#🧬│ai-chat
1 messages · Page 69 of 1
hai
Real arent anime women
No, you have to get Mangio RVC
how do i use ai voice models
ok ty
Using RVC, there’s multiple ways to use it
Are you on pc?
Hello, is it still possible to create model with collab?
Can’t sorry, I can do DMs
Yes
how do i use voice model i js downloaded
Also how much better is mangio crepe when compared to rvmpe?
-colab yep
Colab suggestions for @balmy yew
RVC
- AI-Cover-Gen-No-UI (inference only)
- RVC v2 Disconnected (training only)
🌐 Translated colabs
- PT-BR:
-colabs_br
Audio Separation/Isolation
tysm
qefefwf'
check dms
inference or training?
training
Ayo? @balmy yew level 1 !!! 
Nope
Hello
okay, thanks for the help 🫶
✋
selena

**Method 1 🔎: ** The search accuracy is 100% 💕
Upload the <t:1685905099>
Server uploaded: AI Hispano
Channel uploaded: AI Hispano Post
Profile Owner: aiverse
AIVERSE
aiverse
187
EN 🇺🇸
RVC
0 MB
null
1000
0
sa
as
saas
hi
data poisoning?
:0
How do I make a specific model say what I want?
What’s the program site to merge the zip and the wave files again. Brain fart. EIU or something like that.
Use the TTS option on Ilaria RVC
hi.
It's on #1159513888199540817
Ty Jules❤️❤️
What do you mean by tts? Sorry I'm really new to this stuff
TTS means Text To Speech
You type whatever you want and the model will say it
Ayo? @idle light level 28 !!! 
On a normal voice, no singing or anything
Oh, I mean how do I make a downloadable audio that I can send whenever I want?
Always in terms of making a character voice say what I need
Okay so
Give her time it a few steps. Thx Jules.
gençler farkettiniz mi türkiyeyi kaldırmışlar
Ayo? @midnight matrix level 1 !!! 
You use a vocaloid for singing tts anyways since it's been perfected for decades
To put it simply, there is this website a mod made to make AI covers or use the TTS function. It's called Ilaria RVC https://huggingface.co/spaces/TheStinger/Ilaria_RVC
Hi i'm Issa ❤️
If you want to make your own like rvc. Use diffsinger. Very similar open source
My bad I was trying to find the link
Hi daddy
Poopies. Hugs. Thx for da help other day
Make sure to log in while you use it
Embed failure!
Twist their nuts. Counterclockwise!
Why the Norway flag shirt
And what do i insert into change model?
Ayo? @dusty anvil level 1 !!! 
It's her favorite country. It's part of the lore
Ah no idea, I never used TTS. Ask @hollow tangle 😭
Elaborate please :3
If you want
It's all on her official website and wikis
hi hi
whats happening?
Yes
is still in development, should be deployed today/tomorrow

Oohh okay
You always react with the same emotes or stickers 😭
What it do I'm Producer Dat Boi Ricci and I'm addicted to AI
Can I ping you whenever I want?
Intelligent dance music
👋
wsp
АЛО РУССКИЕ ЕСТЬ
hey
hi
hi
uh
Yes
welcome to the underground
hello
Welcome to everyone
Cheese
Hola soy Gatito Loquito... y me gusta la musica
A question friends, once I download the voice model... can I install some program on my PC to be able to modify my voice with the voice model that I just downloaded?... I downloaded Ed Sheeran's voice and They want to replace my voice with Ed's voice. But I want to do it on my own PC without using online web pages. Sometimes I don't have internet at home and that's why I need to make it local. I will greatly appreciate your help.
The webpage that is given works offline since it’s hosted on your pc
did something happen?
oi @waxen grove
Ayo? @honest rover level 4 !!! 
oi
so basically now i know whats the problem
every time i do the cmd download with slow internet that has spikes
it gets corrupted
so uh i have kind of a side quest and a question to u
does ur w okada work?
yes
Can't upload
why not?
700kbps upload
ur internet bad?
yes
haiiii
someone uploaded a model and pretrain files and inserted em?
I've been thinking if someone could make me a Garreth Weasley RVC Harvest model for my tts experience
also I made a request for that
I'm seeing if someone is going to help me out
Is it just me or does it look like a lot of models are missing?
