🧬│ai-chat

[placid schooner]

Says thare not ready

[whole abyss]

The song?

[placid schooner]

Yeh

[whole abyss]

I can't share files on here

[placid schooner]

Add me then send it

[whole abyss]

Donzo

[elder willow]

hey one of my runtimes crashed because i exceeded the runtime what do i do now ?

[feral halo]

Ayo? @elder willow level 2 !!!

[hollow compass]

how do i stop alien voice?

[ionic crystal]

Ok

[hollow compass]

im trying to turn vocaloid into ai voice

[ionic crystal]

Ok

[hollow compass]

is there a model that works well?

[feral halo]

Ayo? @hollow compass level 1 !!!

[hollow compass]

the ones i downloaded had an alien-like voice

[ionic crystal]

Isnt vocaloid Alien voice

[ionic pumice]

no

[hollow compass]

https://tenor.com/view/monkey-confused-gif-15712129875520007531

[ionic crystal]

[ionic pumice]

https://tenor.com/view/mayonnaise-cat-furry-beecat-meow-gif-26509467

[ionic crystal]

High ahh pitched voice

[storm meadow]

Medium-high end, what’s your GPU

[tacit plover]

🇹🇷🇹🇷

[earnest dragon]

the actual good ones not the bad ones like that haven't been released but they do show demos of it

[wary talon]

does it mean i cant give input of my own audio files and check it works?

[feral halo]

Ayo? @wary talon level 2 !!!

[wary talon]

sorry english is not my first language

[storm meadow]

Paris is in France :/

[tacit plover]

LoL

[maiden pebble]

Can anyone recommend a Docker container for a web UI to run voice isolation on an Nvidia GPU?

[earnest dragon]

I would say that no point in getting it working

[tropic garden]

[earnest dragon]

why docker

[deep nest]

.

.

[silver atlas]

#📢┋

[ionic pumice]

true

#📢┋

[elder willow]

one of my runtimes crashed because i exceeded the runtime what do i do now ?

did my training save?

rvc disconnected v2

[ionic pumice]

no

[elder willow]

wht do u mena

[ionic pumice]

you should use kaggle instead

[elder willow]

yes but did my training save

[ionic pumice]

way faster

[elder willow]

i let it run for 2 hours 😭

[ionic pumice]

no

[elder willow]

why not ???

no checkpoints no nothing?

[ionic pumice]

the trained models are stored in a local server iirc so, no

[elder willow]

fuckk

[elder willow]

u got a guide for that kaggle thing ? im training a model for the first time

[ionic pumice]

-kaggle

[rare sorrel]

📘 Kaggle Notebooks

• 🆕 Applio Notebook by Vidal
• 🆕 Applio Notebook by Shirou
• 🆕 Audio Separation by Shirou
• UVR5 NO UI by Eddy
• How to use RVC Mainline Kaggle by Cauthess
• ✨ RVC Mainline by Hina
• Original W-Okada's Voice Changer Kaggle
• Modified W-Okada's Voice Changer Kaggle

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

More commands: -colabs, -hf, -local, -rvc, -help

[elder willow]

thanks

[slate wraith]

hi

anyone learn in fl studio

[pseudo salmon]

so you guys think i can run RVC on 8gb vram? i have an rx480 gpu that runs great with koboldai

[earnest dragon]

free colab has 4 hour daily gpu usage

[maiden pebble]

Why not? I like the nature of containers.

[feral halo]

Ayo? @maiden pebble level 2 !!!

[maiden pebble]

Simple to setup, simple to remove.

[earnest dragon]

yeah mvsep have instruments at a part of song where audiostrip had just the vocals at the same part

[earnest dragon]

why no just use python environments and stuff it's better because docker uses more resources then the same thing without

[ionic pumice]

wuh :3

[earnest dragon]

windows or linux

[shell flame]

hello guys

there is a problem

[shrewd oyster]

🇫🇷

[shell flame]

idk why but in voice.ai it wants me to add an voice avatar but everytime i add its says minimum 400x400 pixel but i tried a lot photo still says that

[ionic pumice]

ew voice ai

-realtime

[rare sorrel]

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

[maiden pebble]

Linux

[maiden pebble]

Not really.

[earnest dragon]

docker is only good for Linux server installs

[upper hazel]

Hello i m Noobdog Beer

[maiden pebble]

And that's what I'm using.

[ionic pumice]

hello noobdog beer

[upper hazel]

just called me beer is way better :>

[gentle canopy]

hi

[covert lake]

we dont offer support for voice.ai here, we use wokada

bro fr thinks its that deep

[earnest dragon]

docker is only good for server stuff or stuff you are running 24/7

[earnest dragon]

you mean trash.ai

[earnest dragon]

no what's worse https://www.changevoice.ai

[polar flax]

not really surprising tbh, this was also terrible #✨│announcements message
many ppl seem also be enticed to kits.ai as well

[earnest dragon]

and they also have another program that does the same thing with paying but instead is for faceswap with roop/rope/insightface quality

it depends if the person had there ai voice of themselves copyrighted which isn't that common so not really unless that and most times it's only if making money and if not saying it's ai

okay then that should be fine it's more likely for you to never make money really if the song is copyrighted

[maiden pebble]

and i'm doing both

[polar flax]

you can try ask the voice owner for permission and convince them that the voice model training and any appropriate usages fall under fair use
but not only that, if the songs you want to cover are under UMG, SMG, etc, beware that it might risk to get copystriked

[ionic crystal]

Rizz rizz

You wont

[hasty hare]

helloo~

[earnest dragon]

no one would sponsor you for content with other people's music in it so patreon is best but no self promo in this chat and also you don't have the will power of mind/grindset for content creation

[silk jolt]

bro got roasted

just had to edit that like 3 times

[polar flax]

you cant request sponsorship to anyone, but you can make other kind of contents like video tutorials, game playthroughs, product reviews, etc.

[earnest dragon]

no I just had to make it more understandable for discord users that need more explanation or words for something

[silk jolt]

yea mb

sorry :c

[earnest dragon]

yeah but does he have the will power and the mind/grindset to achieve those things

[elder willow]

nah

[forest shore]

slt

[final obsidian]

Hello

[silk jolt]

whats a good programm to use for the ai voices?

[feral halo]

Ayo? @silk jolt level 1 !!!

[silk jolt]

the one i tried to use got flagged as a pickle

[wise umbra]

-rt

[rare sorrel]

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

[paper sluice]

Hello

[final obsidian]

Hi

I have a photo of my father, I want to clarify it, how can I do it?

This photo is with our national football player Arda.

[old gate]

Hello

[hexed lagoon]

Yo I need a really specific model and I can't find a high quality one

I'm just wondering if anyone has a high quality model of house

The doctor

[earnest dragon]

https://rentry.co/ForkVoiceChangerGuide

[silk jolt]

aight thanks

[pulsar perch]

what program do you guys use ?

i use something and it dosen t rly work

[covert lake]

for what?

