#🧬│ai-chat
1 messages · Page 64 of 1
• (Local) W-Okada Voice Changer
This one has more options and is somewhat more convenient to use than Real-time RVC.
Credits: ravencutie21
• (Online) W-Okada - Colab Version
A free, online version of W-Okada that allows you to run code on Google’s powerful GPUs.
Credits: Antasma245
• (Local) Real-time conversion of RVC
More accurate than W-Okada's, but requires a Nvidia RTX card.
Credits: teas.hi
• Russian guide for voice changing
Credits: redoverflow, blackcat76it and tntmaxim
• Text to speech:
Note: If you need further help setting up/using the voice changers, you can ask for help in #🔍│help-w-okada or check the channel's pinned comments.
If you're dumb fix yourself dumbass
Hahahahah embed perms
Thank you! I'll take a look right now
Ayo? @gentle sparrow level 1 !!! 
Chill bro
You're getting a timeout for that

hi . does any one know how to make my own voice model
Creating a model with Colab RVC Disconnected 1. Preparing the dataset: place your dataset files in a folder named after the model. zip the model's folder. 2. Connecting to Google Colab: In your Colab notebook: Run the "Dependencies" section. Connect your Google Drive. Navigate to "Drive" -> "My ...
i tried the colab its too hard for me any easier way
Colab is literally the easiest way
@winged tangle you want to make your own voice model!
yep
no just a famous local
Then do it
i cant its to complicated for me
Ayo? @winged tangle level 1 !!! 
Your best bet would be #1159289738314919936 a model then
ok
Epic embed failure
combustsy
So I realized wokada was the voice changer I had, but I redownloaded it from the link but it still wouldnt change my voice, it just repeats back my own voice
You settings are fucked up then. Did you enable "passthrough" by any chance?
yes I did
But did you like set a the voice model
And did you set your audio input and output settings correctly
Yeah, passthrough shouldn't be used
Only allows you to test your microphone
That would be your issue, It's passing through your voice ignoring the changer
where can i find the tutorial
I learnt through a tutorial on yt by some AI channel
-rvc
Hey, @south willow!
👇 Here are some useful resources to help you learn how to make ai covers
Translation by country
wrong one
can u help me
hi
Whats your issue
im stuck in downloading mod files
Nvm yall, pass through was the problem, for some reason, I thought it was like a way to teste out voices to myself
its taking like 2 hours
ah
issue solved
I'm not sure what I could do to resolve that, maybe try restarting the process?
Also is it not a wifi speed issue?
Mine took a couple minutes
Oh
ill send u ss
Ayo? @south willow level 1 !!! 
In my experience the start_https was taking a while so i just alt-f4'd and started the start_http bat file
Ayo? @vagrant terrace level 1 !!! 
That's normal for like a lot of open source software
Yes
nooooooooo
Search asmr in #1159289700490694666
Oops
If you cancelled the first time download you most likely fucked it up
Oh no mine works perfectly
But then again i used start_http for the first time startup and let it run
Poland Polska?
You might have only borked some sample models then I guess
Interesting, I only tested with the first sample before moving on to downloading a model, so no loss to me
why my voice changer is so slow
What graphics card are you using?
It could be a variety of things
i dont know
Oh ok
how do i check
On the right hand side of the screen there is Chunk, extra and GPU
The dropdown next to GPU should have your installed graphics card
im using gpu1
huh, doesn't say anything else?
If you click on the drop down is there more items?
Ayo? @vagrant terrace level 2 !!! 
it doesnt say anything
Anyway, I was going to ask if anyone had tips to reducing the cut-off?
Are you on a laptop or pc?
im on laptop
Ah ok, this software uses graphics cards to interpret the incoming speech and outgoing speech.
It sounds like you are using an integrated graphics card in which case it's probably running pretty hard to keep up. You might be hardware limited.
I have a game open with medium graphics settings and the voice software open and my gpu is at 87 degrees (I am using a GTX 1070ti in a laptop)
How can i make my own ai voice?
-rvc
RVC Guides (How to Make AI Cover)
Translation by country
i want to make my friends sing a song using ai
I think thats it? Only been here a short time.
Get audio recordings of your friend
im using rtx 3050
ima reinstall it ig
Ah ok, then it's most likely your settings. I don't think you will run into insane hardware bottlenecks
What causes delays is the two settings above GPU, play around with them until you find somewhere that works. The shorter the delay the worse it will sound, so find a nice balance.
Ayo? @glad panther level 1 !!! 
18 - he/him
Is there Zelensky’s voice here?
where is price
Isn't there the model for the kpop girl group twice?
What do you mean
Hi, I'm pretty sure that there is a Rex Orange County model that cover Japanese Denim in this server (or other server). Is there anyone that still have that cover? Thank you
You can make it urself
I mean, I just want to listen to that cover, because it was perfect
i'm not good enough to train a model that has such quality
What was the model name? @ivory shore
I don't really remember, but it was Rex Orange County that cover Japanese Denim
Just the name i need
It used to be in YouTube but it got deleted somehow
Rex orange county?
yup
https://www.youtube.com/watch?v=uNrJLQLs0dA I found this but this is speedup, kinda suck
yeah, I think I saw it in this server (or maybe other servers)
alright, let me see thank you
All good
https://www.weights.gg/models/clm72w3wk0xc2cctcpot7xzhd
I think this is it, but the audio was removed due to copyright claim
Ayo? @ivory shore level 1 !!! 