No, your just blacklisted for having a loli pfp
What?
whats the best colab rn
Is it for real? Sounds strange...
weights.gg has the missing models
Theres also a bot to search here
Too bad they're not well categorized. Having to know the name of the voice in advance is a bit...
Ayo? @elder willow level 1 !!! 
Hello! UCLA English/DMA prof teaching/working on algorithmic aesthetics. Glad to be here—
Hello, how can I combine two photos with AI, for example, how can I add a third person to a photo with two people?
hello searching foe valorant voicelines
I remember a local program similar to the one you use here, but in it I could change the emotion of the voice. Do you know what this software is?
To combine two photos and add a third person using AI, follow these steps:
-
Choose an AI image editing tool like Adobe Photoshop, GIMP, Remove.bg, or Pixlr.
-
Prepare the photos by ensuring good quality and resolution.
-
Open the chosen AI image editing tool and access the editing interface.
-
Import the two photos you want to combine.
-
Open the third photo with the person you want to add and isolate them from the background.
-
Copy and paste the selected person onto the combined photo.
-
Adjust and blend the person to make it look natural.
-
Perform final touches and enhancements as necessary.
-
Save the final combined photo in an appropriate file format.
do colabs take hugging links?
is there a way for rvc disconnected to auto save to my drive?
it just wont autosave there
is it possible to use a model in realtime while singing and it will sound really good ?
hi
how are you
E
whatt
Ayo? @limpid jolt level 4 !!! 
omg
very funny
okay
verry funny
but ridiclous
are you okey cadoiz
how are you
hi

Yo wsg
Hello
Does anyone have a tutorial on how to do a cover on your own PC? since the site I used no longer has a community voice.
Ayo? @rancid palm level 1 !!! 
What do I do once I downlaod an AI
Hello !
Is local not working for anyone else?
Oh, kits... that site sucked anyway
There are guides in #1159513888199540817, they should be easy to follow
ow, so brasileiro kkk
tem algum que vc recomenda eu instalar no pc?
se tu tiver uma GPU q rode
nao tenho uma poderosa mas melhor do que nada
é qual?
1060 3gb
acho q dá pra cover
ss
manda link no pv
Is there a Kanye West model?
Yes
Ayo? @forest quarry level 36 !!! 
ye it was taken down due to copyright
Some people did bad stuff intentionally to take us down...
ai hub is back tho
m
hey there! joined the server to learn more about AI voice changing. Trying to put something together for a costume and i have no idea what im doing
no
So far I got w-okada working on my laptop but its chugging pretty bad. It does have a 1650Ti onboard so I'm surprised its struggling so badly.
Ayo? @sick pewter level 1 !!! 
any tips on improving performance? I really need to get the latency down as low as i can get since reality doesn't give me any leniancy in syncing back up video and audio in post lol
how do i upload a model on google drive so i can use it
i mean
if ur gonna make vids or stuff, i recommend just using a fork
i ve got a question with dataset
nope, trying to use it for real time performance.
at like conventions and stuff on the floor
well a 1650 is technically low end for this type of thing so uh, yeah... i suggest trying out rvmpe + high chunks to keep it somewhat stable
can u check #✨│ai-help
Ahh i was wondering abou that. I'm totally uncalibrated for how much horsepower this kind of thing uses. the W-Okada gitgub was really vauge on the hardware side of things between "oh you can get decent performance on an older CPU" to people saying you need like a 4060
I guess it might be time for a new laptop then. Current one is half dead anyway.
Dunno how CPU effects audio files lol
i mean the git implied you can run the AI on the CPU alone
with decent results. Seems like you really need the power a GPU gives you to do this stuff well though.
honestly i think you can run but like if u want optimal performance then at least a 3060 should do
ktos ma głos spejsona panowie
any 2xxx geforce or newer does the job tbh
waa
hey can anyone help how to get the voice client to stop play back on my own voice
hi guys
hi
so, i made a lucas from ssbb ai model
if you didn't know his voice: he's from mother 3 and super smash bros brawl so i've decided to make lucas's voice model lol
i featured his voice model in my latest meme post: https://www.youtube.com/shorts/6J_FnsKWtMg
Was gonna reupload my models but i forgot i just joined and don't have perms 😭
Ayo? @silver sandal level 1 !!! 
oh neat
do /usermodels
I'm sorry, the user you mentioned does not have any models uploaded in my database.
also, how do i improve my first model, i don't have the model qx role nor the model maker 😭
Ayo? @elder willow level 1 !!! 
is the database from weights.gg?
