give up

https://www.weights.gg/

Search ty dolla, there's a few that pop up

[safe sparrow]

hello

[feral halo]

Ayo? @safe sparrow level 1 !!!

[safe sparrow]

sup

[covert lake]

#1159290752195633273

[covert lake]

whats ur pc gpu?

[covert lake]

You can search rvc ai voice model at:

• #1175430844685484042
• #🔍│find-models
• https://weights.gg/ (login required)
• https://huggingface.co/models (but watch out cus in hugging face there arent only rvc ai voice models)
• https://applio.org/models
• https://voice-models.com/
• https://thevoicemodels.com/ (for Turkish Models, login required with discord and level 2 on their server)

if there isnt one, you can:

• #1159289738314919936
• #1191429836321849435
• make it yourself with our docs guides https://docs.aihub.wtf/essentials/how-to-make-voice-models/

[covert lake]

whats ur pc gpu?

[zealous zephyr]

it will take hella lot of time 😢

[covert lake]

well depends on how much long is the dialogue

[fathom flower]

Hi everyone, is there a voice changer for macbook?

[stark tangle]

Nvm

[covert lake]

yes, wokada that uses RVC models, especially the fork that is optimized: https://rentry.co/ForkVoiceChangerGuide

[fathom flower]

thx))

[covert lake]

yw

[weak matrix]

waiwue

[zealous zephyr]

gotcha

[ivory relic]

@elder willow hi

i have a question

[drowsy peak]

sup

xd\

gotta buy

they don't ship to russia because of some bald guy

[zealous zephyr]

wtf is this real

[covert lake]

?

no

[grim loom]

New here🥺

[past kayak]

Hi guys I got a question. How can I work my way around really strong AI detectors such as Turnitin? I know about humanisers but some of them still can be detected. If anyone can advise, it’ll be much appreciated 🙏

[covert lake]

ai detectors aren't trustable

[past kayak]

I agree, so is there a way around it ?

Because some schools heavily rely on them to be 100% true unfortunately 😔

[covert lake]

telling chatgpt to be as human as possible, its kinda like tries

[past kayak]

I see, thanks bro

[zenith nova]

guys should i post the jameskii model?

[near root]

is there any process to making the most realistic voice model possible that isn't already included in the ai-hub docs?

[harsh dome]

i installed arch btw :3

first time arch user 😄

[undone nest]

Hi everyone, new and happy to be here

[frank sinew]

Hello, can I share videos of youtube here about ia generated music?

[primal karma]

Hello 🙂

[compact hatch]

the hugging face rvc isn`t working it said "gpu task aborted"

[stray cobalt]

Any good french girl model ?

[covert lake]

check https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit#heading=h.mfxmgfqiaevu

[covert lake]

#1159290752195633273

[dim spire]

Hey I wanna use an AI for art, but it wanna feed it my own stuff from scratch, is there a program that allows me to do that?

[frank sinew]

AI for art? Is it the microsoft designer?

[dim spire]

i honestly don't know

[frank sinew]

Is it ai picture generation? Music? What kind of art?

[feral halo]

Ayo? @frank sinew level 1 !!!

[dim spire]

picture

[feral halo]

Ayo? @dim spire level 1 !!!

[frank sinew]

Picture? Microsoft design is interesting, you prompt something and it will draw new picture using ai. Is it what you looking for?

@covert lake hello

What is this promo?

[covert lake]

its to promote your videos

dont promote yourself in any other channels

[frank sinew]

Ok

[frank sinew]

I can't send messaged there. The send button doesn't appear. Why?

[dim spire]

well it uses a database of images in order to do so, i want to make my own database

[frank sinew]

You want to create a data base of images and then use some tool that will use those pictures?

[feral halo]

Ayo? @frank sinew level 2 !!!

[dim spire]

correct

[gilded eagle]

where do i fidn that

[icy pendant]

in your MMVCServerSIO folder

[gilded eagle]

ty!

[feral halo]

Ayo? @gilded eagle level 1 !!!

[abstract latch]

who was deep voice model

[topaz vortex]

Hello, i was trying to run start_http in the MMVCserverSIO but it just showing a loop of "VCClient Wait web server" message loop

Is anyone can help ?

I was trying to open RVC Okada by every available tutorials but stuck in the same spot

[grim loom]

Hii

[feral halo]

Ayo? @grim loom level 1 !!!

[grim loom]

I’m new here

[grim loom]

🤤

[polar flax]

train your own SD LoRAs or find ones as your want

[polar flax]

u mean img2img with your rough sketch?

[dim spire]

thank you ill look at those

[wicked skiff]

hi!

[vestal lake]

@ripe sluice UNBLOCK ME

PLS UNBLOCK ME @ripe sluice

PLS UNBLOCK ME @ripe sluice i BEG.

[silk fog]

Why is weights. GG not working ? It logged me off and locked my account and I can’t get back on!

What is going on with Weights. GG?!

[foggy token]

i was sent to time out in a different server because i said some guy was wrong

[cunning pine]

Anyone here intrigued by firefighting? If so I have a couple photos from my call from earlier. just DM me.

