#🧬│ai-chat
1 messages · Page 148 of 1
Ima make a chart
With your bot, Astra, and wordband
Let’s see which one has the most features
ok. but please try it out all the features and be objective.
no thanks.
The way you talk about the bot makes you seem like your promoting it
i kinda wanna know how to make it better.
can someone make me a google account rq
its asking me for verification & im not doing it thank u
i use pia
how do i verify lmfao
It's usually phone number
have you tried our bot?
give me wordband
.//./
@modest citrus
it's not true since you never tried it out
Ayo? @blazing depot level 3 !!! 
smh
Ok tell us about the bot
@cerulean glade
omg how can you make that graph without even using them all?
Try using it in #1159290666019471421 with /cover
how can we believe in it.
No
Wait let me check the command
I watched your video
This is correct
I’m on your discord
made the video as short as I can
you didn't do anything.
That’s all that’s added vs the video
its been a while since I have made an ai cover. i heard it chagned and u cant use gui anymore so is there a new video guide explaining how to do it?
You can't guess what the bot can do.
Ok what else can you do with your bot then lil Bro
I AM IN YOUR DISCORD SERVER
how long do i have to wait for a cover btw @wintry sinew
and you didn't use any commands
how do i make ai covers i forgor
I’m checking right now
and make any video. You have simple understanding by reading the one line descriptions of the command.
you will love it after you check it out.
Same with Astra
im distruibuting every song i make on this to abacous
why is there so many ayesha erotica models, she doesnt want her music being used
They should be taken down
they wont
Ayo? @elder willow level 3 !!! 
i own a label and its approved on spotify
Ask ai hub to take them down
They will take it down
oh
i thought u meant the songs im distributing lmfao
Thanks for the suggestion, that feature is already in progress, will get it added because you requested. And the messages don't disappear for both bots
is rvc still available in colab? if so is there a video tutorial on how to use it because I have not done anything ai in months
👍
ok im just straight up being ignored. this is probablly the wrong channel but I have 0 idea where to ask at I have not been here in like 4 months
yeah, and no there isn't video tuts, there's stuff in #1159513888199540817 though
this ai is my ass
sorry for not answering btw im just focusing on helping some other ppl
yeah i looked through that had no idea which one to use 😭
Fragment 2
Result 1
Artist: Ayesha Erotica
Album title: Barely Legal
Track: Where You at
Label: MERLIN - Ayesha Erotica
Score: 94%
Result 2
Artist: Ayesha Erotica featuring AlexZone
Album title: Daddy's Princess
Track: Where You At feat. AlexZone
Label: Hot Bitch Ltd.
Score: 79%
Result 3
Artist: Diego Edwards
Album title: Hallelujah
Track: Hallelujah
Label: West Winston Recording
Score: 64%
why am i getting audio analyised with ai covers this is embarrasment
need one that will support my old rvc voice models that should probablly be retrained bc rvc is probablly better now
Ayo? @alpine fjord level 2 !!! 
theres one for ilaria rvc which is a huggingface space (basically, same thing as colab but cpu only)
_ _
ooooh]
well
uh
no
rvc hasn't improved at all
i mean i can upload it to hugging face
it's still same old v2
lmao
am I able to use it without hugging face though? really dont wanna make a new account
seems like hugging face is required, that sucks.
Girl normal voice?
-egirl
Suggestion for @unreal lotus
Searching for the perfect RVC e-girl voice model? 
man
man
man
man
how do i upload a model to hugging face
whats the website
Ayo? @distant oriole level 4 !!! 
im slow
hugging face
yea ik i meant which one but i found it
how did you get model maker without knowing how to upload your model to huggingface, also it has to be done in a specific way.
its been a while man..
check #1159514067187277865 pins then.
also you used to not have to use hugging face when I first got into it
nah im good now
its been months since i have done this i just forgot lol
if its not in correct format for the repo its not gonna be good lil bro. you should check
??
model has been trained and worked in the past no reason it shouldnt now
if the license isnt openrail aka the thing that makes it okay to actually upload the models then thats on you for what happens. just pointing out what should be done.
Ayo? @elder willow level 3 !!! 
???