Wait I'm actually mentally ill
hi
uhm guys can someone help wit the voice changer lagging when i launch a game
hello, can someone help with me
yes
i'm trying to approve my model and i don't have model maker nor model qc
oh

then who knows
can someone help me approve my model but i don't have model maker nor model qc
just get model maker
Ayo? @grim fog level 3 !!! 
how tho
it didn't gave me the model maker role tho when i posted lucas's ai model
where did you post it?
still at#1159514067187277865
oh
"Please wait until a Model QC reviews your model. You'll be notified once it has been reviewed."

how do I get my voice client to stop lagging in games
hola, alguno sabe como puedo utilizar la voz de will smith? o como la puedo buscar?
Hello, does anyone know how I can use Will Smith's voice? Or how can I look it up?
Questions guys, if I was to pay someone to make a custom model for me. How much would that cost? How many reference recordings do I need? The more recordings I have the better?
hi
Hi, I'm just getting into AI voice covers and would love to learn more about the whole process of creating to testing them!
Can anyone tell me if it's possible to do "AI Cover" on Android or a smartphone?
I'm having doubts about this "10 hour" thing. And I don't see this "resume training" cell there. I don't think I can do this without the GPU.
vlone
this wordband bot is stuck on thinking 😭
let him cook
it takes like 5 mins
Ayo? @finite nest level 1 !!! 
hey aint u the one who didnt use the deep face channel correctly
I actually wonder. In RVC v2 Disconnect, after loading the dataset, do I have to click training to resume? I also asked this in #✨│ai-help.
yeah bc some nerd told me to go to #1159290666019471421
but I didn’t know I had to use bot spam
I wasn’t rude WTF?
this is ur fault @covert lake but bye
lets see
Ayo? @sacred plinth level 6 !!! 
i just joined
my fault for telling u which channel to use?
at least i have the ability of being able to read a channel name before just doing some random commands LMAO
Hey
takes one to know one
whats the best juice wRLD model
The one that you think sounds the best
you just have to believe in it
butt
ok
its going to take a while butt fuck it
no thank you
bruh what is that pfp?
you can't say "butt fuck it" and have that pfp
and not make it weird
me as a little kid
that doesn't make it better
idc
general chat moment
idk
yep
“Stop the peace and keep the violence”
whats up dog
wow
epic embed fail
its a gif of the spiderman and lizard poster animated
but sure you can remove it if you don't like it
THANKS SO MUCH AI Hub is back ❤️
Admin team should prepare a backup or something like another place to preserve the community 🙏 #1159516963014451302
this is the backup
Oh ok I didn't know when I was on the first Discord
Ayo? @iron shell level 1 !!! 
Hows everyone
checked. doesnt look like you put much work into the video itself but the audio's aight.
I feel like Timbo lied to me. I can't resume training a voice...
heya
litsa, explain to me how does my model did not meet any quality standards?
Ayo? @elder willow level 2 !!! 
hi what channel should i go to if i wanna learn how to make ai covers? i just downloaded two ai voices already
And sadly, I don't have the pth files...
is it normal for rvc inference to have 0-1% gpu utilisation (on a 3090 btw)
I'm on rvc v2 disconnect, where I ran out of GPU, while I was training the voice.
the pth files?