[earnest dragon]

no one does but I eventually thinking of when I get a new gpu to do the voice and face for Dr. House

[pulsar perch]

can anyone help me ?

[pulsar perch]

voices

[earnest dragon]

use more words

[covert lake]

wdym voices?

be more specific

[pulsar perch]

i want to use a girl voice

but it dosne't work

[earnest dragon]

realtime or not

[covert lake]

ig in realtime for calls

[pulsar perch]

realtime

like on fivem

[covert lake]

dont watch yt videos

they are outdated

-rt

[rare sorrel]

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

[earnest dragon]

https://rentry.co/ForkVoiceChangerGuide

[hexed lagoon]

Ffs, do I have to do it myself?????

[covert lake]

also whats your pc gpu

[pulsar perch]

can you help me ?

[earnest dragon]

yep

[pulsar perch]

please

[earnest dragon]

I gave the guide

[pulsar perch]

is there a video guide ?

[feral halo]

Ayo? @pulsar perch level 1 !!!

[covert lake]

You can search rvc ai voice model at:

• #1175430844685484042
• #🔍│find-models
• https://weights.gg/
• 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/

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]

no

read the guide

[hexed lagoon]

You'd think there'd be a good model just based off the amount of material there is to work off but all of them are either shit or singing

[earnest dragon]

videos guides can never stay up to date

[pulsar perch]

oh

but what is the best girl voice
?

[earnest dragon]

yeah but I eventually want to make a high quality one and stuff with Beatricev2 for speed and then a DeepAscension DFM model for real-time

[pulsar perch]

and do i nedd voice meteer?

[earnest dragon]

follow the guide it tells you which one to get

[covert lake]

you mean upscale? you could use this https://discord.com/channels/1159260121998827560/1221468455723733094

[pulsar perch]

i press on all the numbers?

[covert lake]

follow the guide, also whats your pc gpu

[pulsar perch]

laptop 4050

[covert lake]

laptop?

[pulsar perch]

yes

shoud i use the clocks ?

[silk jolt]

atp just get a pc bro

[hexed lagoon]

Do you got high quality clips of house speaking?

[feral halo]

Ayo? @hexed lagoon level 1 !!!

[covert lake]

i heard laptop gpus are less powerful but i think ur gonna run the program fine

[silk jolt]

why did my msg get deleted lmao

[covert lake]

read the guide, it tells u everything, and if theres any issues u can ask in #🔍│help-w-okada or #1192011222023950368

[silk jolt]

i didnt even say anything offensive

[covert lake]

huh?

[silk jolt]

i said smth about a laptop becoming a frying pan

but it got deleted lmao

[hexed geode]

wsg guys

[silk jolt]

wsg

[hexed geode]

lol

[silk jolt]

i wanna know how people make those models

its like magic

[hexed geode]

which models exaclty are you talking bout

[silk jolt]

anything tbh

idk how people even make them

[hexed geode]

yeah nah thats not a valid response

[covert lake]

u mean rvc models which is what most models in this server are?

[silk jolt]

what was i supposed to tell you xd

[silk jolt]

ah yea rvc

[covert lake]

which are speech to speech, used for ai covers

[hexed geode]

the type of model you mean, like as he said RVC models

[silk jolt]

yea i know

[covert lake]

cus theres also gpt so vits which are text to speech

[hexed geode]

we also have tts yeah

[silk jolt]

mb didnt even those there were other types

[hexed geode]

and even loras

[covert lake]

in #1159496755218501692

[hexed geode]

theres a #🔍│help-ai-art in here too

[hexed geode]

yea

[silk jolt]

oh yea tts i wanted to try them out but my client i was gonna use got flagged as a pickle

so thats a no go

[pulsar perch]

if i increase chunk it gets procesed harder ?

[hexed geode]

bro wat ?

[silk jolt]

can i send links?

[pulsar perch]

yes

[silk jolt]

ops wrong reply

[hexed geode]

higher quality / longer processing time yeah

[hexed geode]

depends, im not sure if your current role even has the perm

[silk jolt]

but a pickle is smth that sends out your data or smth

atleast thats what i learned

[feral halo]

Ayo? @silk jolt level 2 !!!

[hexed geode]

a stealer?

[silk jolt]

dont take my word for it ask a proffesional

[hexed geode]

im just confused, tf is a pickle 💀

[silk jolt]

when importing a python module, arbitrary code can be executed

i dont fucking know

[hexed geode]

thats just a stealer

a payload baked into the python lib i guess

i heard about them doing that

[silk jolt]

i looked it up bro and it gave me a 7 page document explaining everything except what it means

xd

[covert lake]

i think they mean

[hexed geode]

TADC_Bubble??

[silk jolt]

not that

its some other clientbut pretty sure its the same

[hexed geode]

💀

[silk jolt]

i dont fucking know what it means and i dont wanna risk it tbh im too stupid for stuff like this

[covert lake]

what i meant is the pickle next to the voice models in HF, that was my first voice model

[hexed geode]

yeah nah ik, i was just questioning the voice model

[covert lake]

was my first one lol, like 1 year ago

[hexed geode]

lol i see

[silk jolt]

also so i dont keep embaressing msyelf what tf does that mean

[pulsar perch]

i can t belive it works

[hexed geode]

ok what are you even risking here?

[silk jolt]

i runno

[hexed geode]

https://play.ht

if you wanna use tts this is the best website

has free api, you can multiacc on it, really realistic voice cloning

i would say it even sounds better than tortoise

[silk jolt]

how hard could it be to make an rvc model

[earnest dragon]

no promoting paid services

[earnest dragon]

not that hard if you know how to get good audio

[silk jolt]

so its a skill issue

[earnest dragon]

basically is everything a skill issue

[silk jolt]

kinda yea

[covert lake]

never heard of that before ngl

hows it compared to 11labs and gpt so vits?

[earnest dragon]

yeah I have but it's paid for doing your own voices and only use is really making text to speech bots

[silk jolt]

are tts voices even good?

i got ignored 😔

[jaunty nymph]

hi, how can i get install the voice changer

[feral halo]

Ayo? @jaunty nymph level 1 !!!

[hexed geode]

its actually really good quality, dont think it beats 11labs but not so sure about gptsovits

[hexed geode]

its not even paid lmao it has free plans, and im not the owner or sum shit, im promoting it because of how much i love it, so that should mean something

[hexed geode]

i can show you an example

it doesnt have like very realistic emotions ( you can choose the whole text to have a diffrent emotion ), but by default its not so expressive, but still sounds alright in my opinion

[silk jolt]

wtf

thats actually good

yo i need to know how to set that up 👀

[hexed geode]

theres no setting up, its online

[silk jolt]

oh fr?