[earnest dragon]

It's working perfectly fine

[naive dust]

damn burh

i never knew that RVC v2 was 3.1gb

[agile wren]

Pretty standard for an AI package (w/ full environment packed)

[earnest dragon]

Yeah rvc is not that big compared to like deepfake software

[silk fog]

It made me locked out of my account

I didn’t do anything wrong

Look!

It says try signing in with a different account

[feral halo]

Ayo? @silk fog level 2 !!!

[silk fog]

It seriously needs to get fixed !

https://feedback.weights.gg/p/why-did-weightsgg-just-lock-me-out-of-my-account

[fervent bone]

Wow X-Minus does a great job!

[silk fog]

It’s working again! For now at least

[versed frost]

Was a temporary server problem, try logging back in now.

[silk fog]

Oh thank goodness 😮‍💨

[earnest dragon]

bs reformer is better

[fervent bone]

Nah

[earnest dragon]

why

[fervent bone]

My opinion

[chilly lake]

since they require login I refuse

[fervent bone]

For X-Minus?

[earnest dragon]

tried it and I just heard the instrumental in part of my vocals

[earnest dragon]

bs reformer is better then mdx

[lean drift]

yo is there a way to sing like live with the ai?

[earnest dragon]

yes w-okada and if your trash at singing then route your autotune voice into w-okada

[lean drift]

is it free?

[earnest dragon]

autotune no w-okada yes

[wintry gulch]

hi, i need some help understading current AI chatbots for natural conversation.

which ones would usay are the best ?

any recommendations ? mobile/ PC any.

[covert lake]

What? U can't upload files but can share the yt link in that channel

[glossy lintel]

Can someone tell me how to generate pics like a car made of diamond Or adorned in diamonds. I want to have full control over generating pics..

[covert lake]

You can use Dalle3 for free with @elder willow in #🤖│bots or ask for help in #🔍│help-ai-art

[covert lake]

Its mostly about rvc and wokada here, but for Chatbots u could try seeing for llama 3.1 70b or 8b depending on how powerful is ur pc

[elder willow]

Hola

alguien aquí ha usado okada voice changer?, es un programa para clonar voces con IA y demas, si alguien lo ha logrado usar puede escribirme al privado

[covert lake]

#🌍│español

[winter ridge]

Hello

[elder willow]

Hello everyone, name is Lilian and keen to learn more about anything AI related in this amazing community. Thanks!

[idle stag]

hey can somone help im trying to convert and audio file into angel dust voice

or have it say it in angel dust voice

[covert lake]

what rvc are u using and whats ur pc gpu

[idle stag]

its fine i already did it

[feral halo]

Ayo? @idle stag level 1 !!!

[covert lake]

alr

[earnest elm]

anyone have link download blanco rvc ?

[covert lake]

u mean the voice model?

You can search rvc ai voice model at:

• #1175430844685484042
• #🔍│find-models
• https://weights.gg/ (login required)
• https://huggingface.co/models (but watch out cus in hugging face there arent only rvc ai voice models)
• https://applio.org/models
• https://voice-models.com/
• https://thevoicemodels.com/ (for Turkish Models, login required with discord and level 2 on their server)

if there isnt one, you can:

• #1159289738314919936
• #1191429836321849435
• make it yourself with our docs guides https://docs.aihub.wtf/essentials/how-to-make-voice-models/

[hardy rose]

i dont think the doc link is working

[covert lake]

#📰│dev-updates message

[coral perch]

Can somoene teach mw how to use the ai voice

I mean the voice model

[icy pendant]

What are you trying to do, a cover?

[coral perch]

No

Not a cover

[icy pendant]

This is the moment where you tell us what you're trying to do but no stress

[frozen tiger]

Hello everyone
I'm new here
Is it possible to build an ai model using this

https://huggingface.co/papers/2410.07160

[latent echo]

hi

[coral perch]

So. How to use the model ai

I want to change my voice

[icy pendant]

Realtime voice changer.

https://rentry.co/VoiceChangerGuide

Go to downloads for your gpu, upload model is explained on the guide

[ionic pumice]

hmz

[elder willow]

yo

[ionic pumice]

hello

[astral owl]

hi

[ionic pumice]

hello

[real hazel]

heyyy

im obsessed with the world of ai

just riding the wave

currently obsessed with runwayml and rendernet

[jaunty ether]

does anyone have deezer?

i need someone to download a song off deezer for me

its nowhere on sc or spot

[river adder]

you can go online and find ways to fetch them

[jaunty ether]

i tried 2

none worked

[river adder]

for educational purposes there is this website called lucida

[jaunty ether]

Oh thank you very much!

[novel nymph]

Hi

Discord is banned in my country

Goodbye👋

[ionic pumice]

rip

[cinder hedge]

Hi guys, the wayback machine has been hacked

[covert lake]

I mean there are VPNs

[novel nymph]

Now
this is what I use

[covert lake]

Ye VPNs are good

[novel nymph]

Yesss

[ionic pumice]

hello duckduckgo

[ionic pumice]

isn't gamefi like a crypto thing

[upbeat sorrel]

guys hey

I need help Im looking the x model but to use g and d

like svc fork kinda thing

does anyone know where to look ?

[covert lake]

svc is old

use rvc

[chilly lake]

RVC is not a spring chicken either

[covert lake]

spring chicken?