This really gives off a “BLASPHEMY AGAISNT THE HOOOOOLYY SPIRIIIT” vibe
today is oct 30 for me
The ptsd files?
https://youtu.be/h9jc-N39S1E check out my new AI cover ACC on youtube if you want to
I'm last 31th for expert mode, next is November a Hardcore mode. VERY FAST EXTREMELY!! 😁😁
Wrong server, this ain’t terraria
just looing for ai models
Here's this ^^ https://www.weights.gg/
sighs
hello 👋🏽
What up!! 😁
It good for people likely for Halloween season!! 😁
Here's a bit of breaking news the lawsuit that was filed against deviantART, stabilityai and openai has been Dismissed by a judge https://twitter.com/technollama/status/1719086572161617986?t=vtUd_Gwg3t4WXG1gvDH5Bw&s=19
More details on the tweet
good, that lawsuit was dumb
*midjourney
Guys do i get 1 use every day or is it a 1 time only
for the rvc
hey guys does anyone know where I could find the best vocal isolator for songs? I need a song instrumental but none of the ones I found work the best
Is it the pth files?
**Method 1 🔎: ** The search accuracy is 100% 💕
Upload the <t:1687812272>
Server uploaded: AI Hispano
Channel uploaded: AI Hispano Post
Profile Owner: xitsnein
Nein
xitsnein
548
ES 🇪🇸
RVC V2
0 MB
Harvest
450
0
**Method 1 🔎: ** The search accuracy is 100% 💕
Upload the <t:1691508866>
Server uploaded: AI HUB Brazil
Channel uploaded: AI HUB Brazil Post
Profile Owner: chara0954
Chara
chara0954
3744
EN 🇺🇸
RVC V2
180 MB
N/A
200
0
use UVR
or mvsep
Thanks
Ayo? @steel remnant level 1 !!! 
hi
@midnight pine USE #1159290666019471421
oi
bruh wut
Guys do i get 1 use every day or is it a 1 time only
for the rvc
idk guys but im already ready for friday night vibes
in the meantime funkytown vibesa
is there any new ai creater
of what, colab?
cause mine doesnt seem working
just got doxxed but we up
funkytown plays in the background
I HATE TETO
YES
Ayo? @elder willow level 2 !!! 
new phone who dis?
no?
sooo its 1 time only
nope
then
No YT premiere today, but here's some Teto content while you wait for my new cover next week!
Illustration by ChanPo
→https://twitter.com/ChanPotei
Credits:
■Original song: We Like to Party by Vengaboys
→https://youtu.be/6Zbi0XmGtMw
■UST by me
dance vibes

hi I am calysta new member
Hi Hi
Hi, can I ask for help ?
hi, just downloaded okada vcc and how do i test it out? i click start and no voice
dataset is empty
I've installed svc locally, the gui works but I'm encountering PermissionError: [Errno 13] Permission denied
what should I do ;-;
is there additional setup for applio or is it just meant to work out of the box?
Hello
Did you download the latest version, iv'e had the same issue and it fixed itself when i got the latest version
Ayo? @ripe vapor level 1 !!! 
:0
Do you have a link ?
At work arm but go to the huggingface page for okada and download the one at the bottom
Thats the latest one
i cant hear the voice changer , does anyone know? is it something to do with the input/output
Ayo? @shrewd sentinel level 2 !!! 
Hi guys :DD
Tell me, how i can download RVC-GUI (like EasyGUI) for local (not cloud)? :DD
I use RVC-GUI from tiger now, and i wanna have some more options for voice models :DD
Add your headset as your monitor
Under input and output
Can you hear monitor?
its not detecting any volume
how about ping?
is it on buf? buf is 1020ms and res is 22ms
how do i check ping soz
i did it and still, does this happen to you when you first downloaded?
Okay, you use virtual cable? Try change output to your device and check
Idk why but this don't like your devices 😄 I had ping 50k on my first download
i dont use virtual cable but i have it
Hi everyone 👋🏼
Hello
Hello
Do you have the colab link of rvc v2 training?
Where you train a model
This should be able to help you
hi i'm roscall i was one of the original model makers
hello "roscall i was one of the original model makers" (dad joke)
i can't type in the voice models channel
strange
car jumpscare
u sure?
QUICK THROW AWAY YOUR PHONE IT'S A JUMP-sCARe
Restart ur discord or clear ur cache, it might help
Anyone know why my voice will go from lagging with the voice ai audio, to clear, but then once I switch back to my normal voice, and then switch back to the ai, itll go back to lagging
realtime?
yes
like, Ill have the audio going
and Ill be talking and at first its laggy
Hi everyone
and then after a few times of refreshing it it goes clear
but then I switch back to my normal voice on disc
and then switch back to the cable, and it lags again
I ment I can't post pictures in the voice model channel
Idk but u might wanna check #✨│announcements
yea use embeds now
hI!!