[hexed geode]

basiclly unlimited free usage on the webui, but if you wanna use api a 15000 character limit/month

[hexed geode]

yeah thats why i love it so much, and promoting it fr fr

[silk jolt]

damn lemem try it

[hexed geode]

go ahead, no ones stopping you lol

[silk jolt]

i need to sign up right?

[feral halo]

Ayo? @silk jolt level 3 !!!

[near root]

i wonder if giving an inference audio of someone speaking/singing slowly, and then speeding up the output would make it sound more realistic. has anyone tried that or do i have to go to my closet and record me singing slowly

[silk jolt]

i guess its just you idk

[hexed geode]

yeah, it asks for a phone numb but no verification so put a random numb,

ngl they should add more shit to their website to prevent multiacc

[silk jolt]

didnt even all i did was add an email lmao

[hexed geode]

you can signup without a phone num?

didnt know that

[silk jolt]

you can i guess

[hexed geode]

i dont think it would

[feral halo]

Ayo? @hexed geode level 12 !!!

[silk jolt]

so how do i get a voice tho

[hexed geode]

well you can either upload a raw vocals file to clone a voice ( it takes like 15 seconds to train fr crazy ), or just use pretrained voices

i cloned my own voice with my own dataset

some random asmr mommy voice

lmao

[silk jolt]

i dont want it to sound weird

but ur a women?

oh nah nvm mb

[silk jolt]

i thought u only did ur own voice

i fr wanna try my own voice tho

is it max 30 sec or can i do as long as i want for it to be better

or it doesnt matter?

[hexed geode]

lol

[silk jolt]

i fr thought ur voice was good but turns out everything about this is a lie lmao

since i know of this i cant trust anyone anymore

[hexed geode]

im not sure, it may ignore everything after 30 seconds, or may actually use it to train idk

[silk jolt]

il ltry i guess

any specific things i gotta say?or just say random shit

[hexed geode]

lemme tell u smth

that voice you heard

[silk jolt]

yea wsg?

[silk jolt]

mhm

[hexed geode]

was from a asmr JOI video on ph

[silk jolt]

what the fuck

[hexed geode]

it was alot of pain and suffering to clean the audio and prepare the dataset

[silk jolt]

nah bro

what

[hexed geode]

but yeah, it was worth it at the end, her voice is really good fr fr

[silk jolt]

i runno how to do allat tf am i gonna do to clean the audio??

[earnest dragon]

if does require paid to use your own voices and use them as much as you want

[silk jolt]

i think you missed the t

[hexed geode]

it doesnt, you can clone 1 voice for free 😂 dont critisize others if you dont even know what your talking about

and you can use it for free as much as you want, literally unlimited

[earnest dragon]

1 voice is not that much at all

[hexed geode]

but the api is limited, but its free too, just a 15000 character limit per month

[hexed geode]

well its free, its literally free, what do you want? you expect them to pay you instead?

[earnest dragon]

15000 a month onapi is not that much

[hexed geode]

yeah but compare it to 11labs, to use 15000 characters on there you have to pay a good amount im not sure how much but yeah

[earnest dragon]

why doesn't someone at a w-okada tts client and then play.ht would be useless

[hexed geode]

and your getting it for free also, no ones stopping you from multi acc

[hexed geode]

thats not how shit works bro 💀

thats not how ai works

[silk jolt]

yo wait

cant i sample a song and use it for the tts?

[hexed geode]

w-okada is using a voice to voice model, its a whole another infastructure for text to speech,

[earnest dragon]

yeah because if a local free way provided same features and comparable quality then it would be better.

[earnest dragon]

no they have a GPT-SoVITS tts client

[hexed geode]

i think you may be able to, if your cloned voice is singing then what you say in tts should be singing too

[silk jolt]

i guess but im not a singer nor am i a voice actor so it might sound horrible

i think ima just upload some random audio

[hexed geode]

english level 999+

[silk jolt]

whyd u delete that tho xd?

[hexed geode]

replied to wrong msg

[silk jolt]

oh xd

[hexed geode]

anyways @earnest dragon you are right, its just that the open source models are not at the point where they can beat closed sourced ones

and most closed source models are asking for alot of money to use the api or use them overall, so if you conclude all that play.ht is a really good website for what it is

[silk jolt]

will people actually pay just to sound like an e girl?

[hexed geode]

many people will

[earnest dragon]

but until you compare a play.ht voice to a high quality GPT-SoVITS or fish speech voice of the same person then you can say play.ht is better

[silk jolt]

also can i ask tf gpt-sovits is, im new here i dont know allat

[earnest dragon]

it's a tts model that requires little amount of data to make a model of

[silk jolt]

does it sound good tho?

[hexed geode]

well i would say play.ht may be better than GPT-SoVITS, well i cant say that for sure because i havent directly compared them but even if it wasnt better, its super easy to use, its online so you dont need a good gpu or alot of setting up shit, good for beginners

[earnest dragon]

yeah but haven't heard a w-okada tts client demo yet which might sound better

[silk jolt]

whats that

[hexed geode]

it sounds alright, around the same level as play.ht, i dont remember its quality rn, i cant tell which is better

[silk jolt]

also lemme guess the demo is paid?

[hexed geode]

i should be open source

[earnest dragon]

there is colab and kaggle and hugging face for GPT-SoVITS.

[hexed geode]

colab is just unrealiable gets the session termed, hugging face is ok, but to use your own space you need to pay, and kaggle, idk about that, its just always more trouble

[earnest dragon]

It isn't because then they can't make money from it.

[hexed geode]

i personally would for sure go through the trouble to use a better tts

[elder willow]

sdf

[hexed geode]

but people who are not familiar with these stuff will defenitely have problems with it

[smoky tangle]

.

[silk jolt]

i prob would too tbh

[hexed geode]

oh its not open source??

[earnest dragon]

colab is good if you have a module that makes it not disconnect and don't use gui or Tensorboard

[heady ember]

hello

[silk jolt]

hi

[silk jolt]

bro is 1000% a guy

[earnest dragon]

No doubt about it.

[heady ember]

This server you can find all kinds of famous voices

[silk jolt]

who are you talking to? just asking

[fringe vault]

.

[hexed geode]

lol

[silk jolt]

where did bro spawn from

[heady ember]

I'm saying this in general in case someone can answer me.

[silk jolt]

oh ur asking

mb i thought you were explaing smth to someone

[heady ember]

Yeah

[feral halo]

Ayo? @heady ember level 1 !!!

[silk jolt]

i guess theres voices here

[feral halo]

Ayo? @silk jolt level 4 !!!

[silk jolt]

idk if theyre any good tho

[heady ember]

ok

[earnest dragon]

he is probably talking to this server's vast void of people that want to sound like a girl.

