#🧬│ai-chat
1 messages · Page 114 of 1
Well, if you consider it stealing, then don't make a dataset.
buddy boy it is stealing.
its like the boy kisser meme
its copyright itelecual rights. it downplays va royaltys
atternys could have a field day
If you're using it for personal usage (which is what basically 95% of users do), it isn't a crime lol
i make content with it. so its not personal
Are you making money from it?
hi
not enough veiws
Ello
Then you're not stealing from anybody lmao
VAs are making more money regardless, and they will continue to be paid better after the whole AI shit takes over
tried using it and it was fun before i realized i suck ass at talking lmao
what about live embodiment?
like i pretend i am that character and use the ai as a helper for live interacion?
wait so being the character is not ok?
In that context of trying to make money from it, yyyyyyeah maybe not good
a lot of characters are copyrighted by the org that are associated wqith them
oh shi----
example: using gawr gura's voice online on a twitch stream would fall under impersonation
fair use
but ofc if you want gawr gura to read out your fanfiction personally then no god in the world can stop you

i can...

i used the JP rosetta voice and i spoke a bit of jp and suddenly fkin miyuki sawashiro actually came out
Ayo? @elder willow level 1 !!! 
jumpscared fr
hate it when the real VAs jumpscare....
@quartz roost
see
based reaction for teto
Alien pin kalomaze poopbert
real?
She also has a Japanese tts on talqu

釣 り で す た m9(^Д^)プギャー
*This video was made for April Fool's Day.
(公開当時の文章です。本編は存在しません。)
2022年4月1日でめでたく誕生から14周年を迎える重音テト、
遂にVtuberとして活動開始!
まもなく本編公開!乞うご期待!!!
以下敬称略
モデル:Tda式改変重音テト・デフォ服(Tda/vk)
https://seiga.nicovideo.jp/seiga/im8156709
エフェクト:Croquis改造エッジエフェクト(less than equal)
https://seiga.nicovideo.jp/seiga/im3705513
#TALQu 版重音テト
https://...
Ayo? @quartz roost level 48 !!! 
holy level 48

that mustve been so ass to tune lmao
tried making teto sing loli god requiem and it was such a painn
It's a teto tts, you type in and it's automatic
I also made my own for english
o thats sick
i dont know how to do any of that stuff so yall might as well be wizards
Teto is ahead of everyone else when it comes to speech software
Waiting for the teto voicepeak

god i cant wait until i get the certs to upload files lol
https://youtu.be/MnRfDvJYlyg?si=-4JAll4qEipAG6pj
This is teto rvc
HI!!!!!!!!!!! (dije hola saludame)
now this time I wanted to try something more about IA's and so, well, it's about teto ia, but not the one we all know from Synthesizer V, but rather our beautiful teto from UTAU, where with utau and a little bit of ia I have managed to train the voice of all the voices of teto utau with covers that I have mad...
this is different from teto SV i assume
Yeah it's UTAU ai
oh i use teto SV
No it's not available. It's forbidden to share any modified voicebanks of teto
Ayo? @elder willow level 2 !!! 
Tetochan has made the best sounding one so far
I'm redoing this with synthV properly
i can post yt links and not get assassinated right
Ya
Sent
I love that you covered a gumi song
I may have to get her
yea theres a synthV for gumi coming out soon
#shorts #synthesizerv #synthv #utau #vocaloid
does anyone know what ai websites ppl use to make those things like "marvel characters in medieval style" and stuff
aint that midjourney
whats a good deep female voice model
-egirl
egril
lolll
is there an alternative model architecture which can train on both your voice and the target voice so it is better at mimicking the target voice from yours rather than basically just shifting your voice into target voice's frequency range like RVC does at times?
i guess so
or at least some way to get the generated output to sound less like my voice
like if I try a model it'll basically just sound like pitch-shifted version of my voice and nothing like the target voice :/
.
AAAAAAAAAAAAAAAAAAA (had to introduce myself)
👍
ok howabout this?
i dont charge for making datasets but for model training and knowlege?
does covergen colab accept huggingface links?
I'm having an issue with applio, it's saying I don't have a module 'torch'
hello
wsg
Ayo? @steep yarrow level 1 !!! 
How’s it goin!
How u doing?
Guys
I have a question, and whoever gets the most votes, I'll do it
Vigil or Tachanka?
where do u download rvc?
(both from R6S)
Ayo? @brittle barn level 5 !!! 
u can do /poll create
yh
whats up poopy
yo yo wsg
Hey im oni
hai
I thought only mods could do that
i like that little hat bro
shoot that a difficult choice
thanks 
1️⃣ Russian man with LMG
2️⃣ North Korean defector
cmon look at this russian man