???
Ayo? @elder willow level 1 !!! 
just embed a gif or somethin in the original post
How long would it take to train a decent model on an M1 Pro Mac?
no
wait nvm it's paid lmfao💀
bruh
im stupid as hell
kmn
dw i hate this situation just as much as all of u but we can't have the server being taken down again
so we'll find solutions together

Anyone use RVC-GUI with mangle? 
Ayo? @gusty lynx level 3 !!! 
hi everyone
The system cannot find the path specified.
The system cannot find the path specified.
The system cannot find the path specified.
The system cannot find the path specified.
The system cannot find the path specified.
Why you turned into bot?
the system cannot find the path specified
Is it local? Sry, i'm noobie 😄
regjfwioewsgijk1ewojgtfbrewo432rflp5ew4kotiekewgt54yet
kiht 4t4l;eikodfkofldc;,xkferiofkdfje
tjiroghvdfjhnjeslfhuoethweiofjidjkhiofhjeifjdlfjsd;fjeo;rjewofredfddfdsfd
efds;kfdlfjmerklfjerlkfjerojgfrlklefjerklgfnrklgnrklgnrkegnreklgnlfkdlngkdfngkfngkfdngkfdngkrhkfdngkfdgnkfndgsdg'er
fdvgmkdfngkfdnggndgndfjgerjkht3w738wiejdsmkelsrj-239r03diosofjeferfrgvdfgfdghes535rfdewt
jiojeetkjfeijcidsfioesfjesoru32094892r90ewolsdjkl.ldpweir0-wur9037549refeasdxmsdhtuhyergfjrsdivjdfkgjds
fjeruhtu3u89jikdjkfjewiruewifjdjfiaejfopweirowdiosdwsdslfesofjerotodkfwesdfdtgsd
ifgjriutweioskdljdcmihfuesfvnsea3qwwertyghjvb nmnrswrafgfhdhasikxz
ay bro watch out u might get a time out
yes
hi
any good horror character ai voices? i tried looking for John Kramer but only got the italian dub version.
chao
how to make a dataset
Ayo? @fathom crane level 2 !!! 
get isolate audios
but dataset is empty when itry
how am i supposed to make ai models
becuz why is it empty i put the audio in there
Hello, what RVC do you use?
v2
I know, I wonder what link? I can not find it
anyone have Kafka model?
what do you wanna do
you nean kafka from ff7
Is Anyone Having Issues with uploading voice models?
what do you want to do
I not speak english,i use translate
Ayo? @severe ice level 1 !!! 
sadly, all of them are in the voided discord
Sorry
or check weights.gg
ahello
what's good blud
Well, a cover of Jax singing my favorite song, I think that's what it's for, right?
wsg u back
Yes
what do you wanna do

It doesn't work anymore, is there a way I can have my own RVC?
Tenés algún error?
Brave
Ah okay
Y no tengo sesion iniciada de mi huggin
Deberías. Sino te va a dar bastantes errores
Asegúrate de no salir de la pestaña o app mientras se renderiza y desactiva los escudos
Si sigue sin funcionar duplica el espacio. Para eso tocás el texto violeta que está al principio
Si sigue sin funcionar usá el que yo uso https://huggingface.co/spaces/r3gm/Aesthetic_RVC_Inference_HF
Suerte ^-^
Ok ok,muchas gracias :D.
Ayo? @severe ice level 2 !!! 
De nada ^-^ no tengas miedo de etiquetarme si pasa algo :3
gemme that where can i find it
#🔍│find-models or in the website weights.gg
What do you need help with?
hello julia, pls teach me how to make voice models, I've been using kits.ai but it's really limited, and I wanna learn RVC local, I followed some youtube videos but I fail
so sad😭
Oh I know nothing about making models, sorry. There are some guides on #1159513888199540817 and you can ask for help on #📑│making-models and #1159290066074607626
Good luck!!
okay thank you julia!!