[silk jolt]

am i the only one who wants to use my own voice?

imagine all my firends using my voice and talkign to me

too bad idk how to do shit

[heady ember]

They are in #1175430844685484042

[silk jolt]

yea

you can request models in #1159289738314919936 if theres anyone thats willing to do it they just might

idk tho

[heady ember]

to ask ooo

[silk jolt]

i spawned here like an hour ago dont ask me tbh

[heady ember]

Yes, I have seen 90 messages or so

you have written

[silk jolt]

so you were stalking me 🤨

jk

[heady ember]

mmmm no but if you want

[silk jolt]

bro what

[heady ember]

Do you want or don't you have a cat?

tell me yes

[silk jolt]

chill dude

[heady ember]

I want a photo cat

[silk jolt]

yea my cat sadly ran away like a year ago she went out for a walk and never came back 😔

[heady ember]

I am calm

@silk jolt uuuuuu no

[silk jolt]

@hexed geode what tools do you use to edit audio?

im gonna try to do smth

[thin ermine]

hhhhhhhhhhh

[heady ember]

https://tenor.com/view/let-me-out-cage-angry-gif-24075632

I can be your friend

[silk jolt]

no, im good

[heady ember]

do you want me to help you

[feral halo]

Ayo? @heady ember level 2 !!!

[silk jolt]

no, no im doing just fine over here

[heady ember]

but you want to be my friend

[silk jolt]

no sorry :c

[heady ember]

because

[silk jolt]

no reason

[heady ember]

hey I wanted to be your friend

[silk jolt]

i didnt tho

[thin ermine]

crazy rejection

[hexed geode]

audacity

pretty easy to use

[silk jolt]

yea i used that before but i deleted it and forgot its name

[silk jolt]

ty

i tried to make a cover once

that didnt turn out well

bro it feels like a maze opening something that was re-desighned

[warm sedge]

Hello, I can't open the RVC GUI, I get an error, can you help me?

[silk jolt]

go to #✨│ai-help i think

[warm sedge]

I can't post a photo there, can I PM you?

[hexed geode]

oh also

[silk jolt]

im not a helper

[hexed geode]

ultimate vocal remover

[thin ermine]

https://tenor.com/view/smiles-happy-gif-10778999771644652214

[hexed geode]

for removing instruments/background noise/echo and stuff

[silk jolt]

alr ty

i really think they should add a support ticket or smth

[fast crown]

is there a video in downloading and adding models i need help starting

[silk jolt]

you just need to download a model

• extract it, press on edit, and then upload, add the path file to the model and add the index in the index

pretty simple tbh

-edit -upload -model, (add the path file from the model you downloaded and extracted) -index (also just add the index file from the model you downloaded and extracted)

thats as much as i can explain tbh

unless ur using smth else which idk about

dont click that

wont the mods so smth about a seemingly suspicious msg xd?

nvm

its gone

@hexed geode this is awesome

i swear if i didnt know any better i wouldve thought it was real

well that was fun while it lasted

ima go now

bbye

[fast terrace]

what's the best free real time voice changer do yall recommend? (i use w-okada)

[fleet haven]

where can i find people that do commisions

[earnest dragon]

audiostrip.com free trial does a better job at that and even tested against mvsep and mvsep was not as good and also audiostrip takes not as long

[earnest dragon]

https://rentry.co/ForkVoiceChangerGuide

[light bone]

Definitely not in #1191429836321849435 and #1159289738314919936

[halcyon scaffold]

Why my music sounds like plastic?

[light bone]

Because you didn't sort it to the appropriate garbage can

[warm sedge]

#✨│ai-help pls help me 😦

[halcyon scaffold]

Can you help me?

[vestal gull]

guys how can i get a song that already exists to be sang by another person?

[austere sable]

rah

[ruby hound]

hm

[feral halo]

Ayo? @ruby hound level 5 !!!

[earnest dragon]

-local

[rare sorrel]

Suggestions for @vestal gull

🖥️ 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

Credits to Faze Masta and Antasma for compiling these links.

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

[earnest dragon]

because you did do it correctly

[ruby hound]

Idk what's going on

[hexed geode]

cool. will check it out later

[plush trout]

I got covid, bruh

[hexed geode]

nahh bluds living in the past

[plush trout]

Bruh moment

[feral halo]

Ayo? @plush trout level 3 !!!

[hexed geode]

ngl cringe

[ruby hound]

Ugly

[plush trout]

I hope you get aids

[earnest dragon]

how do you get covid if you are inside all day

[plush trout]

School

[ruby hound]

Bro

Are you serious?

[silk jolt]

bro what

[earnest dragon]

oh so you are not a adult yet

[plush trout]

Senior, 18

[ruby hound]

lol

[earnest dragon]

then how do you get covid if you have a shorter day usually

[plush trout]

Idfk

[earnest dragon]

you most lean in towards people then to get

[tidal imp]

Hey guys 👋🏽 It's great to be here with all of you! I have a strong background in product design and user experience. After recently experiencing a career transition, I am now looking to collaborate with an AI engineer/developer on an exciting new project. This project has the potential to be mutually beneficial, and I am eager to find the right partner to join me in its development. If you are interested or know someone who might be, please DM me. Thanks!

[ruby hound]

Covid atually exists?

[tidal imp]

Hey guys 👋🏽 It's great to be here with all of you! I have a strong background in product design and user experience. After recently experiencing a career transition, I am now looking to collaborate with an AI engineer/developer on an exciting new project. This project has the potential to be mutually beneficial, and I am eager to find the right partner to join me in its development. If you are interested or know someone who might be, please DM me. Thanks!

[ruby hound]

Bruh, i got ignored

[earnest dragon]

by who

[ruby hound]

Well

[earnest dragon]

yeah because you are wrong

[ruby hound]

But covid atually exists?

[earnest dragon]

yes it does so its not a question

[earnest dragon]

what kind of project because I'm looking for ai devs or coders to help with continuing a project

[tidal imp]

One is a SaaS based platform and the other is an AI plugin

What kind of project are you working on?

[ruby hound]

Sorry but what because?

[wispy ocean]

guys i installed a virtual cable but its messing with my audio

how do i delete it?

[earnest dragon]

I am working on a Deepfake / ai vfx /audio training tool called DeepAscension which I want to keep fully amd and nvidia compatible https://github.com/Cxsmo-ai/DeepAscension I just need coders and stuff to be able to create the features I want because the previous coders are too busy to help so looking for new ones

[strong solstice]

I just made a sick Ski Mask - Yonkers (tyler the creator swap)

[earnest dragon]

probably one image swap

[strong solstice]

its an ai cover

[inland compass]

Does my graphics card need to be powerful for the audio program?

[bleak niche]

guys how can i do a kanye ai voice clone for a song

help out please

[earnest dragon]

or when I heard swap I thought face because usually people say cover instead but it's okay

[strong solstice]

My fault!