hello ai hub
common tachanka w
its time to grind and get to level 5 todat
yo bro
Vote Booster: Vote now for a 10% boost. https://arcane.bot/vote
lol
not much farther to go
ykw fuck it
feeling great lets get started
Lord Tachanka will be made
what the fuck is up youtube
today we're going to do a prank, gone wrong, gone sexual, almost died, insert 100 other shit
Watchu want? :3
wtf
real
,cs

bruh

ANDREWANDREW CHALLENGE (HE RESPONDED TO ME)
sad

fr
I LIKE BALLS (the edible kind of balls)
wait the context makes it worse
all types of balls are edible
fr?
i mean not tennis balls
has anyone here ever trained on a really big dataset? 😭
but animal wise all balls are edible
then the uhhh food kinda balls
i mean you could
like meatballs
🤨
my current dataset is about 12 hours of raw vocals something like 9gb of data 😭
uhhhh
chill on me
gah dayum
like bull testicles they are boiled and people eat them in other countries
idk if its even worth training that much
why tf
dam
i have a even higher amount
prob around 100k files of raw vocals of a artist 😭
bro is gonna train to 3000 epochs
bruh
exposed
i mean why do i have to train on big dataset if i can ged hq models even with just 10 min
true
well
BRO YOU GOOD??
i mean technically
big dataset is just waste of time
WTH
nah why is my dataset so trash 😔
@amber light
its clown behaviour
this model is going to almost never break, its going to be extremely versatile
ill let you guys know the results
give this man a special role rn
12 HRS DATASET
i hope 😭
LOL
Yes
Ayo? @amber light level 36 !!! 
if its booty cheeks imma cryit took me nearly 20 mins to pretrain

bros make i t the best model ever done
@glass fiber 12 hrs dataset
what artist btw
its juice wrld and he has over 300 sessions out 😭

YES
im bout to get clowned
OMG I ACTUALLY LOVE YOU
it could be bigger 😭
THIS IS GOING TO BE THE BEST
i hopelol
how much for 1 epoch
did you manually clean up the dataset
bro im not sure that your brain gonna take it
pre much its all raw vocals too and outtakes
vise gonna sit there for hours
days
lets not talk abt the bills ..
i just asked in here originally because i got this message before it started training
"That's a big dataset you got there. Log interval normalized to 200 steps from 1846 steps."
you got tensor board and all that stuff?
its using colab 💀
dont do it bro ure gonna look like a folklore after this 
😭
NAAHH
ik colab crazy asl but ion got a powerful enough pc (yet)
do you have at least premium
i plan to get a 4070 with a new intel i9 32gb of ram
yeah your shit gonna die out mid train
Nooo bruhhh 😭 i used to but i used it already
Quality> quantity
good luck with that
fr
fr bro if you really want to just use 2 hrs
for rvc more than a hour is useless
even 20 minutes are good ahha
20 mins minimum
i just realized...
@amber light we found whos gonna create RVC v4
who’s gonna create rvc v4
i could just train per each song