[chilly lake]

a young person; usually used in negative contexts to say someone is no longer young:

[woven ether]

what is the best model pre train

[chilly lake]

default as a base is all you need 99% of the time

[covert lake]

pretrains can help for other languages

[covert lake]

true

Whats gonna replace rvc tho

[chilly lake]

well, I tried Snowie and Rigel that were supposed to have some russian training... imagine my surprise when the OG pretrain did a better job and came out less noisy

while those "trained" models had an accent

true mystery

[covert lake]

pretty weird as OG is trained on an english-only datase, VCTK

was it SnowieV3.1

[autumn tree]

‏Your device is broken

[chilly lake]

yeah, tried both snowie v3.1 and v1.2

[tacit oxide]

Btw if i want to delete program can i somehow delete all the programs i downloaded in 1 click somehow? Otherwise how do i find it?

[feral halo]

Ayo? @tacit oxide level 1 !!!

[icy pendant]

If you downloaded prepackages (rvc1006, applio vx.x, wokada, etc.) You can delete the folder and thats all

If you manually set up stuff youd know how to delete them so i assume its only prepackages

[ionic pumice]

does any1 know the name of this little guy

[wraith ore]

I have a question. Would anyone here know how to plug this audio into an AI and have it make similar tracks? I really like this one and am interested in getting a few more like it.

Track: Christmas Hip Hop Beat by Tini Mc (It is on YouTube)

[tardy vector]

no

currently impossible

[analog cosmos]

do a reverse image search

[ionic pumice]

that's what I did silly

[analog cosmos]

it links to a VN called SharedHouse or something like that, some of the character names are on their itch.io page

[wraith ore]

Crazy to learn! It seems simple compared to the feats I have seen others produce with AI and heard about on the news. Then again, I have only worked with text and image AI so I know nothing of audio.

[analog cosmos]

I hate it thank you

[earnest dragon]

yeah Suno ai paid can do covers and stuff of uploaded audio

[tardy vector]

i badly remade it just for this

[cursive bluff]

How can I make my own custom voice model?

[covert lake]

whats ur pc gpu?

[cursive bluff]

rtx 3070 ti

[feral halo]

Ayo? @cursive bluff level 1 !!!

[covert lake]

As you got a good PC, you can use RVC locally, you can choose between:

• Applio: A fork of RVC with some extra features like Applio TTS, same quality tho
• Mainline: The original RVC

[polar flax]

https://docs.ai-hub.wtf/essentials/how-to-make-voice-models/

[cursive bluff]

thx!!

[copper heron]

It feels like I have comprehended non-existence and reached the end of the world, when I know and understand what everything means. I have achieved everything I could and can retire because now nothing holds me back and I have achieved everything that was possible and impossible

[vivid hound]

How do i change roop cam language from russian to english?

[trim sandal]

what do i download for simple speech to speech?

[covert lake]

whats ur pc gpu?

[trim sandal]

its not that great

what's the like smallest and fastest one

[covert lake]

Facefusion is better

[covert lake]

rvc is the best one and what everyone uses

tell me ur pc gpu and what u want to do

like realtime for voice changing in calls, making models, using models in pre-recorded audios

[trim sandal]

i just want to make a quick ai cover

[covert lake]

could u tell me the name of ur gpu

or do u want me to just give u cloud (remote good pc) which is easier (but got gpu limits, u cant use it unlimited like local which is on ur pc)

If u really want a quick ai cover just use ilaria rvc zero

[stray merlin]

Ilaria RVC: CLICK HERE 🤗

Guide on how to use it: CLICK HERE 📝

Don't forget to thank Ilaria if you find it useful! 💖

[plush storm]

Anyone have a mommy voice model I can use?

[dense hamlet]

I need one too

[turbid cypress]

#🔍│find-models

[summer palm]

i exist

[elder willow]

Hey

[trim sandal]

thank you thank you thank you

[wraith canopy]

.

[covert lake]

@rigid gazelle use #🔍│find-models

[dark steeple]

good evening joined coz i was interested in ai covers of vtubers particularly el muchacho de los ojos tristes sung by korone lol

[earnest dragon]

Roop cam trash

[tacit oxide]

Bruv what is that??

[earnest dragon]

what is what

[tacit oxide]

Nvm i had just werid error

I fixed it already

[earnest dragon]

-tts

[rare sorrel]

This interaction has expired, use the command /guides realtime if you wish to see it again.

[covert lake]

real and its not even on github anymore

[covert lake]

u wanna learn how to do an ai cover? whats ur pc gpu?

[dark steeple]

6750xt

[timid viper]

🐖

[vivid hound]

I cant decide between roop cam and avatarify, i'm still pretty new at this but i installed all packages myself lol...Is there an alternate version better than these two? somebody said facefusion but i want more recommendations please. I'm an online streamer

[prime hamlet]

fivem

[chilly lake]

your choice is either a slow directml build or rvc, or cuda emulation with Zluda in Applio, or using colab/kaggle

kaggle with 2 GPUs will be faster than 6750

[dark steeple]

dont worry im too new to even understand these words

[vivid hound]

help!! Conda is not activating in terminal after installation....All i get is conda is not a recognized term in cmd. please help

[feral halo]

Ayo? @vivid hound level 1 !!!