Congratulations you found it
Now use it and make a cover
@ivory shore btw don't eever post any audio here
What can I do to share a voice model?
@boreal stag go to #1159514067187277865 and do /submit
i used to be in this server already why it kicker me
I used to be a model maker, but for some reason I was banned from the server for no reason.
where can i get the collab link?
I'm glad my models weren't deleted.
Hello
Guys I've a question. So I tried couple of rvc models
Working so good realistic in RTX 4060
It works on social apps like telegram , discord WhatsApp
What about actual phone call? How do we do that
server was banned for copyright
u can try /usermodels to get your role back
The rat incident of October
I'm sorry, the user you specified does not exist or is a BOT.
price from mw2
Ayo? @humble idol level 3 !!! 
Upload the <t:1691784574>
Server uploaded: AI HUB (RIP)
Channel uploaded: AI HUB (RIP) Post
Profile Owner: memerowo
私はMemerOwOです
memerowo
4148
JP 🇯🇵
RVC V2
70 MB
N/A
1000
0
thank you 🙂
Ayo? @boreal stag level 1 !!! 
@boreal stag got damn you got a alot of cows
voices*
Even more than me, that's crazy bruh
I'm sorry, the user you mentioned does not have any models uploaded in my database.
😭
thank you @mortal jackal
price from mw2 would be good
Try to find them #🔍│find-models
i tried
no results only found soap
what
@wintry sinew
wanna check something
okie
Heey
i mean if embeds get removed again in verified u bet i am getting it back
wsg, what you doing here
Checking what I can do with these voice-models.
you can make covers etc
use it in realtime
How can i make a ai cover song of my friends singing it? (i have audio files of them speaking)
is text to speech good? (quality wise.)
make a model
I cant figure out what I'm supposed to do or intsall to use the ai voices. And a link to the Guides section is unhelpful, I still can't figure out where to start. I can't even find the RVC download link. I have been at this for an hour straight.
Link
yeah preety good
pls
wich one of them
what's good one
wich one of the links do i use
Ayo? @storm elk level 1 !!! 
HI anybody have cs go CT ai voice?
how do you use a rvc v2 model
@mortal jackal
Hi everybody
hi
how can i make models of my friends and make them sing a song
aint that the guy from fortnite
E
🧐
hugging face didn't work 
Tell your friend to record voice for about 5-10 minutes, then upload it to the dataset and train
lolbro getting wiped out for meme
i have the audio record alredy, where do i upload it
Go to Drive, then create a dataset folder and upload the voice to be trained
can you come to dms
Oh shit . it is my boi Adom Hs
can you link me the drive
took less than 20 minutes
You use a phone or a computer
stonks
?what does that even mean
fortnite
How about you back it up with a source
after i have voice record in a folder there how do i train it
Best f0?
rmvpe
kk
You can try RVC V2 Disconnected
okay
Ayo? @storm elk level 2 !!! 
Hello mates. Happy to be here
hi
Hi
Oh, I forgot to tell you, if you want to use RVC V2 Disconnected, you need to create a zip file and it must have the voice of the person you need to train and then upload it to the dataset.
okay
How long do we have to wait to get the model maker role?
Hello, but this guide: https://docs.aihub.wtf/other-guides/text-to-speech-rvc-v2 what program does it refer to?
what do i type in the dataset
Just upload zip files
Hi
yo
answer me pls
rvc v2+
any link?
best lip sync for deepfakes
Hi
HI