i think the biggest juice model on here is 90 min dataset
Ayo? @onyx epoch level 5 !!! 
LFG
so each model could be specific to the song
vise fr fr
@glass fiber
then you can use that to get the sound you need
is rvc v3 still happening
wrd brb
alex needs his mfer for help make rvcpp or whatever they renamed it to
or is anything updating with rvc 2
fuck v3 v4 where it’s at
im proficient in html
There's a free open source version of this called diffsinger
either that or we bring in rvc 15
oh i thought that would be in the rvc v32
real
nah maybe rvc 32.327
rvc but it only makes shitty models
Is there any reason everyone says Applio is bad lol
the one in charge
pooprvc
apple
rn
12 mins in and im half way to one epoch 🗣️
poop is the real admin here
Ayo? @glass fiber level 2 !!! 

damn im now head of rvc
Fair, but does it actively hurt performance/response or smth?
real
that was easy

Cute Santa hat
yk whats crazy, i remember the diff-svc days 😭
thx 
i mean there’s an explanation
amazing
just forgot what it actually was
crazy that I thought my shity model back then was crazy
that was only like last christmas too
Diffsvc has improved
fr?
-svc
Yeah I still use so-vits-SVC
It's on par with rvc
don’t it still take forever to train though
Yeah but you get more control
shit id idnt ping
nah the diff-svc colab is shit tho it doesn't work anymore LOL
i mean it is a bit slower but it wont affect the model if ur asking that
i wonder if tts would improve rvc
Your mistake is using colab
yeezybeaver sovits 2.0 colab
just use a tts then use audio on rvc
This guy fucks!
poolab v4
poop-svc
Bruh there's so many real ai tts already. But you can train your own like tortoise
fsfs tortoise cool
poop-vits-svc
Since I've done this:
- Install Applio
- Download a model
- Put acapella version of song into model
- Listen to output
But notably the output is different to the sample of the SAME song from weights.gg and sounds nowhere as good :/
What's the acapella from
And yes, I've messed with params
why is coeiroink so large 
If it's made from uvr it's gonna be shit
SynthV voice to midi sucks with vocals extracted with uvr
But if it's good clean real vocal recordings it does great
https://youtu.be/ATC9Qygio5M?si=-CtYsHPTl7iRXcYB
I just downloaded this and converted it to a wav LOL
I mean sure it's maybe not the best, but it shouldn't be THIS bad (I'll send output and sample in a sec(
This is what it's "supposed" to sound like...
https://generated-samples.weights.gg/cloam7n1p00v2wsj78gpy3b2h/5.mp3
That sounds extractsd
if my math adds up, it will take my juice model around 40 hours per 100 epoch of training 😭
I'll try this into synthV and see how messy the pitch is
Ah, fair
Still doesn't really explain why it sounds basically just like a bad impression when I run it with my voice...
id try it on a sample that is good to test
I just don't think it should be THIS bad
(I don't actually really CARE about this song tbh, I was just trying to do a sanity test and now I feel insane so...)
Ayo? @inner nova level 2 !!! 
I have a clean acapella for you to test
Alright, I'll try it, send it over :p
try a song without any background vocals
idk if you train models but if you want I can send you a mega (ik mega sucks) with all the sessions for juice if your curious @onyx epoch
the folder is about 185gb of his raw vocals
Alr, will do
(Gimme a sec I might have to fix some python stuff)
pls
owo cash
Wah wario!
(my python has died imma just reinstall :/)
hey i am kuro i am new here
f
guys this is what failing to install mangio does to your pc
f
a
- pip installs packages in python3.9
- pip3 installs packages for python3.11
- control panel shows 3.10 and 3.11 but no 3.9? wait wha-
hi
i just installed python
im still trying to uninstall it 💀

?
oh
It doesn't install it system wide
Hellooo
hai
hai
voice joker?
Hello
OMG LADY FURINA-
Hey it is I.
Heyyy I'm your biggest fan!
Furina character or my Furina content? Because I do be kinda Furina on content too XD
Oh sorryy I've never seen your content but I'm your biggest fan as Furina character XD
(I also had a Furina profile a day ago)
What do you all think is the best model
Aatrox
Furina
I use Furina voice model while also "physically" looking like Furina by using an Furina avatar on VRChat.
OH REALLY?
Yes.