Ayo? @alpine fjord level 3 !!! 
what??
the voice shouldnt be copyright protected or anything. nobody is going to come after me
Thanks bro
😔
@kindred scroll
Skull has been crushed
either he just got banned or he left lol
ty pookie 😊
considering it is LITERALLY MY VOICE and it is private i dont think i should need licensing, idk how it works tho
i didnt have to first time i used hugging face but idk
oo ee oo I look just like _______________________________________________
come on someone finish it
how do i get the link to my rvc model?
i just copied link to it and it said link invalid so idk
Do you guys have a Hank Hill voice?
happy new year
do i upload a .zip or just a bare folder?
its the pitch
fundamental frequency of a voice (should have been taught in physics
)
say u were using a deep voice model on a high pitch song, you would have to raise the pitch
Slow down, they do have Hank Hill voices #🔍│find-models and you can read the basics here https://rentry.org/ilarvc_inf_guide
do i upload the zip of the voice model to hugging face? or the contents?
Check weights.gg
Zip
it takes so long to upload 😭
Nvm got it working
how long is it supposed to take bro 😭
anyway it keeps giving an error when uploading to rvc
"ERROR - The download failed. Check if the link is valid."
how should the link look?
What was the link?
bruh
Change blob to resolve
Ayo? @alpine fjord level 4 !!! 
Depends on the Wi-Fi 🙏
its done already
Hey does anyone happen to have a arbys guy ai voice?
so when i try to run the start http why wont the black screen from windows load? do i needa wait for something to install?
If not someone should make one think it would be funny to make meme with
is it true that literally no one knows what model duckus uses

Wsg
lil bros models get exposed within a day, you just arent looking hard enough, its just bc he doesnt make his own. go on weights.gg most people will literally rename their model "duckus voice" for clout.
wait what the fk are epochs
Information for new members!
Voice Changer help
#🔍│help-w-okada
Cover & training help#✨│ai-help
Searching for models#1175430844685484042/#🔍│find-models
https://weights.gg/
iirc they’re just a way to give a number to how trained a model is
I could be way off the mark lmao
I know I already said it but please someone make a arbys guy ai voice I really wanna make some memes lol
It's there
Then wait and see if someone takes your request lol
Yeah, more doesn't mean it's going to be good. That's why you don't see model masters everywhere
Is colab a good source for training?
There’s a lot of technical shit to it afaik
Depends on the dataset and allat
If you aren’t capable of training locally yeah
What does that even mean training locally
I mean I trained a voice before but the whole model came out to almost a gig
Like training on your own device, installing all the software for it
You gotta have a pretty high end pc far as I’m aware
And this is done on a webpage or a application
anyone online to help me out with my issue
Application.
Ohh I would figure, but yeah if I could I'd train the arbys guy voice myself just don't know if the webpage would be a good training page
There are people who are better with training online, but the little I’ve tried to train went well online

You just gotta make sure to have constant backups and make sure your colab doesn’t disconnect
Well the one I did, did work it's not like terrible outright but guess it couldn't hurt to try ig
Just don't know why the hell the index file is like 800+mb💀
For my other voice
i need somi link
HAPPY NEW YEARS
Happy new year y’all
Happy new years guys,looking forward to more voice models this year!
Hola Amigos!! Este es mi mensaje especial de Año Nuevo. Ninguna porqueria copiada y pegada de
esas que dedicas dos minutos para felicitar tanto a familia, amigos, compañeros de trabajo o cualquier
otra persona. Esto vale para ustedes que siempre
estuvieron a mi lado, que me conocen muy bien y que
me quieren por quien soy. Para ustedes que hacen
parte de mi historia! Porque los quiero y para mi ustedes son el mejor equipo de waterpolo del cual
tengo orgullo de ser parte. Feliz 2007 y felicidades para todos.
Enviar mensaje
Merry 2024.
ayo where the mods at!!!???
@timid musk (sorry for the ping)
HAPPY NEW YEAR!!! (I'm putting this in every server I'm in even if its dead. Rejoice of 2024!)
Yo
Ayo? @hearty copper level 1 !!! 
fixed
gm!
hello
What other activities can you do to get tokens?
hey
anyone know how to make those raspy plankton ai voices? a link to a model or smth?