Ayo? @jaunty heart level 3 !!! 
julia
i cant find that
Hi
👋
2 am

hi
hiii
helo
Hi guys whatz up
Hi
May I ask a few questions?
If we search for a model And how can I download the sounds?
I've tried it, but I don't quite understand it.
If anyone can help I would greatly appreciate it.
bro my ngrok token could not be found becuase it says this site can't be reached
am i scammed
where do i get the ai voices
It's my birthday =)_ _
'sub
hii im new here
Finally got my Locust Drone voice trained. Had to make it 220 Epochs instead of 250.
i just realized that I have been using the official version of rvc, do the forks really offer anything to warrant using them instead?
Hi
Mangio fork is infinitely better
yo hiiii
hi
hi
:0
cool gui + some cool features

ello
which version do i download? Is the training version for training only, or does it do what the non-training version does as well?
🙏
https://huggingface.co/MangioRVC/Mangio-RVC-Huggingface/tree/main this maybe, I may be wrong tho
https://rentry.co/VoiceChangerGuide this might help
This guide is written by: Raven
If you edit this credit yourself here :
This is for the AI Hub Server
W-Okada FAQ - has faster answers but its just general purposed to get it done with.
Anyone CAN DM me by clicking my name to be direct linked to my discord account but only if it's new information...
Both
I have a question. I visited here a few months ago and there were many different models on the voice model channel. But when I visited recently, the number decreased a lot What happened?
Server got wiped by discord in between
Supposedly because of copyright infringements
oi
question real quick
when u laugh in the voice changer does it sound trash or just me
trash
hi
Trash, it's usually because it's not really good at it or the model has no laughing sample
o i see
Is mangiro a real time voice changer
Ayo? @polar trellis level 1 !!! 
Nipe
Nope* w okada is real time voice changer
Not mangio
hi
hi
q DA FD
WAT
i
also the .15 one is
in the system should i set both as the cable? like input and output?
In the system ?
ya
Your sound output should remain your headset
like sound settings in windows
hi
And input, well that depends if you want everything to use the modified voice or not
so i put cable if i want it for everything right?
yes
and in discord i put input as cable and output as my headset riiight?
also thanks man for all the help
hi
hi
Doesn't including laughing in the dataset just make the model noisy or smth
hi
hi
Hey fam, when doing an AI cover, how do you get the vocals to match intrumentals pitch?
So it doesn't sound off key
hi
Hi
Hi
Who know Azahriah?
Ayo? @elder willow level 1 !!! 
hi
https://open.spotify.com/album/20vgwLm4Zo6tL7DKO05mfJ?si=Sd3BdyUVRbqkSbpYTX2Xcg
First ones to officially release an AI album on all platforms
hi
Hi
Hi can anyone make Rick Ross AI plz!!!
Does anyone have an Ethel cain model?
Hey guys, does anyone know if theres a voice that is good for taking medium-deep voices (guy) and is able to turn them into a kinda-beleivable girl voice?
is GIGABYTE GeForce RTX 3060 Ti Vision is alr?
In my case, it hasn't
Hi
Hi
Hi I Like Make The Content Of Ai Anime Singing Another Songs On Youtube
hihi
what if someone does an experiment where they do an ai cover of an ai cover, and then do it over and over again to see what happens to the vocals
Ayo? @wise trench level 3 !!! 
As long as you dont want to run Cities Skylines 2
A long time ago, when graphics cards were expensive, I was thinking of upgrading to 3060ti. However, I ended up waiting for graphics cards to become "cheap" and ended up getting a EVGA 3090 instead
Ayo? @unborn ermine level 17 !!! 
And ultimately, the only reason I upgraded from my RX580 8gb, was so I could do AI stuff
is there a working colab?
is EVGA 3090 that good?
Well EVGA no longer makes graphics cards. the 3090 ti / 3090 were the last production runs they had. I wish they still did, but I respect their business decision
A 3090 is a 3090 tbh. You probably dont need it, its just what I ended up making the jump to. a 3060 ti was what I was thinking of upgrading to when graphics cards were on hold lists and massively over msrp- a 3060ti was basically a better 1080ti- which is my frame of reference I can easily understand and compare
Hey fam, when doing an AI cover, how do you get the vocals to match intrumentals pitch?