A friend of mine who knows how to do this RVC stuff too; he made a Furina voice model for me and I like it more than the Furina voice models that pop up when I searched Furina in #🔍│find-models but he told me to join this server if I want to find any public voice model that should work with the same software.
NICE LOL

I'm mostly Furina in Silver Wolf's clothings but I have many Furina avatars.
Do you also look like Furina irl?
No I'n a dude lmfao
Oh shit

I'm a grown adult man with hairy arms and a beard so I would say like not look like Furina at all, very much the opposite.
WHAT
But I think is beautiful thing we can become anything we want to through voice models and VRChat avatars, if I want to be a cute girl online then I can be a cute girl online, lol
True

I'm huge Furina fan so it's awesome I can "cosplay" as Furina this way without being cringe or made fun of for being very much an opposite person.
My Furina content also kinda pops off so that's really nice too.
That's great to hear!
Ayo? @elder willow level 8 !!! 
Suggestion for @reef badger
banned
:0
I just like making video content as an girl from Genshin in online VR adventures. Lmfao
where could i find tutorials for bert vits 2
Wait are they actually banned or did they just delete their own messages? Lmfaaooo
na i banned em
Oh damn
Wouldn't you like Furina being your wife or do you want to be Furina herself?
Either are cool but neither are really possible in the end so in the end being Furina in VR with Furina voice model is the closest to these options so that's what I'm kinda doing. But I am also myself, is not like I'm trans or want to become a girl, I just like making Furina videos as accurately as possible.
Ayo? @reef badger level 4 !!! 
I see!
WITH THIS TREASURE I SUMMON
Do da homework
heyo
EIGHT-HANDLED SWORD DIVERGENT SILA DIVINE GENERAL MAHORAGA
5
nuh uh
Oohh, cool cool, almost there.
i wish i was over level 5
God i need to find a way to not make it sound as robotic
rip no perms
YOU CAN DO IT. I BELIEVE IN U.
Vote Booster: Vote now for a 10% boost. https://arcane.bot/vote
✨ Tip: Use /card to modify your rank card
F
shit im level 1
OMG YOU'RE ALMOST THERE TOO!!!
So close to lvl5.
fr
Fr fr
Use booster 
@wintry sinew
nuh uh
i was active during early aihub era but now im not even lvl 5 sadge
f
Vote Booster: Vote now for a 10% boost. https://arcane.bot/vote
i mean like 3 levels
yeah i need to talk more here
ye