[earnest dragon]

Kanye is controversial and stuff so I don't care for anything kayne has done since he became controversial

[devout sigil]

Alright I am a little confused. I want to train a model using preexisting audio I recorded. I would perfer a free version just to test an idea I have, but if the answer is in front of my face, let me know ;w; i think im getting confused with the TTS ones when I need a voice changer instead. Where I can put vocals into it, and it generates off my vocals.

[earnest dragon]

-local

[rare sorrel]

Suggestions for @devout sigil

🖥️ 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

Credits to Faze Masta and Antasma for compiling these links.

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

[bleak niche]

but he was still a good artist before all the weird shi he said but what if i asked for drake voice clone help can u help me

[earnest dragon]

still don't care for kanye

[bleak niche]

drake?

[zenith hamlet]

"i don't even know what i did,praying on you since was a kid" yea

[strong solstice]

Are you doing an AI cover where you swap songs using YE or just making a new ye song using his voice + a beat?

[earnest dragon]

Drake is weird but not as controversial as Kanye but would only do voice if also I was doing face

[bleak niche]

ye new ye song with a beat

[bleak niche]

lol

[devout sigil]

-overtrain

[rare sorrel]

Overtraining

You can detect if a model is overtraining if the TensorBoard graph starts to rise and never comes back down. An overtrained model will sound robotic, muffled, and won't be able to articulate words well.

Check these resources to learn more about this topic

• Epochs & TensorBoard from AI HUB Docs
• TensorBoard from 🍏 Applio Docs

[earnest dragon]

then you use udio and Kanye ai voice

[bleak niche]

idk how to do that like what websites on links do i need to use

[feral halo]

Ayo? @bleak niche level 1 !!!

[zenith hamlet]

And use mvsep bsroformer to separate it probably

[earnest dragon]

applio has built in overtrain protection that makes it so
it stops it when it sees no improvement

[earnest dragon]

no mvsep bs former I tested today and audiostrip.com free trial does a better job if you want the test audio I can upload them and send

[fallen ruin]

@covert lake Doesn't count spamming If I sent five images once 🖐️

[zenith hamlet]

What is audiostrip

[earnest dragon]

its a proprietary ai audio separation website and was founded and stuff by the queen Mary university of London and also they did a thing were they paid members money to use there songs and stuff as training data

[zenith hamlet]

What's their model's sdr

Or like the arch

[earnest dragon]

don't know all I know it's vb splitz 3 and 4 which is there own model

[devout sigil]

Where do i go for support relating to this, because I have already hit an error

[earnest dragon]

did it already overtrain

[devout sigil]

"This app cannot be installed" followed by access was denied

[earnest dragon]

send a screenshot in #✨│ai-help if not level 5

[hollow verge]

Hi
Guys is there a free vc changer for android

[earnest dragon]

no

[covert lake]

https://media.discordapp.net/stickers/1159265813270896700.png?size=160

You can just send multiple images in only one message instead of multiple messages

You are REALLY taking it deep acting like that meme wasn't about the n word

[strong solstice]

I just did h00dbyair with Kanye

[feral halo]

Ayo? @strong solstice level 1 !!!

[strong solstice]

was insane

[earnest dragon]

but he was saying there is a limit for one message

[feral halo]

Ayo? @earnest dragon level 14 !!!

[earnest dragon]

okay what

[strong solstice]

Wanna hear it XD sounds so sick

[earnest dragon]

okay

[strong solstice]

need perms

[covert lake]

the limit in one message is 10 images, he sent 5

[earnest dragon]

#✨│ai-help or send to https://pillowcase.su/

[earnest dragon]

then he's a dumb or just doesn't know

[strong solstice]

Bet

https://pillowcase.su/f/e2dd4b01be1909555615e310efb8b66a

[bleak niche]

no wayy

[strong solstice]

Its not perfect I did it very fast

[covert lake]

a mute is NOT that deep

[bleak niche]

itsss fireee bro

[earnest dragon]

yeah the way the kayne vocals are make him sound fruity or sus

[covert lake]

not like i do really care either, its discord

[strong solstice]

Yeah I am trying to fix that as of now

Evil Jordan with Kanye coming soon

[earnest dragon]

can I request a song for you to try to see how it would sound

[strong solstice]

Yessir

anyone can I am making them quick

Evil Jordan sounds pretty cool

[earnest dragon]

https://pillowcase.su/f/2d0d3c379fb8f0387d4d2439a1463c5d https://pillowcase.su/f/43699f1b655a0fda606306107af7b867 audiostrip.com vb splitz 4 separated

[strong solstice]

its not in english tho?

[elder willow]

tho

[strong solstice]

will that still work?

[feral halo]

Ayo? @strong solstice level 2 !!!

[strong solstice]

I speak en only lol

[elder willow]

me too but the full filthy frank kenya song is a banger

[feral halo]

Ayo? @elder willow level 1 !!!

[elder willow]

i love songs not in english

[strong solstice]

French songs are goated

#noirdesir

[earnest dragon]

it should work

[strong solstice]

https://pillowcase.su/f/03e8cf08f4684f12af7e12995fdb30d8

My Eyes w/ Ye

[novel spade]

how do I stop my mic from making weird sounds with the voice changer?

[wise quiver]

• lower microphone gain
• use noise surpression (the second one)

[neon ingot]

anyone got a good model to troll as a guy

?

[earnest dragon]

are you sure you're not a guy already

[neon ingot]

do you know one?

[earnest dragon]

no because was does a guy need a voice changer to troll as a guy

[neon ingot]

its for my girl friend

[earnest dragon]

are you sure you have one and discord only doesn't count

[storm meadow]

Just look for a random dude model

[neon ingot]

no i mean a friend who is a girl

[earnest dragon]

either way I ment

[novel spade]

thanks

[sacred bone]

hi

[opaque adder]

hello! Im here to change my voice

[gentle hollow]

Ah well, you came to the right place. There are a few ways to change your voice: one of them being vocal training through pitch, tone and other changes in vocal cords - and another, of course, doing the same thing but with the computer’s system

[earnest dragon]

-local

[rare sorrel]

Suggestions for @opaque adder

🖥️ 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

Credits to Faze Masta and Antasma for compiling these links.

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

[earnest dragon]

https://rentry.co/ForkVoiceChangerGuide

[toxic abyss]

anyone know that one website that is basically a catalog for ai voices but lets you use the ai voices straight from the website

I think the ending was .gg

nvm found it, it was weights.gg if anyone was also wondering this

[lament tinsel]

what are good settings for chunks (i need it to sound good)? i dont mind performance cause i have a pretty high end gpu

[earnest dragon]

this discord is the catalog #1175430844685484042

[wind stratus]

weights.gg

oh nvm you found it

[slim dirge]

👋

[dire bone]

Plz for Minato Aqua voice model owner, delate it because today is her graduation

[elder willow]

hey

[earnest dragon]

isn't graduation mean it's okay to have

[strong solstice]

who wanna hear my eyes with kanye instead?