Guy mickey guy
Hello chat
I'm new here
And this server is so cool 🔥
It's pretty easy to find stuff here
What.
what being able to read does to a person
Btw how i can make own ai voice
I wanna import my voice from real life
(don't blame if i have bad grammar)
Alright thanks I'm just new here
i kinda can't find the channel where i can get info about it
Ayo? @ember cradle level 1 !!! 
the one id say to read is called "training models with mainline rvc" its good enough but other guides also are fine, if you dont have good hardware look for one on colab, rvc disconnected specifically
my hardware is very bad
all the information needed is in guides
so you'd want to use colab for it then.
since it runs all online, meaning your hardware has no effect
its free.
rip
You know that is?
i kinda forgot what colab trully means
how i can remember it's when you do smth with someone
i have memory loss
It an abbreviation of Google Colaboratory
oh ok
A product of Google that allows anyone to execute and publish arbitrary python code
Delete that, please
wth
helper role, you can simply delete it yourself and time out
ahh hell naw wth
Ik. Just tryna give them a warning
where did i slide into 😭
Guess they won't do it
men are hot
the first thing that happens when you join tells you to look at rules.
bro is so dirty 💀
and its the very first one
i just wanna be innocent and happy...
facts
some people have brain rot.
especially me when i see toxic people
toxic people who say disinformation etc
like why people just can't resolve problems like normal people without becoming enemies instantly i mean people don't even listen to each other to try to understand the situation
That requires people to have this thing called common sense and or human decency. Which unfortunately, most people do not have.
i think 5-35 procent of people do have
it's just hard to find those people
and i wanna change topic
cuz it's my first day being here
Understandable
uhh okay?
Well if you do need further help, go to #✨│ai-help since that's where you will be if you get stuck on anything for making your own model.
where exactly can i watch all info about making own ai voice?
Ayo? @ember cradle level 2 !!! 
all the info would be in the #1159513888199540817 section
you just have to find the specific guide you need.
the mainline RVC one is good, but then you also want the colab one on RVC Disconnected.
that should give you more than enough information to be able to clone your own voice and or other peoples voices.
ok thanks
i just wanna make my own voice
but evil voice not standart
i have an evil fnf voice
which i made by recording myself
so talk in that specific way when you are recording your audio for it then.
alright
or just pitch shift your normal voice after its been made so its deeper
wanna hear the voice btw
i didn't pitch i just made deeper voice
0% of effects has been used
except pitch
cuz i had to do pitch for every note
that is what pitch shift is, you can do it normally just by changing how you talk, or you can use something to artificially do it
and can i have your dm's ?
your seem nice to me
i also search for friends who's doing ai stuff etc
ok
I'm good, I'm not adding anyone, just helping on the short time I want to be here.
can i?
Nope, I'm good.
alright
and i wonder if someone do play fnf here
(i actually don't play fnf anymore i just do remixes or covers)
i went to osu 4-5 weeks ago
ugh
i did play fnf for 3-4 years
you went?
i mean i don't play fnf anymore i quit it
i switched the games
i think i said it wrong
i meant instead of playing fnf i play osu!mania now
will i get banned if i still use this for training 💀 https://colab.research.google.com/drive/1TU-kkQWVf-PLO_hSa2QCMZS1XF5xVHqs
YES
anything with a gui yes
Ayo? @elder willow level 4 !!! 
-colab use RVC Disconnected for training
- AICoverGen-WebUI, modded by Hina Google Colab
- AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
- AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Ilaria RVC, by thestingerx Hugginface Spaces
- 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
HAPPY NEW HEAR!!!!!!‼️‼️‼️‼️
But idkw the model i trained with Disconnected sounds so different from the old colab
YES HAPPY NEW HEAR 👂
so im taking the risk rn💀
Set sample rate to 40k
Use 4 batch size
Then it should sound the same
Happy new year everyone 🎊😎
Ayo? @wooden plaza level 1 !!! 
Do not just outright recommend a sample rate, actually teach them how to find what to use. And if you are going to outright recommend one it's going to be 32k.
i want one discord nitro
So get it
if i already subscribed can i use it?
alguien tiene la voz de takanelui
Subscribe to what?
colab pro
You don’t need to pay
I mean for the old one😭
Did you try 4 batch size and Lower sample rate?
so i'm taking the risk💀
i always use 8 batch size, i havent tried that
rvc disconnected would be more up to date, so it would not sound worse if anything itd end up being better, but you still have to do your part and actually clean your dataset
Happy New Year 🎉
thank you
i'm know
i wish you could preview voices
i wanted to see what foxys voice sounded like to use it in something but it had an error somehow
a server error keeps occuring
well
i guess thats too bad then
This is actually due to a common error on Windows, it can be easily fixed by pressing Alt + F4
im not a child i wouldnt fall for something so dumb
AI voice changer for Core i5 12th is suitable?
you should have a gpu.