umm hi?
can anyone tell me what tool do people mainly use for A.I covers?
does w-okada voice changer work on mac?
https://discord.com/channels/1159260121998827560/1167844377046040769 ilaria rvc (huggin face space)
https://discord.com/channels/1159260121998827560/1175615544037212200 AiCoverGenNOwebUI (colab)
one of these two
👋🏽
duplicate the space for no queue in the 1st one
says its unknown
thats for applio btw...
can you help me out with installing it on mac?
i rlly want to and it wont work...
soooo what's going on?
every time i open the start cmd, the whole terminal process doesnt happen
it just gives me a short message and says process completed
i opened the other thing first then tried, but still no luck.
hmmm what mac do you have? like what year
idk what year. i have a macbook pro
2019...
damn it
well you can use Colab though
any other way i could use it or another good voice changer simular to this one?
Ayo? @elder willow level 2 !!! 
i got a 2012 and thats how i do it
I tried that also, but i couldnt get it to work.
ill try again tho
alrighty
lmk if anything goes wrong, lemme get the link
don't need to use the download model step, you can do that manually.
in the gui
ok. im running step one
finished step one
what exactly happened??
what's the log saying
.
.
yo
hello
pp moment
🙏
bro this is so hard to do on a mac lmafo
Which software?
Do you have a M1+ Mac
intel
So you can only do online methods
Otherwise it will be slow
-rvc View the guide for RVC Disconnected
RVC Guides (How to Make AI Cover)
Translation by country
ok thank you
Ayo? @silk moth level 1 !!! 
how do i get guide maker role i have a guide for og rvc
guys, tell me 5 characters from any show/series/whatever, its for something
real quicke
Waluigi, Mysticat, Roblox Chezburger, and Smoke Detector
Kalomaze
a ok
heyo
Ayo? @twin socket level 2 !!! 
does anyone know what voice model duckus uses?
.
anyone got carti model for like
tryna feel how u feel, brad pitt, those session yk
which one work
Check the website weights.gg
It has a lot of carti models
ty bro precciate it
@limber slate
Information for new members!
Voice Changer help
#🔍│help-w-okada
Cover & training help#✨│ai-help
Searching for models#1175430844685484042/#🔍│find-models
http://weights.gg/
Hey do I need a role or level to request a model? Because the text is still grey
Yes
Oh crap for some reason I can't post images here
where can we share what we made
Kinda?
huh
If you uploaded a cover to YouTube you can link it in #1159290752195633273
oh ight
It’s a original song you can post in in #1159289417215787018
ass
Yes
What methods/applications does everyone use to improve their covers?
what is epchos
epchos lmao
hey
Hi
What is better? pm, harvest, dio, crepe, mangio-crepe, rmvpe or rmvpe_gpu??
Information for new members!
Voice Changer help
#🔍│help-w-okada
Cover & training help#✨│ai-help
Searching for models#1175430844685484042/#🔍│find-models
http://weights.gg/
does oshi no ko suck or no
I see my crush for real
Excuse me but can someone help me I'm having problems downloading the voice mod
want the link?
What software is that?
Sure ig
Yeah
https://huggingface.co/wok000/vcclient000/tree/main this is rvc v1
Ayo? @lean rock level 1 !!! 
the huggingface link for it
Do I just open it or what?
can i still use google colab?>
do u have a dedicated graphics card?
-colab
Suggestions for @old stone
- AICoverGen-NoWebUI, by Ardha Google Colab
- RVC Disconnected, Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- RVC-HF, by r3gf Huggingface Spaces
- AICoverGen, by r3gf Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
- Advanced RVC Inference, by neuclya Huggingface Spaces
You can find more info on the #1159513888199540817 channel. If you can't find your answer, feel free to ask for help in #✨│ai-help. Credits to Faze Masta and Antasma for compiling these links.
Yeah
should i follow what the bot says
Idk tbh
Ayo? @leaden kindle level 1 !!! 
or you
Goodnight everyone
how do u not know??? check ur settings type "device manager" and tell me what it says undr "display adapters"
Uhh you can do both if you want. If you have any questions you can ask in #✨│ai-help