[dire bone]

In vtuber community its not

[earnest dragon]

isn't that mean they retired or left

[polar flax]

typical asian idol industry

[polar flax]

then why Lulu Suzuhara (Nijisanji) models still exist?

[zenith hamlet]

They can't continue?

[dire bone]

Yes

[simple flare]

yo how do we use the voices

like make them say stuff?\

[placid schooner]

Can someone make me a song

[feral halo]

Ayo? @placid schooner level 3 !!!

[placid schooner]

About furrys taking over

[simple flare]

brah

[placid schooner]

Plz

[opaque adder]

got it all working great !

[simple flare]

what do we use to use the voices i now downloaded????

like what voice software

[placid schooner]

Can someone make flixy a song

[simple flare]

cause im lost guys

[earnest dragon]

-rt

[rare sorrel]

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

[earnest dragon]

-local

[rare sorrel]

Suggestions for @simple flare

🖥️ 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

Credits to Faze Masta and Antasma for compiling these links.

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

[shrewd notch]

is there anywhere i can get voice models for xtts ?

[verbal panther]

why isnt my rvc working

[feral halo]

Ayo? @verbal panther level 1 !!!

[final cove]

hi

[earnest dragon]

no but why not fish speech

[shrewd notch]

Is that better? I need something I can use with python

[feral halo]

Ayo? @shrewd notch level 1 !!!

[earnest dragon]

fish speech is the current best local tts and only needs 15s of audio

[shrewd notch]

I see

[earnest dragon]

https://www.bilibili.com/video/BV1wz421B71D/?vd_source=c0346336e29f5d074384ac4d4104f6f0

[shrewd notch]

Thank you

[earnest dragon]

https://github.com/AIFSH/ComfyUI-FishSpeech

[sharp knoll]

Do you guys know any text to speech bot with famous voices?

ty

[normal prairie]

yes

[earnest dragon]

no one uses ai jukebox anymore

because of how outdated it is

no udio is better

[feral halo]

Ayo? @elder willow level 3 !!!

[elder willow]

hlo

[lilac forum]

hi

[late bolt]

Hey if anyone here is interested in being a partner manager/seeker for my server pls dm me i need help!

[feral halo]

Ayo? @late bolt level 1 !!!

[mental zealot]

Guys, does anyone have a voice model Sweetie Fox?

[mental zealot]

I would be very grateful to you if you can really do it.

It's okay, it will be at least something

[feral halo]

Ayo? @mental zealot level 1 !!!

[mental zealot]

Yes, just a minute

[rocky elm]

Guys I made a voice model on weights gg, how can I post it here?

Yea, can I post a voice model here?

Ok

[worthy oyster]

hi

[flint turret]

hello

[frail valley]

.

[rough wasp]

.

[cinder kestrel]

Does anyone use ESRGAN on google collab for upscaling videos and images???

[subtle sinew]

hey

do you know an english/ american personnality with an italian accent ?

[kindred remnant]

How do i post my model

[fallen ruin]

https://tenor.com/view/rengoku-gif-21338177

[novel escarp]

hey, guys
does any AI model fetch the all data of rental cars in Houston?

[azure timber]

Hello World!

[zinc vine]

Does switching the advance settings in the Okada voice changer from F16 to f32 do anything noticeable for anyone? I tried both and don't hear any changes in quality. Nevermind, I'll stick with f32.

[snow sedge]

As long as you're fine with the tradeoff for x2 vram consumption and slightly more gpu usage

[zinc vine]

Yea. I saw that it's around 2GBs of VRAM taken which completely fine since I have a 4080 16gb.

[pine mural]

.wk

[indigo pike]

Bladee in AI HUB

👑  ⁦Nay⁩ - 107720 plays
 2.  ⁦kalr⁩ - 38267 plays
 3.  ⁦nex⁩ - 28627 plays
 4.  ⁦++⁩ - 25839 plays
 5.  ⁦peri⁩ - 23926 plays
 6.  ⁦maxi⁩ - 19021 plays
 7.  ⁦vici⁩ - 14344 plays
 8.  ⁦Monk⁩ - 13719 plays
 9.  ⁦garrenc⁩ - 12812 plays
 10. ⁦buga⁩ - 12069 plays
 11.  ⁦sniffmanjr⁩ - 11965 plays
 12.  ⁦mustard⁩ - 11550 plays
 13.  ⁦cesar⁩ - 11089 plays
 14.  ⁦libann⁩ - 10570 plays
** 696.  ⁦gublik⁩ - 4 plays**

cloud rap - drain - glitchcore - underground hip hop Artist - 961 listeners - 687322 plays - 715 avg

[sour vigil]

.wk 100 gecs

[indigo pike]

Error while replying: The bot is missing permissions.
Make sure it has permission to 'Embed links' and 'Attach Images'

[sour vigil]

brah

[covert lake]

what?

are you trying to share the models u made? check https://docs.aihub.wtf/extra/model-maker-role/

[covert lake]

https://docs.aihub.wtf/extra/model-maker-role/

[covert lake]

u should read https://docs.aihub.wtf/extra/model-maker-role/

[covert lake]

!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.

[covert lake]

zero shot yea ig

[covert lake]

whats your pc gpu and use the models how? like in calls realtime or ai covers

i dunno if pillowcase.su takes copyright laws like youtube, it would be better you upload it as a yt video and share the link in #1159290752195633273

[exotic dagger]

Léo!

LÁO TOÉT

[barren quarry]

C'est mon anniv aujourd'hui 18 ans les gars !!

[ionic pumice]

https://media.discordapp.net/attachments/1072878802612535396/1254314198901588039/Screenshot_20240622_184907_YouTube2.jpg?ex=66d00da1&is=66cebc21&hm=de4efc64c08a1f951da124a51c227303ef83f409b6fbecfdfbb11c64bc19f48a&

[ionic crystal]

Server would be dead if people stopped asking for help desperately

[ionic crystal]

They wont cuz they know if they shoot you the bomb explode

With no further information

[ionic pumice]

Ha

Ha

[placid schooner]

Hewwo

Can I have some help

[tall topaz]

[rare sable]

hi

[storm meadow]

#✨│ai-help #🔍│help-w-okada #1192011222023950368 ^-^

[rough laurel]

hi, I'm looking for a coder for my projects, if you want to help, write to me

[storm meadow]

What is the project?

[rough laurel]

Games

[livid ingot]

Hi

[rough laurel]

Hi

[elder willow]

sup

[ionic pumice]

who is bro talking to

[feral halo]

Ayo? @smoky tangle level 1 !!!

[smoky tangle]

.

.

.

.

......................................................................................................

https://discord.com/channels/1159260121998827560/1159514067187277865

[placid schooner]

No dfb

[raw osprey]

Every "furry" is nothing but a useless piece of shit

[indigo mantle]

Which software will be best to use RVC models and generate TTS for short vids?