So it doesn't sound off key
Omg
Wtf, 90 hours?
Who are you training
Just make all the mp4 files one
Then search up mp4 to mp3
Oh yeah definitely do that
You don't even need that much
can't you just put mp4 directly into uvr5 then clip the output .wav in audacity?
10-15 minutes is more than enough
Go to convert io
I think they have 100 gb
Remove the silence first
3090ti, it will still be a pretty powerful graphics card at least, compared to my laptop's crappy graphics card 
Or you can also ask chat gpt for that
Yuh
Ayo? @elder willow level 1 !!! 
You can go with 1 hour of dataset
Just keep a check at tensorboard
The tensorboard will tell you when it is over trianing
no fuckin' way
The only game I cant run well is Cities Skylines 2- but people can barely run it on a 4090... So thats more of a problem with the game's optimization lol. I still enjoy CS:2 a lot tho.
I dont see me having to upgrade in awhile. I have a 3090 and 13700k. I would assume a lot of 3090s are even cheaper now then what I paid
But for 1 hr of dataset, i think you would need 100 epochs or something
I have never touched local
Youll probably be happy with a 3060ti
So I don't know, I can only help you with colab
3060 ti is a W
Where the spooky scary skeletons brohams
Yea def run local if you got a 3090. You can run anything local
As long as you got the storage
My stable diffusion folder is 282 GB
Both, if you choose to convert to mp3 you can export it to wav by using audacity
You have to convert it to wav anyway
this is cray
i just check the game cities skylines. it sounds like my laptop. burst into flames any second if i push it too hard
i can save up enough money to buy a pc with a decent graphics card. i'm still using my laptop's Ndvia GTX 1050
Ayo? @elder willow level 4 !!! 
Training 90h

Havent messed with the voice cloning much recently. But I believe I was using Mango fork (the one by kalamaze) and MMVCServerSIO yes
Not even joking your pc will explode and will be homeless once you get the electricity bill lol
If any of you make an ai cover with any of my model, please @/ping me.
You should prob search on Google or ask chat gpt
Search what?
j
https://youtu.be/RTMptDxXlJ8?si=4F2LRfPw43ySY82r This AI Voice is very Powerful
Epic embed failure
It's a pity that the voice didn't make it public, it was like Dimash sang but with his vocal style
Wait what happened to the old ai hub discord
I accidentally deleted it srry
What role
Nah bruh you are really evil for that
We should cancel you
You are a furry
I had a level role I could access a separate general chat of
Oh for that you gotta gain lvl 5
Again? 😔
Yea😈
Naw just run !verify
Big brain
chandler bing voice model?
I'm still chugging on with a 1650
B
1650, 1660 are so mid
Eh
With 2023 and ai to the context
Can u train
Oh yeah obviously you can run games
Rvc disconnected works fine so I'm sticking with it
Even a 1050 would do it
That seems about right
It's not like I don't wanna upgrade
What will you upgrade it to
What Best Pitch Extraction Algorithm For high pitch Song?
Any upgrade's outta the question rn so I haven't considered
Rmvpe overall
Idk rvmpe or mangio-crepe
Get ya money up
I've heard rvmpe can be noisy or raspy
it has it built in

o/
Haiii >.< Hiii ^-^
hi
Hai
hi
sup gamers
hi
hey people
hey
hi
hi
hı everybody
hi
Hi
im sort of getting tried of anime and cartoons models. as well as kpop models here. idk why.
oh
Ayo? @lucid bison level 5 !!! 
the wordband bot doesnt work 👹
Ok
why ain’t it working?
is there a covergen where there's no mdx 😭
I've asked many times for models, but people seem to ignore other more interesting requests. not only mine, but other people's as well. In exchange, just yet another member of some k-pop groups gets almost a hundred voice models, or someone from an anime / cartoon.
I have nothing against it, but it's getting tiresome, to be honest.