Oyasumi
Oyasumiiii
It says uhd
uhd 600?
Idk it just says uhb graphics
yea u probs cant run this ngl
if its uhd graphics
Damn it
same graphics i had on my laptop and i couldnt run this at all
u can try the collab version
and see if that works for u
np
Do I go through the link you sent to get the collab or what
Thanks
yo
Hi!, a question, regarding using the voices (Tsukuyomi-chan and Amitaro) that voice-changer includes as audio for YouTube videos, is it necessary to give credit to the models, and would this be in the description part of the video?
i want some voice
how to search the model
In #1175430844685484042 or website weights.gg
They are in #1175430844685484042 or website weights.gg
thx
Ayo? @errant nexus level 1 !!! 
Yes. Always give credits in the description and/or video
how about the #🔍│find-models
You type /search and use the ModelAI bot. Not the Applio bot
Thx
hi
is there no dashiexp voice model? i thought there was one did it get deleted or smth
hi
Hello
i have done a rvc conversion in months --- does the google colab gradio web UI still work?
It's 8 AM for mee gn Lady Furina
Thanks for answering. And are there models that are 100% open source? Or with fewer limitations?
Ayo? @warm wigeon level 1 !!! 
I personally use harvest
High quality but long time
It takes 2-3 mins for me tho
how do I get model maker i could not find it anywhere
Where is the tutorial or download at I dont see it
what are you talking about?
Ohh you want to make a model?
i thought u meant the role
Information for new members!
Voice Changer help
#🔍│help-w-okada
Cover & training help#✨│ai-help
Searching for models#1175430844685484042/#🔍│find-models
http://weights.gg/
diluc voice ?
Ayo? @hidden root level 1 !!! 
could someone please assist me? i am very close to finishing making an ai cover, its just i dont know how to correctly download the voice (yes i checked guides)
Ok, I tested this and basically it sounds absolutely horrible and nothing like Cave Johnson
also I tried the same audio through weights.gg inference and it sounds so much better so... what gives?
I GOT A PC
First time?
Yes
Oh! Great to see a new PC user
I get my mouse and keyboard tomarrow
ı need a AL PACINO voice model
bro all them are viruses
What did you experience to say that?
oh viruses
From the voice models?
Ayo? @peak fog level 1 !!! 
Antiviruses might detect them as a harmful virus but they're harmless, it's like antiviruses detecting a 3rd party program as a harmful virus but that program needs to use that coding for its job
well u give me some girl voice changers?
I don't use any girl voice so you can look around in #1175430844685484042
yeah idk how to use them
You wanna change your voice in mic to a girl voice?
yeah bc im trolling my friend
how do i use it/
You can ask for help in that channel
I'm assuming Okada is compatible with RVCv2 models too?
well what software do you have installed for RVC?
Applio?
Mangio?
Okada?
that's just a model, it needs the actual software to use it
He means did you install any voice changer program?
i use a macbook
Install the program Okada, it's the best
bru no img perms
Yea
check help rvc
Tomarrow I get my mouse and keyboard (both in 1(not trackpad)) so I can set up my pc
Goodnight
Gn
-egirl
Suggestion for @peak fog
Hello!
Hi everyone
Hi
how much delay does the voicechanger have? I'm guessing it's almost realtime?
depends on your PC specs
I run it on a rtx 3070
desktop card?
yeah
okayy thanks!
Anyone know if RVC can be trained with audio pairs so that it works better with a specific input voice?
saw what?
Information for new members!
Voice Changer help
#🔍│help-w-okada
Cover & training help#✨│ai-help
Searching for models#1175430844685484042/#🔍│find-models
http://weights.gg/
try asking someone in a help channel
huh?
dead chat
that doesn't help me unfortunately, that's how you NORMALLY train RVC, which as impressive as it is, just doesn't work well with my voice
it just seems weird to me that RVC can be trained with only target voice with no input voice pairs whatsoever...
(Yes, I know that's like the whole point of the architecture but it doesn't like my voice so...)