I think, my CPU is not blows
the voice changer runs mainly on gpu
hlo
I use GPUs through AWS, but it's all expensive
you do not need to pay for anything just use free kaggle or free colab. but if you have a gpu on your computer, you can run it locally
yall know any good models for a proudcer tag?
Hey guys
can you guys help reccomend me a narrator voice
something that would come from a true crime narrator
hi
hi
hi
Happy new year
hi
is there no ty dolla sign model?
just grab a random tts and put like a bunch of effects on it, should work just fine
Hey its me WesPes
hi
Ayo? @lime monolith level 1 !!! 
new year still not verified
Ayo? @ember spoke level 3 !!! 
sigh
i miss being able to talk in that juicy succulent verified chat
not stuck and forgotten in this ghetto pooron chat to be forgotten
is there a place to post your covers here like in the other server
Any effects ideas
reverb and saturation probably
anyone receiving this error?
"NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check."
this happens when im trying to use img2img
tried to reinstall too but doesnt work.
in the meantime is there is any website that can do img2img well?
Hello, I am making a model that in each epoch takes 1:04 minutes, so it will take 17 hours because it is 1000 epochs What do I do to continue resuming the model when Colab removes the GPU for 24 hours?
i mean the error literally tells you what to do
i guess you could find a colab notebook and use that tho
i've done what it tells me to do, only works for once
but the 2nd try will ended black blank images
same thing is happening to some people on github x,x
what hardware u got
tickle my gooch until it chafes 😏
Ayo? @slim dawn level 1 !!! 
Can it be picked up by simply putting the same values and pre-processed data?
BYE BYE 🗣️
Verified closed?
Wat

whats the name of the websites that have all the archives voice models
tq
clifford
Ayo? @next cosmos level 1 !!! 
is there a website that removes backing vocals from a vocal track?
Can someone help me
what is it?
Do you any place I can go to make ai songs for free
I think uvr5 does that
Ayo? @edgy leaf level 1 !!! 
Suna is free as far as I know
Is there a way to train a voice model with no google colab and no nvidia cpu?
is it free?
I don't know, I don't think so. It's called suno by the way, not suna
Bruh how do I train a model I have been trying for 2 days and keep getting errors
So it's basically impossible
I found one
What error, and what are you using
I am using rvc v2 locally and I get the error since I have no nvidia gpu
Yeah, you can't train without an NVIDIA GPU.
How do I train it then?
- AICoverGen-WebUI, modded by Hina Google Colab
- AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
- AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Ilaria RVC, by thestingerx Hugginface Spaces
- 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
Which colab link mine keeps disconnecting
Ayo? @edgy leaf level 2 !!! 
RVC Disconnected is the one to train models
Do I run all?
You're probably using one that's been restricted then
Thanks but do I run all?
YES
Very helpful to get the thing right
Thank you so much
How many epochs do I train it at I have about 1 and a half hour of voice
And if you have any issues just lmk
Honestly, that is way too much dataset
How much do I use?
How long does voice models take to download
wait can u have the audio only in one file or have to have it in multiple files?
only one file
k
stupid question but how do you compress a file into .zip
nvm i figured it out
one last question what is the optimal no. of epochs for a 40 mins
hi guys
h
h
h
go for 500 and check the graphs for every model
if it's alright then keep training to 750, then 1000
Idk how to check graphs
-overtrain
Hey, @edgy leaf!
👇 Here are some resources to help you identify if your model is overtraining
All-In-One Guide on how to make a good model
This guide explains how the D and G files works and much more: https://rentry.org/RVC_making-models
Credits: LUSBERT 
Automated Overtraining Detection (AOD)
Will be available soon in #1159513888199540817
Credits: grvyscale
im going to get a 2k loan in 5 months im in the queue

whats the best pc build in that price
How do I reset a google colab data?
Ayo? @edgy leaf level 3 !!! 