[chilly lake]

CUDA is Nvidia's framework for compuration using a graphics card

AMD has nothing like that on Windows

but there are several shims that make possible to use AMD gaming GPU for machine learning stuff

[pliant parcel]

DM me if you are an AI programmer and interested in music generation. I've got an idea for you...

[elder patio]

where do i chat with an ia cuz i feel lonely

[pliant parcel]

Chat with a human

[analog cosmos]

real

[tacit falcon]

you scared him away

[ashen glacier]

Hello guys anyone willing to work from home with AI software upgrading company?

[chilly lake]

what does " AI software upgrading company" even mean?

[analog cosmos]

he didn't want to chat with real people

[stable elm]

guys there is whay that i can decrease voice delay?

[tacit falcon]

whay

[stable elm]

way

my bad

[elder willow]

-rvc

[rare sorrel]

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[elder willow]

Hi
whenever i use the illaria rvc it gives me a error message @covert lake

[covert lake]

What error message?

[elder willow]

when i use inference in huggingface

[cinder aspen]

ruben sim thick of it

[elder willow]

it gives me a error

[feral halo]

Ayo? @elder willow level 1 !!!

[elder willow]

how do i fix the error?

[covert lake]

show a screenshot of the error in #✨│ai-help

u should get an error message at the top right of the screen

[heavy zephyr]

i need voice Rabbids Invasion

[covert lake]

You can search rvc ai voice model at:

• #1175430844685484042
• #🔍│find-models
• https://weights.gg/ (login required)
• https://huggingface.co/models (but watch out cus in hugging face there arent only rvc ai voice models)
• https://applio.org/models
• https://voice-models.com/
• https://thevoicemodels.com/ (for Turkish Models, login required with discord and level 2 on their server)

if there isnt one, you can:

• #1159289738314919936
• #1191429836321849435
• make it yourself with our docs guides https://docs.aihub.wtf/essentials/how-to-make-voice-models/

[keen portal]

Is there a channel where you can help with line distribution??

[covert lake]

line distribution?

[delicate spoke]

Hi everyone! anyone here building in the AI Web agent space? Curious to learn your take on it and your thoughts.

[scenic plover]

developer to learn Generative AI

[hot yacht]

Чкт странно как-то

[elder willow]

Is there a ai that can change the style of music eg Elton john style to Rick astley style something like that

[feral halo]

Ayo? @elder willow level 2 !!!

[quiet nimbus]

What are these information's about dataset and batch size and 32k and else?

[feral halo]

Ayo? @quiet nimbus level 1 !!!

[quiet nimbus]

where are we supposed to insert them

[gray zealot]

in turkey discord is banned and ım connected with vpn boys

[vocal leaf]

Thought vpn was also banned

But no I guess

[earnest dragon]

yes it is an deeplivecam is basically roop cam just rebranded

[earnest dragon]

just use Rope Alucard next or Rope Live which supports one image or the superior dfm

[earnest dragon]

they do have hip and stuff but not many things support it and zluda is now being developed again with a company funding it and focusing basically only on ML workflows and stuff

[chilly lake]

HIP SDK has a lot of features missing, like STFT

OG zluda gonna take a long time to recover back to what a forked version does now

the author said at least a year

[earnest dragon]

Yeah at least zluda is being worked on with ml focus not the crappy non ml support

[earnest dragon]

did you see the new blog post this month for zluda

[chilly lake]

yes.. current zluda fork by Lee works fine with stable diffusion and rvc

[frank citrus]

⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⠴⠒⠛⠉⠙⠳⡄⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⢠⣴⣾⣿⣷⣦⣄⠀⠀⠀⡤⠟⠚⠛⠛⠓⡆⠀⢷⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⢀⣿⣿⣿⠿⠟⠛⠛⠳⣄⠸⣇⡀⠀⣀⣀⣰⡇⠀⢸⠓⢲⡀⠀
⠀⠀⠀⠀⢠⣶⣾⣿⣿⣿⣷⣦⣤⣤⣤⣼⠆⡟⠻⠿⠿⠿⠟⠁⠀⢸⡇⢰⡇⠀
⠀⠀⠀⠀⢸⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⡇⠀⠀⠀⠀⠀⠀⠀⢸⡇⢸⠇⠀
⠀⠀⠀⠀⢸⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⣇⠀⠀⠀⠀⠀⠀⠀⠈⣧⣸⠀⠀
⠀⠀⠀⠀⠸⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⠇⠀⢻⠀⢀⣀⣀⣴⠀⠀⠀⢹⠅⠀⠀
⠀⠀⠀⢀⡀⠀⣾⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⡏⠀⠀⢸⡇⣿⠀⠀⠀⢸⠀⠀⠀
⠀⠀⣠⣼⣿⣦⣿⣿⣿⣿⣿⣿⣿⣿⣿⣷⠀⡇⠀⠀⢸⡇⣿⠀⠀⠀⣼⠀⠀⠀
⢠⣾⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡿⠀⣿⠀⠀⢸⡇⢸⠀⠀⠀⡿⠀⠀⠀
⠈⠉⠁⠉⠻⢿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⢿⠀⠀⣸⠁⢸⠀⠀⠀⡇⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠉⠛⠿⣿⣿⣿⣿⡿⠟⠀⠀⢸⡄⠀⣯⠀⣸⠀⠀⢰⡇⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⣼⣿⣿⣿⠀⠀⠀⠀⠀⢯⠀⠘⡇⢹⡆⠀⠸⡇⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⢀⣾⣿⣿⡿⠃⠀⠀⠀⠀⠀⢸⠀⢠⡇⢸⡇⠀⢰⠇⠀⠀⠀
⠀⠀⠀⠀⠀⠀⣰⣿⣿⡿⠋⠀⠀⠀⠀⠀⠀⠀⢸⠀⣼⠀⠀⡇⠀⣼⠀⠀⠀⠀
⠀⠀⠀⠀⠀⢰⣿⣿⣿⡄⠀⠀⠀⠀⠀⠀⢶⣯⣭⠤⠿⣠⣖⣃⣀⣹⡄⠀⠀⠀
⠀⠀⠀⠀⠀⠘⠿⠹⠿⢿⠆⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠉⠉⠉⠉⠉⠀⠀⠀⠀