idk what any of these words mean but it might be possible to do whatever ur trying to do
k
PLEASE IM BEGGING YOU IM SO LONLEY
Ayo? @royal maple level 1 !!! 
its 4:20 in the god damn morning
do i know u?
nope
help me pls, What's the problem?
[VCClient] wait web server...230 http://127.0.0.1:18888/
Information for new members!
Voice Changer help
#🔍│help-w-okada
Cover & training help#✨│ai-help
Searching for models#1175430844685484042/#🔍│find-models
http://weights.gg/
Sorry
go to rvc help wise guy 😠
noone is mad
huh
im just tryinna point this guy in the right direction
its all good here
smart people hang out there
oh i was joking lol
dw i got that
😭
i just cant tell what the newcommer thinks
im only in this server to make songs then fan girl about them
i think i might be gay tbh
weird flex but ok
yea i bet you dont listen to steven universe ai covers for hours at a time your like not on my level
😱
@grand oar this is geniral.. not #1159290752195633273
get a load of this guy
get a load of dez nuts in yo mouf
Ayo? @modest horizon level 4 !!! 
huh
🫢
Ayo? @royal maple level 2 !!! 
hi
If I train a model using a diverse range of emotions, will it result in a higher-quality model, or will it risk specializing in just one emotion?
if the emotions are in different tones it'll be alright
just don't get any screaming for too long, can cause problems long term
so i have to delete screaming?
bet you don't make 49 ai covers of the same character
mostly, yeah
it can't train screaming?
it can mess with the normal voice
I guess the buf: 600ms is the delay?
this is the chunk delay yeah
res: is your realtime delay
are there ways to train it with every emotion?
It can if u have it in dataset
I'd recommend training multiple models for each emotion if they aren't small
Meow meow
any guides on how to do it?
like if you have 5 mins of each emotion i think it's worth it to generate multiple models for each thing
i mean, there's a guide to train models in #1159513888199540817, the dataset part should be easy though, just use audacity and unite the samples with certain emotion and train a model with those
also, screaming should be separated from speech?
Hello
where can i post covers lol
can i combine them after generating?
theres only originals
generating the dataset or model?
after generating models for each dataset
alr cheers
if you wanna combine the models, you can
but I'd recommend just using one each for certain audios, or maybe inferring twice with two different emotion models
is there a large guide on how to prepare datasets, how to train models correctly and then combine them in one?
not really combining them, it's something very specific that is only in the mainline rvc fork
only to train models and preparing good datasets
**Method 1 🔎: ** The search accuracy is 100% 💕
Upload the <t:1695315432>
Server uploaded: AI HUB (RIP)
Channel uploaded: AI HUB (RIP) Post
Profile Owner: davisudo2000
davisudo2000
davisudo2000
10194
EN 🇺🇸
RVC V2
72 MB
N/A
300
0
pal those are rookie numbers
Ayo? @elder willow level 9 !!! 
could be but mine's are high quality
last question. are there any ways to speed up training voice model? except computer upgrading
can someone help me, i downloaded voice changer client from github, its working, i can hear it through my own headphones, however, i can not get it to play through discord or other apps
thank you
Ayo? @soft girder level 3 !!! 
np ;), glad to help
Is replay a good app to use? I hear applio is bad
You need a Virtual Audio Cable.
The one by Muzychenko or the one by VB Audio will work
Ayo? @forest quarry level 53 !!! 
What voice changer applications are there?
W-Okada and the built-in realtime voice changer in RVC mainline/mangio fork
would u be able to help me set that up? the yt tutorial i watched didnt show how to connect to virtual audio cable
i make good shit but i can never think of songs
here's how the input/output stuff will go for it to work:
W-Okada
Input: Your Microphone
Output: CABLE Input/Line 1 (on VAC by Muzychenko)
Monitor (optional): Your Headphones
Discord/Other Applications
Input: CABLE Output/Line 1
Output: Your Headphones
can i screenshare and u help me downlaod and walk me through the set up? its ok if not, sorry im a noob
sorry to bud in but what is the voice changer capable of?
and the downloading part of the virtual audio cable doesnt really take that much effort so dw
change your voice to anything else in (almost) realtimeo
omg what
do u have the link to download the cable
Ayo? @rotund echo level 1 !!! 
Audio Processing and Audio Programming by V.Burel
can i use rvc models?
thanks bro
and theres no other way to do it without a virtual cable? even if i have a goXLR or anything?
Whats the most fantastic girl model to use?
SO I CAN BE STEVEN UNIVERSE AND SAY CORNY LINES IN VC?
Is it kinda like voice mod but better?
Are all rvc models in weight.gg or does it depend
...yeah
much much better
mostly
yeah no other way, and idrk what a goXLR is
-egirl
Suggestion for @elder willow