can som1 help me
Hi
Hello there
@elder willow interesting
Ayo? @lost harness level 1 !!! 
Hi
Ayo? @steel onyx level 1 !!! 
@quartz roost https://www.youtube.com/watch?v=xtPzddOoUKI
「自分色を紡いで、未来を描くんだ」
曲名は「みくいろパレット」と読みます!
第12回プロセカNEXT応募楽曲
Music/Lyrics:そともぎ
https://twitter.com/Soto_mogi
Vo:初音ミク
Gt:秋葉敬治 (https://twitter.com/kossy42 )
Ba:そともぎ
Pf:そともぎ
Dr:デクノボウ
Mix/Mastering:はるお (https://twitter.com/haluo_haluo )
illustration:水豹 (https://twitter.com/azarasi_haru )
Movie:瀬戸わらび (https://twitter.com/yk__rm )
dude what the actual fuck i saw an ad of this song
and the moment i heard the pre chorus, i 100% knew its gonna be good
why is this not on spotify holy shit this has catchiness level of greenlight serenade

Ayo? @radiant dagger level 1 !!! 
@hollow tangle hello good sir can i have imag perm
done
Hello
hi
Hi
thx guys
hmm 
Hi
someone plEASE FIND SPOTIFY OF THIS SONG
THIS IS WAY TOO UNDERRATED
Saved for later
this is exactly like when i discovered MIMI
at the time they got only 100k subs, heard shijima ni saku and kuninaru
really liked it, now they got over 1mil now
hope this artist really gets popular because they deserve more than 6k subs
hey watch the El Clasico?
yo hi
Who wins I think Barcelona is win
AI Moment