What do you mean
Like I accidently deleted some files now I can't run it so I wanna reset it
Wait I can ask the new colab ai
you can disconnect and restart runtime
That still retains my files
it doesn't
literally disconnects and deletes current runtime
so you can start a new one
It will not disconnect even if I leave it in background for like 1 hour?
@forest quarry
Tell me someoneeee
-howtoask
How to troubleshoot
- Give context: Throwing your error/issue alone will not help. Tell what step are you on, what were you doing before issue, your settings, your terminal, et cetera.
- Be relevant: You come to ask for help, not to complain. You can't ask for RVC help in voice changer help and vice versa.
- Be polite: DONT DEMAND FOR HELP OR I WILL FUCKING MURDER YOU. Ask politely, the server was not made for you.
- Don't be an attention whore: If people are ignoring your help, kindly ping them. If they are ignoring you, either they don't know how to solve it, they are busy, or you are an asshole.
WITH ALL DUE RESPECT, DO NOT BE AN ASSHOLE AND HAVE COMMON SENSE.
Ask in #✨│ai-help
Thx will do that now
@idle light are you to hlp
Ayo? @elder willow level 1 !!! 
will a rtx 4070 bottleneck with the ryzen 5 7600x
no what
this aint gonna be GPU bottlenecked xd
What
Hello
salut, c'est quoi les paramètre pour la meilleur qualité ?
Hello everyone
hi
Ayo? @vapid inlet level 1 !!! 
what are the settings for the best quality?
in what? training or covers?
hola
cover
use rvmpe as your pitch extractor
thats about it, other stuff you gotta test out and see what's best
- Applio-RVC-Fork, by IA Hispano GitHub
- Mangio-RVC-Fork, by Mangio621 Huggingface
- RVC Studio, by SayanoAI Huggingface
- AICoverGen, by SociallyIneptWeeb GitHub
- Replay, by Replay Team Website
- Original RVC, by the RVC-Project team GitHub
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.
okay thanks 🤝
Does 100 epochs sound fine?
Rmvpe is the best sounding but take more GPU usage
if it's a long dataset, maybe?
ok
either way you can resume and train to 200, 300, y'know
you're welcome :)
If I stop training and I restart with same zip will it continue training?
if it has d/g files and the feature extracted stuff it should be fine
actuallu
actually
A guide perhaps?
Ayo? @edgy leaf level 5 !!! 
i sent one didn't i? I think it has the resuming steps
No it doesn't I think
Could u help @vivid tiger I am not too good with real time conversion
If it doesn't, it's simple:
after you exported model n stuff and u wanna come back, run dependencies, set training variables (make sure everything is the same), then do the resuming steps, and just train the model with the same batch_size
K
-colab
- AICoverGen-WebUI, modded by Hina Google Colab
- AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
- AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Ilaria RVC, by thestingerx Hugginface Spaces
- 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
Wait I can run rvc v2 on hugging face?
yeah, u cant train tho
only inference
Bruh
random question, is there an AI cover maker that uses RVC voice models, requires no python, is something I can just download and launch an .exe from it, and requires no sign up/can sign up with discord/no wait time. (very specific, I know)
Easy gui sucks I tried it
Why not jus use rvc?
Why can't u run python?
.
what about what now?
also the main problem is it's not easy to use
Just extract vocals and instrumentals then put it in colab
I dont want it to be a website tho
I am pretty sure that specific website doesn't exist
-local
- Applio-RVC-Fork, by IA Hispano GitHub
- Mangio-RVC-Fork, by Mangio621 Huggingface
- RVC Studio, by SayanoAI Huggingface
- AICoverGen, by SociallyIneptWeeb GitHub
- Replay, by Replay Team Website
- Original RVC, by the RVC-Project team GitHub
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.
da heck is replay
He doesn't have storage
I mean I could try replay
yeah try replay
Jus download uvr5 and run rvc locally done
No net
Only need song
hmm
replay is kinda easier tho
main problem with rvc is I have no clue how to use it
Idk what replay is 💀
then lemme try it
See a tutorial on it
check their website lol
Rvc is one of the easiest things I have learnt
Bruh idk what ur talking about
again, if you don't know what replay is, check their website
just sent the link to their website
I didn't know their website in the first place lol
Ayo? @gleaming mist level 4 !!! 
yeah exactly
It's been 2 hours do I stop it
yeah I'd recommend stopping
what is the most convincing like 16 year old girl voice that doesnt sound bad 😭
Searching for the perfect RVC e-girl voice model? 