[chilly lake]

if there's an improvement in the og fork, maybe by the time they get it working for ML we'll have some competition by AMD and S.C.A.L.E.

regardless, if you got a GPU < 7900 you're kinda not AMD's priority

they are not going to make 'gaming' GPUs viable with HIP SDK or ROCM

[earnest dragon]

yeah but missing cudnn and stuff because was wondering if replacing directml with zluda in DeepFaceLab would work fully with all features and better performance.

[earnest dragon]

with the new combined architecture yes they would but we don't know when that is going to happen.

[chilly lake]

you can always try and compare

Zluda does not take that much effort to plug in

but i'm notsure about tensorflow.gpu

probably not?

[earnest dragon]

yeah but fp stuff and cudnn but directml works so wondering if there is a better directml way except a full pytorch rewrite but I don't want to do that just first looking for someone to implement non nvlink and non crossfire multi gpu support so that you can do amd multi gpu or nvidia or cross.

[weary quiver]

Hey I am trying to make an AI streamer, I have everything set up beside a real time tts, I tried to find solutions for this but I can't find anything, does anyone have any starting points ? I know there is the classic tts but I wanted something more enjoyable to listen to. I was willing to learn how to train a custom model but I couldn't find if that would actually work and how I could proceed. Thank you 🥺

[stone roost]

gbchgj

[earnest dragon]

what is the goal for the real-time ai

[weary quiver]

either streaming or just an ai assistant for now. I have everything set up beside the tts because I didnt know where to start.

[lean drift]

how do i install rvc im completly new

[feral halo]

Ayo? @lean drift level 1 !!!

[lean drift]

it looks super confusing

[cursive spade]

Hi guys, I'm new on training AI Voice models, what does the D, DUR, and G stands for on the outputs? (i.e. D_1000.pth, DUR_1000.pth, G_1000.pth)

[heady pond]

Do you know if the program to change the voice can be used with an RTX but with 4 VRAM?

[molten stone]

You can try online methods using w-okada's voice changer colab/kaggle

[weak scarab]

Has anyone tried streaming or doing youtube with an RVC model?

[chilly lake]

DUR - no idea, D and G are training weights for discriminator and optmizer

[dusk totem]

hi, how do i open a ticket regarding someone banned from the server?

[earnest dragon]

A bunch of people have

[earnest dragon]

what did they get banned for it probably was a good reason

[dusk totem]

my friend didnt even send a single message since he joined lol

or never talked with someone in the server

[earnest dragon]

did he do something in vc

[dusk totem]

he never talked with someone here or sent a single word

[earnest dragon]

are you sure he isn't lying

[dusk totem]

yeah?? and mods can check the audit log

https://prnt.sc/ZAKnOElQrdep

here is a ss

[earnest dragon]

just ping nick or one of the mods or admins

[dusk totem]

@forest quarry hey can u help with that?

[forest quarry]

tell him to appeal, they should have gotten a DM from one of our bots to do so

also, would you mind giving his id here

[dusk totem]

852774964276494346 that's his id

[feral halo]

Ayo? @dusk totem level 1 !!!

[forest quarry]

confirming, is it @floral swiftr?

guh

xanaxluvrr

[dusk totem]

no, @floral swiftr

yeah

[earnest dragon]

that's his alt it seems so he is here

[forest quarry]

seems like they were banned for being a "nazi"

[dusk totem]

that's not him

[forest quarry]

probably related to their pfp or bio smth

[jovial viper]

had a bit of fun with the voice models and played it through some train announcements XD

[feral halo]

Ayo? @jovial viper level 1 !!!

[jovial viper]

before: https://www.youtube.com/watch?v=tk1NiU1GwPo

[earnest dragon]

yes then why are the discord names basically the same

[forest quarry]

if he changed it we can unban him

[dusk totem]

so if ur name is jacob and someone is named jacob its basically your alt?

[earnest dragon]

#1159290752195633273 only

[earnest dragon]

no that name is more specific

[dusk totem]

his profile is clean and u can check it urself

[jovial viper]

i cant upload videos to channel stoopid

[dusk totem]

lowkey, dont use ur brain again tbh

[forest quarry]

alr I'll do that then

[dusk totem]

ty for ur help

[earnest dragon]

yeah only links because they want stuff of this server for legal reasons

[forest quarry]

there we go

[earnest dragon]

was there profile clean and wondering what his ban reason was

[forest quarry]

yeah it was just "nazi"

since i believe they changed it I'll unban

[acoustic sparrow]

.