hi
Hi
Hi

turk
**Method 1 🔎: ** The search accuracy is 100% 💕
Upload the <t:1691779454>
Server uploaded: AI HUB (RIP)
Channel uploaded: AI HUB (RIP) Post
Profile Owner: sztef
Sztef
sztef
4129
EN 🇺🇸
N/A
242 MB
RVMPE
400
0
OH MAY GAHH!!!
Ayo? @mild saffron level 2 !!! 
so, if using Voice Changer Client Demo, on F0 Det, use crepe right?
Crepe is slower
🥞
But crepe is effective.
Crepe and Dio?
does replay not work with amd gpu?
Hello could someone please help me
when i try it, my voice not lagging while i'm playing a game
Ayo? @vale lion level 4 !!! 
How can I make a sample sound clean in Easy voice changer?!
Oh. Voice changer wise?
For real times it’s slow for slower computers
why did so many voices dissapeared?????
When made locally?
Server went bye bye
againn?
Someone spammed full tv episodes? Mods didn’t delete em server went bye bye
Happened about a month ago
goshhh such damn idiots
hi im noob
Voices can be found using the search bot or weights.gg
alr thx
There’s some in #1159289700490694666 too
i checked that and there arent as many that is why i asked
yo btw
?
when i used to change voice with celebrities i only used the link .zip how can i use the download?
ik what i am saying?
What method do you use?
hi
rvc 2
danggg
Mangio Local?
Which one?
Confirm that the voice you are going to use is consensual first and you will never have any problems
Because most don’t work anymore
can i send the link?
Sure
Doesn’t work anymore
whattt? why?
There is a few things you can do
like?
I'm a professional unconsensional voice changer user
@thin topaz easyGUI on their profile
Or you can use HuggingFace Spaces
Or you can use WordBand
yeee hugging face links
I’m talking about HuggingFace spaces
spaces?
Which is like Google colab but on HuggingFace
Guys i have a question. After downloading the .zip file of the ai what do i do?
A newbie here
https://huggingface.co/spaces/TheStinger/Ilaria_RVC I use this one
You CANNOT train using HuggingFace Spaces
oh
You need RVC to open it
Are you planning to do real time or covers?
i dunno ig both
There is still a colab for training tho
which one?
-colab
Google Colabs
RVC
- AI-Cover-Gen-No-UI (inference only)
- RVC v2 Disconnected (training only)
🌐 Translated colabs
- PT-BR:
-colabs_br
Audio Separation/Isolation
hey whats a good female rvc for a guy with a deepish voice?
Where can i download it from?
Ayo? @rapid root level 1 !!! 
Is your pc good?
Not really
GPU?
Anime girl voice fr
And are you going to do real time or covers?
i acc tryna do it too
Intel HD graphics 630 :v
good luck
Your out of luck
Use colab or HuggingFace spaces
Why??
Or wordband
lmaoo
recommend some to me my g
Intel Integrated Graphics
yoyo so how can i import the voice on huggingface ilaria rvc?
How do i do that?
Download model
after i have trained once, what do i do then?
Idk i never used one
so hard to find a good convincing girl model tho tbh
i have it downloaded
You can’t use real time or training with HuggingFace, https://huggingface.co/spaces/TheStinger/Ilaria_RVC
what yo tryna do boyyy
Don’t download for spaces
You put in the link in download model
ive trained voice ai in RVC Disconnected what do i do then
wanting to use it in cs2
Should end in .zip
https://youtu.be/7NCD2CAXwmA?si=v426LmbpBtBxLVfv I've fooled so many with teto already
Feel free to support me on Ko-fi: https://ko-fi.com/adamedmond
Uploading this first because DEATH cover got blocked many times, probably because of the music video thing so I'm doing the plan B video right now 👍
SVP distributed as well if anyone would like to try it out with other voicebank! Feel free to use it and don't forget to give credit w...
wdymm? wanna hop on a voice channel?
yes i have tHAT
Is it a link or a file
alr wish u luck
zip file
Should be a download link like this https://huggingface.co/TIMBOVILL/TIMBOVILL/resolve/main/Mysticat.zip
What model are you using?
that sounds pretty good, lots of effort went into it
Ayo? @elder willow level 1 !!! 
It should be the download link for the model
dani mocanu.zip
Ayo? @glass dust level 4 !!! 
Grindset #76: be a buff black man irl, and a soft anime girl online (profit)
i downloaded it from weights.gg
Ok