why the fuck you need a voice model of a minor?
catching discord predators and confronting them afterwards 💀
im conflicted
from one stance i would love to help you but i dont deal and let no one deal with model of minors
then any convincing girl voices?
Ayo? @eternal burrow level 1 !!! 
if you want i sell high quality ones if youre interested 
how much?
here you go: https://ko-fi.com/ilariaowo/shop
also you can find free ones there, including as #1175430844685484042 but tbh they are not super hq most of the time, ill suggest you to check around and drawing your conclusions based on your case scenario
ill consider it, thanks
no prob!
oh yea also i forgot
also check weights.gg
for free ones
ight, thanks a lot lmao
it does
google colab or local much better
NAH IT TRAINS ON THE WORST PITCH EXTRACTOR
Just use colab
TO SAVE COSTS
Which one
.
NO INDEX TRAIN ON KITSAI?
LITERALLY NO INDEX
YO💀
$9,99 you get all those wonderful features
WHEN WITH $10 YOU COULD BUY COLAB PRO AND TRAIN SHIT FASTER

suuuure, it ain't easier
but with the guides it's pretty easy
colab better and it free
yeah
i swear to god i heard the ai voice speak to me 💀
just make lots of account and read the document
fr
-colab
- AICoverGen-WebUI, modded by Hina Google Colab
- AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
- AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Ilaria RVC, by thestingerx Hugginface Spaces
- 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
maybe ilaria rvc
Using Google colab on phone will be difficult tho
So I'm trying to get RVC Studio to work on my PC, but when I run conda-installer.bat like the readme tells me, I get a blank CMD prompt and nothing happens
Ayo? @fierce cipher level 1 !!! 
I tried linking to the github but discord said no, it's SayanoAI's RVC-Studio github
Is there something obvious that I;m missing
I'm kinda new (and stupid) when it comes to this
is felt here
this server was a lot better a couple months back before yall showed up
sorry
copyright exists
and that's the reason the first server got killed in the first place
¯_(ツ)_/¯
first server was a lot better
people were nicer and actually helpful
and less mods who banned everyone who spoke ill of anything
yah old good days

my genuine reaction
that just showcases my point about this server
It's just a meme bro, lol
true
nicer in not being such an idiot
false
nah, the mods on the old server were actually chill
^
blud came here complaining about things we dont have power on
Isnt the mods the same from the old one? 😂
not our fault we cant let yall post copyrighted stuff
yeah 💀 most of em
this is what i mean
if you get offended by that message i really dont know what to say
me too, i dont know if im incredibly stupid or im missing context
discord banned the ai cover sharing, staff is the same as the previous one, no one was ever “sinlenced”, we are objectively more “nicer” than before since now at least the help channels and stuff like that actually works and verified was a shit hole
so im kinda confused on whats the issue here😭
fr
Ayo? @elder willow level 3 !!! 
depends on the matter, we have to follow discord tos, dont think we like doing some stuff
but if we want to stay online we have to comply their rules
@hollow tangle Hey, can you help me find a certain model?
Ayo? @meager nova level 1 !!! 
hold on i got a question
#1175430844685484042 and/or weights.gg are your best bet
whyd u ping me
sure, what is it
is yeezybeaver on this server?
yes
he was the nicest dude on the last server
hes demoted to contributor now because hes always off discord
like on one of my ai mixes, he left a nice comment and was overall a sound dude
Thank you! Im looking for a model with a deep/rough voice, a narrator voice that you would see in True Crime documentaries/videos
tbh youll find nicer people in the verified channel, this is literally a filter before people actually speak there
I'm thinking of the Max Payne voice models lol
mhhh, i dont know tbh
He's a goat
pleaseeeee
i dont have permission to speak on that channel
lmfao not a bad idea
fr
Get level 5 from talking
man i said i dont know saying please wont give me ideas 😭😭😭😭😭
just check models and listen to the samples
gotcha
Pulls a gun to your head I INSIST
Nuh uh
also, is colab still a thing for ai vocals or is there a gui now?