[forest quarry]

well... ive already done so

[acoustic sparrow]

that guy isn't related to me in any way

[earnest dragon]

yeah they deserve the ban if they even joked about that which is not a joke

[acoustic sparrow]

I dont want to be disrespectful but you look retarded

[cursive spade]

I am trying to train on a TTS project, specifically Melo. Is the number after _ the epochs? Why is it so big compared to when training on Applio? For example: G_1000, G_5000, etc

[feral halo]

Ayo? @cursive spade level 1 !!!

[earnest dragon]

What is your problem I just said that deserves to not be unbanned because you shouldn't joke about that stuff.

[acoustic sparrow]

It is the staff's decision and it has nothing to do with you, so respectfully mind your own business ❤️

[jovial viper]

what 💀

[polar flax]

https://tenor.com/view/ai-sponge-embed-perms-gif-2722780943144073359

[forest quarry]

you've just been unbanned so please don't give me an extra reason

sorry, its 10pm, dont wanna deal with much stuff

brazils drawing with chile

thats enough for me

[earnest dragon]

yeah all I was saying is most servers would give no second chance that's what I was saying is all

[forest quarry]

fair, imho though, i think it's good to give second chances for certain things

[acoustic sparrow]

It is 4 AM for me, and I’m being respectful w him I didn’t talk w disrespect im just saying to him to care about his own problems

[feral halo]

Ayo? @acoustic sparrow level 1 !!!

[earnest dragon]

but for nazi stuff that's different

[acoustic sparrow]

Lowkey why does it matter to you

You’re taking it like it offends you so much

[chilly lake]

number of steps

if during your training in Applio your steps are very small, then you're doing something wrong, like training on a huge file without splitting it

[earnest dragon]

use GPT-SOVITS for tts

[acoustic sparrow]

https://tenor.com/view/yungviral-gif-18022495009404817544

[earnest dragon]

https://tenor.com/view/embed-fail-dissapointed-sad-teary-seal-gif-3391051642056876120

[polar flax]

Applio isnt for training TTS models

[acoustic sparrow]

Maybe cuz i aint on dc 25h per day

[earnest dragon]

why Diddy gif

[acoustic sparrow]

Because ur his next victim

[earnest dragon]

yeah you're weird.

[acoustic sparrow]

😭😭😭 yeah callin me weird for that is weird too

https://media.discordapp.net/attachments/1125818983870906420/1131601604722839643/ezgif-3-3900fa2eb6.gif

[earnest dragon]

no for saying diddy stuff like that and also the nazi stuff you got a second chance for is weird

[cursive spade]

yes, I am training TTS using MeloTTS. Just wondering why the outputs have different numbers. Very new to AI training 😓

[acoustic sparrow]

I wonder why u care so much about that nazi thingy

Are you jew? Thats the only explanation

Also regarding diddy stuff pls learn how to take jokes/memes as not serious stuff lol

[earnest dragon]

No but it's antisemitic

[acoustic sparrow]

Even if it is, there is better topics to talk about and needs more attention

[polar flax]

do you want another bonk?

https://tenor.com/view/bonk-gif-19410756

[acoustic sparrow]

I asked a question about my pov

[earnest dragon]

the only joke about Diddy was the baby oil and the rest is not a joke

[acoustic sparrow]

Who said that lol

[earnest dragon]

you were making Diddy jokes that are stuff you shouldn't be joking about

[acoustic sparrow]

Learn how to not take jokes seriously

For the second time

[earnest dragon]

it seems like he might be wanting one let's see

[acoustic sparrow]

Hmm, someone is mad over me getting unbanned?

[earnest dragon]

yeah but certain stuff is stuff you shouldn't be joking about in society

[acoustic sparrow]

Well thats my jokes and if u don’t love it just ask to not joke w u

[feral halo]

Ayo? @acoustic sparrow level 2 !!!

[acoustic sparrow]

Easy asf

[earnest dragon]

so you think the thing you got banned in the first place for is fine just wondering

[acoustic sparrow]

Anyways just fuckoff me idk why u care so much aboutme getting unbanned maybe go tweet or talk about the genocides it deserves more attention

[dusty glacier]

You mean 𝕏??

[acoustic sparrow]

Prove that i did that n ill answer

💁💁

[earnest dragon]

Nah 𝕏/twitter is for people like you because you like the lack of moderation

[earnest dragon]

the mods have proof of the ban reason

[acoustic sparrow]

I dont even have an account on twitter or ever installed it

[dusty glacier]

You mean 𝕏??

[frosty bluff]

What's Twitter?

[acoustic sparrow]

X

[earnest dragon]

You mean Bluesky

[frosty bluff]

Ohhhh, Why didn't u say so?

[frosty bluff]

I did, infact

[acoustic sparrow]

Cuz why not saying twitter

[dusty glacier]

You mean 𝕏??

[acoustic sparrow]

It doesn’t matter

[dusty glacier]

You mean 𝕏??