thanks! planning on upgrading my pc soon and im wondering if amd cards are worse when it comes to latency&&quality at the same chunk
ok lemme check
Then it will copy the link
okk wait a sec
ok im doin it rn
This is probally my last cover made on synthV 1.9, since I finished it like, TWO MONTHS AGO, and never posted, but well, here it is. I love natalie sm.
MUSIC: Marilyn McCoo
NATALIE DESIGN: Vimalion
ART: silmist
SVP/TUNING/MIXING: KAISKE
I will probally upload the SVP later.
That's just how good teto is
wassup folks hope y'all are doing great (:
teto as in the character?
bro i dunno why it says error
google shows me an image of a vocaloid char and it honestly reminds me of this https://www.youtube.com/watch?v=kuNixp-wvWM
but like when i copy the link it says .rar instead of .zip
what video is tat
The prefix for AI HUB 🎃 is /
lmao
help me
She has the best voice of all the vocaloids
what does 600 echos ?
Wicked I’m a ♥️♣️♦️
CREDITS
Music/Lyrics: ASTEROID (KIRA, monii, Johnny R)
Illust: Isamu
Movie: HISSYA / OKIKU
Special Thanks: Sena
SPOTIFY: https://open.spotify.com/album/65LLOqvUeS8QZC54G9vlqQ?si=tW5KOEjaRT2SGQVw8mHaRw
APPLE MUSIC: https://music.apple.com/de/album/wildcard-ep/1713029...
An epoch is a training cycle, basically it means how much the ai looks over the dataset, so in this case the ai looked over the dataset 600 times
MORE DOES NOT MEAN BETTER AND LESS DOES NOT MEAN BETTER
Anyone know a good way to make laughing sound normal?
Is your pc strong?
yes
GPU?
hello guys i hvaing out of memory error on my rvc cmd what should i do
Quality over quantity
timbovill can yo help me?
Get a GPU with more vRAM
the voice changer wont fuck up my pc right ?
i7-7700 and geforce gtx 1080
Looks like the person zipped it as a rar file instead of a zip
so what can i do?
You probably won’t be able to train but you can still do covers
its enough
You can use RVC Disconnected for training
can you help me in dms
isnt there any optimised way to get results?
Not for training
Yeah but I don’t really know
ГДЕ ССЫЛКА НА СКАЧИВАНИЕ
You can ask in #✨│ai-help
so what can i do if i have the .rar version?
Reupload it on your account as a zip
On HuggingFace then copy that link
Dunno why they uploaded it as a rar tho
oh ok
and how tf can i reupload it on my acc? :))))))))))))))
yo timbovill can you help me in dm?
@storm meadow
how can i listen to myself while using the voice changer
Ayo? @mint scroll level 1 !!! 
"Listen to this device" in audio settings
i just made my first ai cover on pc and man the difference between having to use sites on mobile and actual software on pc is night and day
only thing is it takes much longer on pc
But it’s more worth it
might be cause im using an integrated graphics chip lol
That’s why
ryzen 5600g, its all i got money for
doing almost anything on a mobile is awful, even if i can unfold mine into a bigger screen lmao
what i used on mobile for isolating vocals were minus-x which wasn't bad but uvr is just so much better
Ayo? @stray crescent level 1 !!! 
what websites work good for 1-2 minute long covers
oh use the colab
it crashes with 1-2 minute for rvmpe
Ayo? @subtle glade level 1 !!! 
kk
Wordband or HuggingFace spaces
huggingface spaces crash for longer rvmpe ones
e
Doesn’t for me
which one u use
1-2 minutes is short not longer so it should work
is there any way to fix the thing that happens on google colabs when u convert an audio and then it gets detected by google and u have to connect to a new runtime again
Or wordband
Ai-Cover-Gen?
ya this one is the one that breaks
Logged in?
You can’t unfocus the tab on spaces
That’s why
i dont need to duplicate the space right?
Shouldn’t need to
kk ill test out
yo timbovill check the dms
Yo
Glad someone made a Japanese cartman ai voice model, now all wee need is some ai Japanese voices for other South park characters like Stan, Kyle, Kenny, Butters, Wendy, Craig, Tweek, and Randy.
what site should i use for making ai covers?
sorry
Ayo? @elder willow level 1 !!! 
Hi, nice to meet you, my name is Izaya
You're in the queue! Your cover will be ready soon. Number of people in front of you: 0 (Status: 🟢)
@sour night, hello, boy
wha
Ayo? @zinc lion level 2 !!! 
Ilaria
For what
hello
yes