In what ways?
because it literally changes every aspect of ur voice
not just pitch
it changes your VOICE.
how long is normal for the initial native client to start
takes like 5-8 mins, dw
Alright thank you!
is it realistic
Oh hell nah you just trolled my ass
Ayo? @elder willow level 1 !!! 
With a high quality voice model, yes
did you read the whole page?
im guessing you didn't
I learned the hard way
Oh okay now I see it, thank you!
theres a silver lining to -egirl
saying "yo just search for one" wont do shit unless we grab ur attention
uhm
Fair enough. I get what your saying. I was in Voice Models just before trying to find any girl voice changer but they sounded awful as I was adjusting the pitch etc. But ofcourse not all voices are pitch trained!
theres also different places to search for models
Any reccomendation for my similar searching?
read the green text
Right okay

Found 1 search on E-girl from 11 days ago lol
noone names a model "egirl"
Well I type the keyword E-girl & 1 result was found lol
i- thats not what i meant..
Why not?
I guess it can be a little more specific
Tf
-egirl
Suggestion for @novel haven
question my task manager shows 0 percent usage in gpu when its selected
Skill issues
wait
that's an actual command now?
lmaooo
yeah
I was rickrolled lol
Does anyone use a NVIDIA TESLA M40 for AI purpose? Any recommendations? Bout to grab it
i hope the link is a troll
It is
epic
I get what you mean. Just trolling around. You can find the egirl model in there
mom of mia
i once made an ip logger with turbowarp because i was bored
idk why, i just felt like saying that
yeah thats an interesting fact
i can't wait until the point that a company makes an entire operating system dedicated to a single game to prevent cheating
that's secretly just major malware
nah, it'd just be really funny to see
wait, what

ping me when u figure it out
i'm going to look for it
on what operating system?
all i can find is valorant, but i'm not referring to something like that is if that's what you're talking about
i mean a literal entire operating system that you have to install
even though it's somewhat close, because it is basically malware & runs at kernel level, it's just not an entire os
r u sure?
what operating system are you using?
theres a reason i ask
garuda?
come to think of it, windows also kinda falls under that definition as well
you're playing a game called "get your computer to actually work" while it makes it a living hell to do so
ever heard of Choc.kun?
choc.kun deez nuts? idk
fuck you
just read note 3# in my bio
i am concerned
fair
bc i can do what i want

even menh is subscribed
i once uploaded unity tutorials to a porn website because it was funny
idk if they're still there or not, but i also made a scratch tutorial lol
they accepted my verification request that required you to "show a video of yourself" or smth, because i kinda did
i showed the player model & a tmp element
this is what boredom does to a man
why does none of mine sound good or even slightly realistic? i have chunk set to 896 and extra on max
what gpu do you have
3080ti
hmmm i think its probably the voice model you're using
try chunk 96-112 and extra 16384
maybe im getting unlucky ive tried like 3 though
^^^
and increase/decrease tune depending on the voice u wanna match
why my extra only go to 131072
probably you didn't download the cuda version?
oooh that could be it maybe?
@rotund echo did you have a dropdown selection for ur gpu
ye either my gpu or CPU
Ayo? @rotund echo level 2 !!! 
ight so it's the right version
👍
Hi 
Does anyone have a spreadsheet of all huggingface models?
Ayo? @teal sinew level 2 !!! 
Helpers must be tired of answering the same questions over and over again.. What are the guide and help channels for?
Does any K-pop fan around here know where I can find Dreamcatcher's JiU ai voice model? I can't find it on any website; I've searched here in discord, in hugging face, Weights...
Ayo? @mellow kiln level 1 !!! 
Nah it doesn't make me tired. What tires me is people that refuse to understand certain things, thats all
You're happy with your job then
hi
Ayo? @spiral kite level 1 !!! 
is there any collab working with AI voice changer?
yeah, check w-okada pins
which channel bro?
OMG thank you so much bro
I think we both are alone. XD

Information for new members! 