[acoustic sparrow]

Lol

[earnest dragon]

bluesky is run by the previous twitter owner so that is the true twitter and 𝕏 is just what Elon musk wanted. it's basically the same as what the signal app creator did sold whatsapp to Mark Zuckerberg then made signal which is what he wanted whatsapp to be

[dusty glacier]

You mean 𝕏??

[weak scarab]

Hey guys

what program is the best? I use W okada rn

[earnest dragon]

the fork of it is better

[weak scarab]

What is that?

[feral halo]

Ayo? @weak scarab level 1 !!!

[weak scarab]

brother pls hit me up with da sauce

[solar agate]

they not like us they not like us they not like us

who agrees

[weak scarab]

why do the voice samples in the #1175430844685484042 sound like 10x better than when I import it to okada

like wat

[solar agate]

bruh i have a question for you

[weak scarab]

what

[solar agate]

why so fenious? fein! fein! fein! fein! fein!

[weak scarab]

what

[solar agate]

are you in the thick of it?

[weak scarab]

what does fenious even mean

[earnest dragon]

-rt first link

[rare sorrel]

This interaction has expired, use the command /guides realtime if you wish to see it again.

[solar agate]

alright about this one

from the screen to the ring to the pen to the king wheres my crown thats my bling

what u think of that one

@weak scarab do u eat lunchly while watching hawk tuah podcast aka talk tuah?!

[feral halo]

Ayo? @solar agate level 1 !!!

[weak scarab]

you think im 15 or what

lunchly is like the worst thing

[solar agate]

based off of this sentence i think youre 15 and a half about to turn 16

lunchly is for everyone!

[weak scarab]

I wont lie u type like an ai

[solar agate]

woah oh oh im in the thick of it

im using this ai for my youtube channel

i have 1 subscriber

[placid perch]

@opal marsh hello genai i mean lusbert
crap

[graceful violet]

get ready to witness a new era of humanity

[weak scarab]

oh damn

interesting

[weak scarab]

damn its better u right

[feral halo]

Ayo? @weak scarab level 2 !!!

[earnest dragon]

tesla promises are unlikely and also nvidia and stuff has been doing it first.

[weak scarab]

do you have any idea why sometimes it crackles? lioke the voice becomes super high pitched randomly

I went through the "crackle fix" at the document but its still happening

[earnest dragon]

crackling is caused by settings being to low or higher for your pc to handle

[weak scarab]

oh so my pc cant handle it basically?

I have 288ms and 2.7s extra rn

3070

[earnest dragon]

yes it can as long as the settings are set correctly and did you try the reccomended settings for your gpu

[weak scarab]

I will just adjust it and fine tune it but wow this is wayy better than the regular w okada

[feral halo]

Ayo? @elder willow level 4 !!!

[finite pawn]

Yo I urgently need a RVC google colab ai trainer, anyone have any suggestions?

[feral halo]

Ayo? @finite pawn level 1 !!!

[finite pawn]

Or anything that could train my wav file

[broken raft]

-colab

[rare sorrel]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], by Eddy, Hina and Gdr Google Colab
• UVR5 NO UI, by Eddy Google Colab
• UVR5 UI, by Eddy Google Colab
• Modified W-Okada's Voice Changer, Google Colab
• 🆕 FaceFusion UI, by Nick088 Google Colab
• 🆕 FaceFusion NO UI, by Nick088 Google Colab
• 🆕 EasyGUI, by Rejekts Google Colab

ℹ️ Note

While the Colab free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[broken raft]

@finite pawn

[finite pawn]

thank you

[cursive mirage]

rvc model wont work for me; I recorded my voice with the changer on and it's my voice normaly. ( I am new to this so I might just be doing it wrong.)

[finite pawn]

it works but an rvc model that actually works is seldom

[cursive mirage]

is there any way to trouble shoot it?

[earnest dragon]

A bunch of money and time and gpu server would be needed and also tons of training data.

[proven python]

bro someone give me a good, voice model

they are all badly trained

[earnest dragon]

of who

[proven python]

female i guess.

[earnest dragon]

Then no you're a guy wanting to sound like a female

[proven python]

its only for a youtube video im making.

any good ones here?

[earnest dragon]

no to have it run on it's own and stuff and play games and talk and have vision you need a gpu server.

then what would it do then

still the requirements are high still

so just a chat avatar so basically the same thing just locked to the chat

it can't fully act like a normal person using the computer and stuff and locked to chat means a llm with a chat avatar.

then there is livetalking on github that works but would need a beefy pc for it.

[chilly thicket]

https://tenor.com/view/feaky-meat-tounge-gif-3271019151004286960

[polar flax]

https://tenor.com/view/furina-embed-fail-gif-10775077421402251353

[night lake]

https://tenor.com/view/embed-no-no-embed-megamind-megamind-meme-gif-25261934

[tribal cipher]

aloo

[polar flax]

they are still just humans, but with voice changer/TTS, but even if they are an LLM, yea they could respond to chats but they need to be real human to do game playthrough, practically

[earnest dragon]

no for the games they actually Train the ai to play the games it's self

[polar flax]

I still doubt their ability with computer vision and stuffs, compared to hard-implemented playthrough AI by the game dev to do speedrun

but if well-trained computer vision AIs may eventually be real, they could beat real humans in competitive games and create new issues

[valid fox]

is there any watermark remover for a local install?