BROTHER NOAH
hi
Noah?
no
proto-rap?
boy tag
Like as a role?
Then I don’t know what you mean as ‘tag’
femboy tag

um, can someone tell me how to get the bfdi two voice model?
hi
weights.gg or search bot
eh?
sup
what?

‘Can I eat this one too?’💀
Who you think you are?
Fucks like, "Do you know who I am?"
Fucks fail to understand
I'm like, "Eh"
Who you think you are?
Fucks like, "Do you know who I am?"
Fucks fail to understand
I'm like, "Eh"
weights.gg or search bot
I love that song
what does search feature ratio do?
@hidden grotto search
I'm femboy
I have no idea how to use this

ok

The higher the value is, the more the sound of the model it will use. So for example, if the model was built using a dataset where there was background noise, your converted audio will have that background noise
whats the best girl voice
ah ok, thank you for explaining this to me
Basically the more the value the more it will sound like the model you are using. Set it lower if the audio glitches and stuff
You are welcome ^-^
where to downlad
💀 too lazt to answer

i cant find latina
Ayo? @south willow level 2 !!! 

Why do you need e girl voice?
Are you serious right now @wintry sinew
i want to troll my friends
How?
Hey
Ayo? @brazen sorrel level 1 !!! 
No one's ever seen me feel shit but eh
Lil' bits of eh gleaming, like Piss Christ
Cover me like skintight
Why my rep is like why they might be right
I'm kinda sketch, I might
Shoot a glance at the desperate like
Then I forget shit like Death Grips, like eh
I wave them off, I wave them in
Gotta flake, I fuck 'em off 'cause I ain't them
I'm like, "Eh"
…
Just look at weights.gg or #1159289700490694666 until you see a Latina girl
life is a highway
Guys I have a question. Is there any English dubbed anime voices available?
alr ty
There surely is. Search for the character you want

Ayo? @tacit osprey level 1 !!! 
On #🔍│find-models or the website weights.gg
the biggest bear
are there any goku voice models?
I dunno
Yeah, a few of them
Search on #🔍│find-models or on the website weights.gg
Click on the magnifying glass and type in whatever you are looking for
Not using voice.ai 
That’s the best setting
And on the website type on the search bar what you want
I love kits.ai 😍😍😍 I love using an ultra buggy, limited and paywalled app 😍😍😍
I even had a pa
myan
Ayo? @indigo epoch level 2 !!! 
paaaaaaaaaaaaaaa
I'm boutta BLOW
simply get a PC with a 3060
simple and easy as
😀

AMBATUBLOU
Guh...AMBASSING
AMBATUKAAAAAAAAAAAAAAAAM
😀
Dawg I'd shoot the plug the milisecond that mf showed up in one of those
https://youtu.be/8DTysiESVYI?si=QtloaMf6PEqBOTrC NOTTI OSAMA AI X SKI2CLICKIN - 41K
Ayo? @green adder level 1 !!! 
(Doesn't have a plug)
imagine wearing a VLONE 🤣 😭😭
we staying in the hood with this one
this shit sound ASS

chill
omw to buy a vlone shirt for 1ski
dawg I need the F16's flying over the city center to either bomb something or don't
swear to god these asshats have been here for 4 DAYS and haven't done shit but fly around like dickheads
dammit all my tabs shown
pizza anxiety
🍕
I use ezgui colab by rejekts
is there a way i can get to ai devotees faster cause server shut down a while ago
be active
ok
also guys how the hell do i separate backing from lead vocals ._.
is it 6 karaoke in the vr arch
jesus
I'm looking 4 2pac's voice model
Ayo? @white badge level 1 !!! 
should i remove echo and reverb first or separate lead and backing