[viscid nebula]

Hello

[valid ledge]

yo anyone have a good female voice model

[polar flax]

use chatGPT

[polar flax]

https://tenor.com/view/boykisser-meme-gif-3960706660224756602

[raw osprey]

Wtf is this degenerate gay shit yout sending while you are considered a model creator

[ionic pumice]

https://tenor.com/view/jojo-siwa-karma-boomerang-jojo-siwa-gif-13668189400813376244

[raw osprey]

@timid musk

Can you fix the server

[urban aurora]

@paper blaze where i can download best RVC?

[indigo mantle]

boop

[potent cargo]

yo is this plug ai serve

[elder willow]

that moment when your model makes even lesbians thirst for you 🤣

[potent cargo]

?

[elder willow]

ways to make rvc models without a gpu (aka with a cpu or online) for free?

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

There are better but my model is trained with Applio and is driving people wild in games rn lol

other non-colab options?

[feral halo]

Ayo? @elder willow level 1 !!!

[upbeat pivot]

guys help please

[elder willow]

You could use runpod

but it's not free I think

it's cheap tho

[upbeat pivot]

i need to use an ai model that's not in this server, how can I create one?

[elder willow]

There's so many options @upbeat pivot what kind of voice quality and such are you after?

[feral halo]

Ayo? @elder willow level 2 !!!

[upbeat pivot]

wdym

[feral halo]

Ayo? @upbeat pivot level 1 !!!

[elder willow]

That's the basic question you need to be asked to even be directed to the right place

[upbeat pivot]

wdym for "voice quality"

[elder willow]

Because some methods provide better quality of course

I meant voice quality as in better linguistics and pronunciation, sound quality, and such, not sure how to be any clearer than that for you.

[ionic pumice]

-docs

[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
• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[ionic pumice]

the cleaning out for datasets hasnt been updated in a bit tho

[elder willow]

thanks kar

[fathom trench]

How to download

[wheat grove]

can someone help me to use it

[covert lake]

i think u can also ask in #1159290752195633273

[covert lake]

There are different Text To Speech (TTS) AIs:

GPT So Vits: RVC isn't as good as GPT So Vits for tts, as rvc is for speech to speech, but gpt so vits 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

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

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]

whats your PC GPU?

[covert lake]

?

[covert lake]

cpu its not possible, on cloud/online u can

As you dont got a good PC, its better you use cloud 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)

[urban aurora]

nvidia

[covert lake]

so you want to train a voice model yourself? whats your pc gpu?

[covert lake]

yes, but which?

[urban aurora]

3080ti

[covert lake]

alr that rtx is good for both inference and training locally

check https://docs.aihub.wtf/ -> RVC -> Local -> Download either mainline (original rvc) or Applio (rvc fork with some extra features, same quality) on your pc

[urban aurora]

oko

[feral halo]

Ayo? @urban aurora level 1 !!!

[urban aurora]

okok

[covert lake]

for any issues ask in an help channel

read https://docs.aihub.wtf/extra/model-maker-role/

[covert lake]

download what?

[covert lake]

use what? and ask help in help channels

You can search rvc ai voice model at:

• #1175430844685484042
• #🔍│find-models
• https://weights.gg/
• 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/

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/

and don't share the dataset in any channel bc of copyright

[elder willow]

can rvc be used for inference if i have an ok cpu?

[elder willow]

other options? like, is the weights.gg one-model-a-day training still available?

[covert lake]

only for inference but it will be slow, its better u use cloud things like ilaria rvc zero which is free and runs on ZeroGPU (A100)

[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! 💖

[covert lake]

weights.gg training is not really free anymore

For training on weights.gg, you can:

• Pay
• Make a creation for 5 days streak to get a free premium training, you can repeat this
• Use referal code, getting 5 friends to subscribe via your training

However, weights.gg just uses RVC, you can just use that for training rvc models for free, if you want to do this you can tell me and I can tell you more info

Kaggle would be the best free cloud option

[fathom lark]

Who wants free nitro dm me

[gentle canopy]

hi

[elder willow]

ilaria might be the best for me

is there an ilaria rvc ver for training?

does gpu vram actually matter? is it feasable to train a voice model locally with 2gb vram?

[hollow tangle]

yes

[covert lake]

no

[hollow tangle]

no

[hollow tangle]

thank you i know im the best girl

[covert lake]

ilaria rvc mainline ye #📰│dev-updates message

[indigo mantle]

Went with Applio. It's amazing and super easy with TTS feature. Thank you 😄

[feral halo]

Ayo? @indigo mantle level 1 !!!

[fathom trench]

I downloaded zip in it dist and then what?

[covert lake]

what program and model are u talking about?

[trail forge]

hi guys have you tried the new flux.1 model?

[covert lake]

are u following a yt video?

[covert lake]

your welcome

[covert lake]

ye its very good, especially the dev one

[fathom trench]

Voice changer

[feral halo]

Ayo? @fathom trench level 1 !!!

[trail forge]

yeee finaly text and hands :D also no political censorship

and i can even run it localy woooo

but it is sloooow

[fathom trench]

vcclient or MMVCServer downoloaded

[covert lake]

i think you are talking about wokada

if u want a guide

-rt

[rare sorrel]

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

[covert lake]

1st link is the wokada fork optimized

2nd is the original wokada

also whats your pc gpu

[fathom trench]

Thanks

[covert lake]

damn how much vram do u have to run the dev one locally

[trail forge]

well i have a 12gb 3060

[covert lake]

yw, for any issues ask in #🔍│help-w-okada or #1192011222023950368

[trail forge]

i think the model is 15gb or something so it is sloooow

my ddr4 ram not fast xd

but i usualy run it at like 256*512 resolution and then upscale it using real-esrgan

[feral halo]

Ayo? @trail forge level 2 !!!

[trail forge]

have you tried running it localy?

[covert lake]

i have integrated graphics 🔥

i use it on cloud

[trail forge]

ohhh

i need to win the lotery and buy a 4090 :(

moreee tensor cores!

[covert lake]

real

[trail forge]

can someone send me 2000$ for a 4090?
i can...pay...with...ai voices and images :D

it is very good deal

nobody? :(

[scarlet spruce]

chat gpt aint work

[gloomy pecan]

hi

[trail forge]

hi

[grim gyro]

Hey guys

[trail forge]

hello

[grim gyro]

Hi

[trail forge]

noo i cant upload pictures

hmm maybe i can monetize the ai

[wooden vine]

i need help..

[kindred remnant]

Ok so apparently my model is not eligible for upload

[wooden vine]

at RVC my res ms is soooo high

[trail forge]

slow pc

[wooden vine]

whats the problem

[earnest dragon]

nah more vram with cuda cores is better

[wooden vine]

really..