I'd make my own, but my computer's really weak and old for this shit.
yeah this usually happens with ppl ignoring 😭 sometimes i give up instead
check help
I feel like a lot of people only know the idols\anime and cartoon characters here, and some actors. as an exemple, William Hurt or Jeff Goldblum, it's like they're non-existant to most of peeps here. :\
wordband bot isn’t working! why not
Hi
wordband isn’t working
apparently I’m rude according to @analog crypt tter
they don’t know what they are talking about
Hi people
Hi I know this is random but I love your bio 😭
How's the adobe animate game going?
lmaooo
hi
Using characters is parody. But when you start impersonating real people is where things get ethical
oops
I use voices I have been personally asked to use
ive only trained real people if their yt videos count
Oh my god I don't know if it's okay for her to say the n word 😭
Asterian gave her the pass
Edit: omg, idk what to say, I'm absolutely flooored right now at seeing the amount of views and likes this got in 5 days. I'm honestly crying out of happiness right now. 10k views and almost 2k likes...holy moly. just....Thank you all so much for the support!
AAAAAAAAAAH THE FRENCH BREAD CHIMERA ARRIVED ON SYNTH V AAAAAAAAAAAAAAAAAAH!!!!!!!
*...
Give me a base file
@wintry sinew embed failure!
:[
I don't get the hype about flower personally...
Kafu can be good... but is really limited in what she can do
I love that gif 😭
Um excuse me?
Ayo? @elder willow level 12 !!! 
Same 
Painful but fun 😎
day 1
She sounds fine but people force her to do things that don't fit her style
What do you mean? She is perfect in every song
Nothing is perfect
Who's the criatura on your pfp? It looks familiar
Clock from an object show called "it's time for the"
Ayo? @ancient ferry level 1 !!! 
Her voiceback is to powerful
She have to much energy
does anyone know why the server got reseted? i mean the previous voice models are not here anymore
Copyright
yeah the previous server died
day 1
what is weights.gg
Ayo? @wide burrow level 1 !!! 
a
bella
people posted full episodes of TV shows then reported the server for copyright
tfw weights.gg just sends a weights.gg link that doesnt work as a model
from what I heard the day it happened
Download the file
hi
major hassle tho
it used to just be a huggingface link
I've never used a colab or cloud lol
When i try to make a song, the vocals sound weird like this, anyone have any suggestions?
i cant add a file, vocals just sound really weird
meow
Hi tetski
Eighty hours? Professional ai voices have like 2 to 4 hours
Quality over quantity
Cut the silence. Focus on range, variety, and the best quality
Get as many unique phonemes as possible so it will work better in all languages
Someone had made a script to do just that
That's why it costs a minimum of $50,000 to make a professional one
And up to a year
Why 😦
Buying the rights of a song?
What? No a voice provider spends weeks in a studio doing multiple takes and they cut it down into just the best pieces then carefully train each vocal mode ai model
original:《Vltava》Bedřich Smetana
tuning:Creuzer
Asterian has 8 vocal modes, they had to modify their engine to handle his voice
Cut all the silence first, I'll try to find the script later. Then you're going to want to take the best bits
My best teto model is about an hour of data
Her synthV has like 2 hours of data
I've seen people make a model on as little as 2 minutes
You will want it in uncompressed wav
You technically can but just make it right the first time
Take some tips from the Utau community on making voicebanks
That has expanded into "diffsinger" which is open-source ai
What type of singing is it
Ah, so limited range?
20 minutes and 200 epochs is about the minimum for a good model
I've only done two so a model master can give additional tips
@north bluff
can i post your knife model on fakeyou
and say its not mine and credit you?
🤨
When I train in the "RVC v2 Disconnected", where do I find my model?
hi
You have to click on the Export model button after training
Then it should be localized in your Google Drive, inside your model's folder in "rvcDisconnected".
Ohh, I didnt see that 😅 My bad, thanks!
Np! Glad to have helped :)
what is the best voice model?
do you set the amount of training its gonna do
does anyone know?
or do you have to do it manually
¯_(ツ)_/¯
You set the epoch count
And the batch_size too
So you set how long the training is gonna be and how accurate to dataset you want it to be in the end.
batch_size should be 8 or 16
but id recommend trying out other values
and total epochs, for a first session, maybe 500?
Welcome to the server
Thanks, pal
Lol I read that as say it's mine and not credit you
...