which is better to do first
just try both
First remove reverb, then backing vocals
To get the best results
each process takes like 3 mins on my integrated card smh
oh nice, thanks!
not to mention the colab needing a new runtime every time i convert an audio
how do i fix that bruv
yeah im on ryzen 5600g
ohhh makes sense
no, uvr
i thought u meant like intel uhd
cause thats what my laptop had (not that i use my laptop but i just didnt know u had a 5600g)
separating vocals from instrumental - 1 to 2 mins
removing reverb and echo - 2 to 3 mins
separating backing from lead vocals - 2-3 mins
mod
Ayo? @zinc lion level 3 !!! 
thats pretty fast
but i also have to convert the backing vocals and lead vocals in the colab
i say fast because its faster than me converting a pth voice model to onnx
tell me
for some reason installing the colab takes 2-3 mins each time when it should only take 30-45secs
Ayo? @stray crescent level 2 !!! 
papa
Mine take 40 minutes 😦
and after converting an audio, if i find out i have to change the pitch cause it doesn't sound quite right for the voice model, i have to install all over again cause google disconnects me ._.
Even more
oof whats ur cpu
Uhh idk
if someone knows how to fix tell me pls
Radeon hd 7480?
maybe ur ram has an impact on it too, i have 16gb
same
Ayo? @indigo epoch level 3 !!! 
cpu, not gpu
Dang
damn
oh shoot are u using gpu acceleration in uvr @idle light
I'm screwed
troll
I don't even know what that is 😭
u're troll
dont use gpu acceleration, very very bad idea for integrated cards
Opening YouTube takes half a minute
its the first checkbox
liar
Honestly it's better
bro i just open youtube in chrome xd
I mean I can record a video if you want me to
wait for real? its faster?
my 6800hs infers 4-5x faster on integrated graphics
trust me i have a laptop that takes 6 minutes to boot up
ok, let's
they arent lying
blud has a better chip than me
i have 5600g aint no way its gonna be able to handle gpu acceleration
why so?
So then, should I turn on gpu acceleration or not?
Still, the CPU is really beefy, I wasn't expecting integrated graphics to be faster and yet
it is
ill give it a shot tomorrow
Yeah I'm also curious
i have an r5 1600 so i rely on gpu acceleration
lol imagine if i try to run uvr on my phone through wine
Why through wine ?
snapdragon 662 moment
You could run it natively
its the only windows emu my phone can handle
bro sd662 cannot handle windows 10
why would you need windows
what
pytorch for linux/mobile is a thing
i aint tryna use that
you can use the uvr weights without the uvr app
rather use uvr for much simpler interface mane
how can i record my voice with the voice changer
Ya and you'll murder performance by putting useless translation layers
use it as a virtual microphone then record
im prolly better off sticking to pc
For sure lol
yes u are
AudioRelay?
i cant tell if ur being sarcastic or not
also does running an audio through uvr multiple times hurt the quality or nah
doesn't matter im exporting it in mp3 anyway LOL
I gotta replace the windows drive, I moved it to another PC and I haven't bothered setting the new one up lol
oh ok
not sure if i told u yet but i tried dualbooting linux and my drive corrupted mid installation
never touched linux again after that
do u guys think i can run rvc v2 locally on my pc 💀 using the colab is a hassle
How did that even happen again ?