[trail forge]

or too high settings

i think but i am no pro :D

[trail forge]

but i need ai cores not some basic cores

[wooden vine]

i have rtx 580 lol what do you suggest me about the settings

[earnest dragon]

cuda cores work with ai

[trail forge]

:(

[covert lake]

huh?

[trail forge]

you have what?

[wooden vine]

i mean rx 580 lol

[trail forge]

ohhh

[earnest dragon]

yeah that's the whole reason cuda is nvidia only

[trail forge]

hm

[trail forge]

well my 3060 strugles with real time voice changing

[trail forge]

i know, but i tought tensore cores are good at ai

[feral halo]

Ayo? @trail forge level 3 !!!

[trail forge]

so i need the most of them

[wooden vine]

my chunk at 576 rn

problem is the res ms its 26k

[trail forge]

i dont really remember but it also depends on the model and things

[wooden vine]

sometimes 30

[trail forge]

well the rx 580 is .... kinda slow

that is why i want someone to give me a 4090

[wooden vine]

lmfao you dont need it

[feral halo]

Ayo? @wooden vine level 1 !!!

[trail forge]

i need to run more flux

it takes a looooong time

[wooden vine]

what about mine hahahaha

cant even work

[earnest dragon]

yeah but you get way more cuda cores in gous so unless the lower amount of tensorcores is more powerful then the higher amount of cudacores then cudacores are better unless that

[trail forge]

xD well the gpus cost the same around the world....but the wages arent :(

[earnest dragon]

yeah because not enough vram

[trail forge]

we need a comparasion 1 tensor vs 1 cuda

[trail forge]

i need the 80gb A100

[earnest dragon]

but that will never happen and you only get the good one if you go the smx platform route

[trail forge]

:(

it is a shame that the 4090 cost like 2.5 average wages

nvidia does make a lot of money

maybe i can make my gpu hmmm profit

[earnest dragon]

yeah smx a100 80gb is 20k without any of the smx stuff needed to use it

[trail forge]

whad

expejnsivte

how hard can it be to desing a gpu?

nvidia, amd, intel, pampkin!

[earnest dragon]

smx

[trail forge]

smx?

[kindred remnant]

Doesn't match the source

[molten wyvern]

i need no introduction name endy

[trail forge]

what does that mean?

[earnest dragon]

is a different connector they made for them that has higher bandwidth and stuff and currently 2 l40s are better then a smx a100 basically same price but 96gb vram

[trail forge]

oh

i think the 4090 24bg will be enought

maybe you could...you know...send me 2k ?

pwease

[earnest dragon]

yeah but why get that when 5090 is coming out

[trail forge]

hmm if you pay for it it doesnt matter :D

[earnest dragon]

no if I had the money I would be spending it on my own pc

[trail forge]

:(

i can be your ai anime waifu

i will have some...like 15s of ping in my voice

please :(

[earnest dragon]

no that's weird unless I get free money but it's still weird

[trail forge]

oh ok, but i want free money

maybe i could work

but thats hard

[earnest dragon]

nah you are not getting free money

[trail forge]

:(

maybe i can sell someone the pictures from flux

they look realy good

[earnest dragon]

nah who would buy ai images but but non one image deepfake models yes people would buy

[trail forge]

i tried making like anime characters and they look like i just drew them
and real people look also amazing, if i use lower resolution you wont be able to recognise it

[trail forge]

yeee deepfakes go brrrr

[earnest dragon]

yeah that's why I mainly do that

[trail forge]

i made my friend into his favourite youtuber :D

i mean i can

but 25$ for something so easy...and people do it free

[feral halo]

Ayo? @trail forge level 4 !!!

[earnest dragon]

but that was one image probably so not the same

[trail forge]

it was a video

deepfacelab

[earnest dragon]

okay then what res model

[trail forge]

well dfl just replaces the face, and i made my model, i think it was 256*256 but that was back when i had a amd 8gb card

but the video was 1080p

or maybe 720

WAIT DFL WAS DELETED BY GITHUB

[small mist]

Hi, I need AI chatbot devs. Pls DM

[elder willow]

does anyone know how to get rvc training for free? (outside of colab and kaggle)

[trail forge]

you can host it

[elder willow]

will it work on 2 gb vram amd gpu?

[trail forge]

ehh i dont think so

maybe i can train it for you

if it is under 12gb

[elder willow]

dk

i kinda feel like you either need a beefy pc or money

but, if you really are that kind, yes pls

[trail forge]

sure thing

[covert lake]

don't do commissions here

ask in #1159289738314919936 or #1191429836321849435

[covert lake]

ask in #1159290752195633273

[covert lake]

no it wont

no need to ask it multiple times, its not possible locally in any ways

[elder willow]

i'll just wait for others to make the models for free then

[feral halo]

Ayo? @elder willow level 3 !!!

[covert lake]

those are the only 2 cloud ways for free, no other way for free

[covert lake]

you can make a request in #1159289738314919936 for free

[earnest dragon]

yes but I have my own fork that is better https://github.com/Cxsmo-ai/DeepAscension

[elder willow]

made 3

[earnest dragon]

no

[trail forge]

leme just donwload that

for safekeeping

[earnest dragon]

yeah amd and nvidia supported and why safe keeping when it's never going away because no deepfacelab mentioned in github post

[trail forge]

hm

and is it dfl?

sory me dumb

and why doesnt your youtube work :(

[twin bramble]

why is my voice so laggi

the voice changer takes to long to talk when i talk

[trail forge]

what gpu do you have?

[feral halo]

Ayo? @ionic patio level 1 !!!

[subtle cloak]

I am somewhat new with AI and am looking to generate videos. I'm a software developer and am trying to find a good open source project to start from, not something like tensorflow or pytorch, but a GitHub project where someone has already done a lot of the work already. Can anyone recommend a few projects to look at?

[earnest dragon]

yes its a deepfacelab foker and my yt channel doesn't work because I think it was falsely takendown when I clearly put ai in the titles and put the real video link in description but I can always make a new one

[earnest dragon]

all of those projects will use tensorflow or pytorch it's needed

[subtle cloak]

Right, but what are the names of the popular projects? I really don't know

Is there a list on some website somewhere?

[earnest dragon]

local ai video is currently nowhere near good enough

[earnest dragon]

but ai face and stuff and lipsync is

[subtle cloak]

Have a link to lipsync?

[feral halo]

Ayo? @subtle cloak level 1 !!!

[subtle cloak]

I don't want to pay to use unboring.ai/reface

https://github.com/ajay-sainy/Wav2Lip-GFPGAN

[earnest dragon]

those are trash DeepAscension/DeepFaceLab are the best and for lipsync it depends on how good of a coder are you

[earnest dragon]

not that good

[subtle cloak]

Like I said, no idea what is good or not

I'm a Senior Software Developer, have been for over 20 years

So, fairly decent

Mainly Python, but have been learning Rust