[regal token]

Hey

[regal token]

You can always blur it out with ai tools

Is it a photo/video?

[odd sorrel]

Are you able to use the ai voice models in text to speech? And If so, how?

[feral halo]

Ayo? @odd sorrel level 1 !!!

[polar flax]

try other model checkpoint/loras and add signature, watermark in the negative prompt

https://tenor.com/view/sml-jeffy-sir-i-dont-know-what-youre-talking-about-i-have-no-idea-what-youre-talking-about-what-are-you-saying-gif-24510567

[covert lake]

There are different Text To Speech (TTS) AIs:

GPT So Vits: RVC isn't as good as GPT So Vits for tts, but gpt so vits (few shot tts, which means needs just a lil training for models) can't use rvc models (and viceversa), and its only limited to: english, chinese & japanese, if you wanna check gpt so vits instead, read https://docs.aihub.wtf/tts/gpt-sovits/

Freemium 11labs: A easy way to do TTS is https://elevenlabs.io/, you can't use RVC model on this but its a mostly premium easy way for good quality TTS

FishSpeech: FishSpeech is a 0 shot (no explicit training needed) TTS, if you got a good pc you can use it locally else use their site

With RVC Models:

RVC is natively for Speech To Speech, but forks such as ilaria rvc mainline & applio have built in tts (using Microsoft Edge TTS to make a generated tts audio, which i suggest you to choose a tts model that is the same gender and language of the rvc model you wanna use, and then convert it with rvc)

If you wanna do tts locally with RVC Voice Models (if you got a good pc):

• You can get Applio in our docs
• While Ilaria RVC Mainline here (no guide as of right now)

If you don't got a good pc you can do tts with RVC Voice Models on cloud:

• Ilaria RVC Zero (Running on A100 GPU, free fasted rvc on cloud) and the guide

• Use Applio UI Colab (with google colab T4 free daily limit gpu)

• if you don't wanna use edge tts, you could try another tts ai from our tts index and use the output as an input in rvc

[covert lake]

As you dont got a good PC, its better you use cloud (remote good pc) for training an RVC Voice Model:

• Google Colabs (4 hours of daily gpu for free, not much hours, but easy to use):
  • Applio (ui)
  • Mainline (UI)
  • RVCDISCONNECTED (no ui)
• Kaggles (a bit harder to use and needs phone number but gives 30 hours weekly of better gpus):
  • Mainline (UI)
  • Applio by Vidal (UI)
  • Applio by Shirou (UI, no guide as of right now)
• Lightning.ai (Kinda hard, needs login, no issue with web uis or anything, but only free 15 credits monthly):
  • Mainline (UI, No guide as of right now)
  • Applio (UI, No guide as of right now)

its better to use kaggle

id also say to check ur pc gpu in case its good enough

#1159290752195633273

else i will delete ur promos in every other channel

[covert lake]

it was also a fork of “roop nsfw”

[covert lake]

whats ur pc gpu and what do u want to do?

[covert lake]

try checking https://docs.google.com/document/d/12hCYJqNCFl6jWKoVvCxtwt2V6nSoilgi5La8dkZa1KY/edit#heading=h.xweoq2pdv4uj

[covert lake]

its better to use gpt so vita if u care about quality

#1159290752195633273 or else it will get deleted

[covert lake]

what are u using?

and whats ur pc gpu

[onyx root]

Hello!

[boreal arch]

help

anyone familiar with js

[hushed cedar]

anyone here uses tensorflow to make ai?

[hushed cedar]

i know some js

[frail pulsar]

Heee heeee

[glossy flare]

hello

where can i find ai models for voice changing

[covert lake]

You can search rvc ai voice model at:

• #1175430844685484042
• #🔍│find-models
• https://weights.gg/ (login required)
• https://huggingface.co/models (but watch out cus in hugging face there arent only rvc ai voice models)
• https://applio.org/models
• https://voice-models.com/
• https://thevoicemodels.com/ (for Turkish Models, login required with discord and level 2 on their server)

if there isnt one, you can:

• #1159289738314919936
• #1191429836321849435
• make it yourself with our docs guides https://docs.aihub.wtf/essentials/how-to-make-voice-models/

Also dont look at yt tuts

tell me what u are looking to do and ur pc gpu

[left thicket]

🫡

[cobalt urchin]

j

[sour gulch]

can someonet tell me how to do

it

[analog silo]

66

[covert lake]

how do what ?

and whats ur pc gpu

be more detailed

[slow nexus]

Does anyone know how to make a hierarchical structure in Neo4j, I have made structure(a basic ontology of one level) but I have to put more semantic meaning in my knowledge graph I have restructure the whole basic ontology to make it more meaningful(like to make classes and subclasses). I know the approach to built it on protege but I can't import rdf on neo4j. If anyone can help me please share your viewpoint. I have to knowledge graph on a Indian legal domain.

[sour gulch]

NVIDIA GeForce RTX 4070

[feral halo]

Ayo? @slow nexus level 1 !!!

[covert lake]

What do u want to do?

[sour gulch]

get a voice changer