I don't recall the requirements for UVR, but you probably can. It might or might not take a while though
idk but at like 40% my pc crashed
i would assume rvc is much more intensive than uvr
Boyman
well u literally have google's cpu and gpu running the code for you in a colab 💀
yeah lol
and it still takes time
Do I look like a boy?
how tf do i know
so i would assume that a 5600g would rather struggle
It shouldn't be taking out windows with it though

Unless the partition selection went wrong
It's definitely going to be slower
well windows still worked
ONE GOLD!
i just need a way to upload audios faster (my upload speed isnt the issue bruh i have 200mbps), and how to not get disconnected from runtime
DID WE FLY
Ayo? @stray crescent level 3 !!! 
wowowow
i dont understand what julia boberts is saying

NO WAY
@indigo epoch hey would you mind giving my first ai cover on pc a listen, im open to feedback
send it
Biden blast
ight dms
Obama Mama?
I can do that!!! I am learning audio mixing!!
Can I?
sure bet
Awesome!
Okay, I have a lot of criticism to give
First of all, what DAW did you use to do the mixing?
Ah I see
hii
Morning!!
Where did you get the instrumental and vocals from?
uh youtube
Was the video on very low quality already?
then ran them through uvr, separated vocals from instrumental with kim vocals 2 then deecho and dereverb
ah that might be a problem, the site i used to download the audio didn't have a bitrate function
Ohh dang. I see
I got you
Next time you download a song from YouTube use this website https://y2down.cc/es/youtube-wav.html
Select the WAV format
When converting audios using RVC or UVR, it's best if you use wav audios. What's special about the wav format is that it has all the data about the file, so that way the apps can convert with much precision and quality
reason why
wav is lossless
Second, lower the room size of the reverb and the Wet value. The wet is for the loudness of the reverb
And third, export the final audio as either flac or wav. Mp3 compresses the quality of audios
@stray crescent got it?
Hi, please help, does anyone have a voice model (Pomni.ipynb) for google colab?
First, don't use Google Collab. It's very unstable and Google has been banning people that use the collab spaces
oooo
Use RVC or the huggingface spaces
alright thanks for the tips
-rvc
RVC Guides (How to Make AI Cover)
Translation by country
Problem with using wav is
it takes too long to upload cause of the file sizes
huggingface spaces are buggy as hell for me idk why
Then use flac instead. Wav and flac are formats with lossless quality, but flac has a smaller size
like whenever i try downloading a model in the space
itll say "an error has occurred" in the output
hi
interesting, will try
What link do you paste? From mega, google drive, or huggingface?
Of the model
and i also tried duplicating a space, didn't go so well, waited 10 mins for it to build then only to be greeted with a blank white screen
huggingface
Are you using Brave browser or an adblock?
Wow, thank you so much, I didn't know about that.
nah im on chrome
incognito mode
You are welcome! And in regards to the models, you can search them on #🔍│find-models or in the website weights.gg
there's actually a google docs with realtime updating full of models
You are logged in huggingface, right?
yep
maybe its just an issue withh the one i used
nope i dont
Like, does the tab stay on your screen all the time?
yes
buddy, you're the best.)
can you send me a link to a space, ill give a try later
Hm then try doing it from another device
Thank you ^-^ good luck!
also can i send u one i made on mobile
i think its my best one from mobile tbh
not that good tho since i used x minus for isolation lol
https://huggingface.co/spaces/r3gm/Aesthetic_RVC_Inference_HF this is the one I use. It even works well on my phone!
thanks, bookmarked, ill give it a shot
You are welcome!
You think so? Why?
minimal voice glitches, and the fact that i actually got to preserve the backing vocals and reverbs and all the effects so i didnt have to make my own
n-Track Studio
Hm I never heard of that. In which audio format did you export it?
and i got the inspiration from a youtube video, not to brag but they had to use a different version of the song in the second half cause theirs broke due to the backing vocals
mine didn't so i didn't have to use a different version of the song
Oh okay
vocals and instrumentals wav but the final product was mp3
There was an issue reading this thread from AI Hub! Make sure to go to AI Hub > #voice-models, and copy a link or message id from exactly one of the threads there.
Oh and does the app allows you to export it in wav or flac?
it's actually quite advanced for a mobile app, only problem is you gotta turn off the internet when you use it cause it has ads
Oh no that's awful 😭
Rooted android master race
This isn't an RVC2 model! Please make sure to use an RVC2 model from AI Hub. (Make sure the model is tagged with RVC v2)
Hi tetski
Lucky patcher master race
Thank you tetski :3
Yo, where can I access the Google link where I can make my AI cover ?
I wouldn’t use colab for inference
Don't use Google Collab. It's unstable and Google has been banning it and the people that use it. So be careful
-rvc
RVC Guides (How to Make AI Cover)
Translation by country
Use HF spaces or wordband
Hi
@wintry sinew most normal bio in AI hub
Poopmaster is normal
oh shoot @idle light
unfortunately, saving as wav is locked behind a paywall and my free trial ran out
They did????
Yeah
Aw shoot 😭😭
They say it’s seans fault tho
Send it on mp3 then
Sean who??
Seans fualt
@tardy hare
He said somthing rude so pixie throw a cup at him
seán McLoughlin
This isn't an RVC2 model! Please make sure to use an RVC2 model from AI Hub. (Make sure the model is tagged with RVC v2)
Unless they have a picture on the wall saying “DO NOT SERVE THIS PERSON”

hi
they dont have it in my country
Iceland moment
We are the McDonald’s
Sigh You know, someday if there's a RVC Studio tutorial. I could make my own voice models. I just hope that the process is easy.
Ayo? @short ravine level 3 !!! 
Like every 15 miles or less there’s a McDonald’s
Respect da Mickey D's!
wth is RVC Studio?
Just use Mangio!?!
yeah I know..
Go ahead and see for yourself. https://huggingface.co/datasets/SayanoAI/RVC-Studio/tree/main
Squishy I just got word from the authorities that you were banned from chick fil a and you are in debt with the chick fil a mafia
What happened??
How did you manage to get a restraining order from any chick fil a on your proximity 😦
Well, I could use Mangio. But I want to make the voice models speak, not sing. RVC Studio could use some light shed on it.
RVC STUDIO IS A THING!?
Yes, sir.
link
ha ha ha
just like they said
Séan deserved a drink in da face
Are people allowed to say slurs in the server?
Chick-fil-A staff didn't like it