LMFAO
colab is banned by google
Webui was banned
yeah, but ppl r using huggingface stuff now
-colab
- AICoverGen-WebUI, modded by Hina Google Colab
- AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
- AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Ilaria RVC, by thestingerx Hugginface Spaces
- 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
same thing at the end of the day
For free users
oh look theres me
How many epochs for a 9 minuite voice model?
here's all the working stuff
300
No ui still works
go for 500
check graphs
Is it good
i disagree
oh of course you do 😭
important thing its the dataset is super clean
I have a budget pc
i just have a bad expeirence with huggingface
not my fault we have two different training methods
why?
then dont train locally
im just saying, it's rare for short datasets overtraining so early
idk, it just never works for me
trust me, they dont overtrain normally but you can hear robotic stuff more easily
strange
ofc it's all dependent on like
the dataset quality
BUT
if you have a decent dataset
500 would not be bad at all
and plus u can use earlier versions of the model
or even merge the best fm/mel stat ones with the best g point
im too cool for all of this
that'd be amazing
A better moana voice model?????? pls reply
do you know which is the best huggingface model to use for ai vocals?
good question, but I don't know any really good ones
actually now you gave me an idea, merge a lot of vocalists' models
and make an unique voice
👀
,HDH
I'm thinking to merge Michael Jackson, Freddie Mercury, and any other really good singer maybe
nice
ive seen people do that and upload it onto spotify
its technically not them singing it
It's a good idea, makes a unique voice that can't really be replicated normally
I'll do a bit of research on how to merge the models, maybe I'll come up w something
i would try it, but isnt it copyright?
it technically wouldn't be, since it's a new voice... but I would not risk it
I want someone to cover a song called "3 Nin No Uta" No not that one That one from "Koraru No Tanken)"
i saw a dude get like 1mill streams with that
I want someone to cover a song called "3 Nin No Uta" No not that one That one from "Koraru No Tanken)"
damn, that's big
okay man
til this day, it hasn't been taken down
Im not sure if the song is copyrighted......
it probably is
How can i send it
send what?
The 3 nin no uta song
i think I'd recommend not sending it
Its in japanese
ya know the rules, no copyrighted stuff
Is there a website that checks for copyrights on youtube
no
you'd have to upload the song privatley and see if youtube picks it up
https://www.youtube.com/watch?v=aTNFMIz8oAE&pp=ygUPY29yYWwgbm8gdGFua2Vu Here's the song link if anyone would like this song to be covered....
anyone?
im confused, so you want someone to ai cover this song?
Yeah thats what i meant.....
An Ai Cover Of 3 ri no uta
can't you do it yourself?
I need someone to make a better ai cover of this song
How do you stop models from sounding robotic
yh but you saying ai cover could mean anything
Ayo? @elder willow level 4 !!! 
like whos voice do you want for it
Maybe Shizuka From Doraemon?
anyone?
idk get it tho
Look the character up
you could try and make it better yourself
Oh
it depends on the vocal stem
https://www.youtube.com/watch?v=aTNFMIz8oAE&pp=ygUPY29yYWwgbm8gdGFua2Vu Can anyone listen to this song and see how it sounds?
and quality of the actually epoch
@hollow tangle is there a producer role for this server btw?
this is made with AI?
makes sense, it sounds too real
I have found 3 results for the search mazlum kiper...
Owner: Ozan Efe Akpınar
Uploaded: <t:1692201510:R>
This model can be found on the Kits.AI platform, visit their page for more information.
Did you like the song?
it's nice I guess
i don't understand the lyrics so I can't really say if I like it or not
What Genre was the sonng in?
what do you mean?
Like Jazz,soul,Funk etc
Hello
oh , I don't know
I have found 3 results for the search mazlum kiper...
hello i need help with an AI model, im using xxxtentacion model but for some reason the audio i convert never sounds like him yesterday i didnt had that problem anyone know how to fix it?
Ayo? @jade valley level 1 !!! 
hi! Are you training locally or on cloud?
im a hard noob so im not sure if im training, i just try to convert audio to xxxtentacions voice
maybe try a different model
sorry, I meant converting
hi guys happy new year
there should be a couple of other xxxtentacion models
there are only 3 models and only 1 for rapping
ello, happy new year for you too!!
Happy new year to you too!
but i think u can use the others for rapping too, nothing stopping you
ok i will try it out appreciate it
how to search for free ai voice models
HAPPY NEW YEAR AI HUB!! 🥳
Buy e-girl, mommy and v-tubers voice models here: https://ko-fi.com/ilariaowo/shop
You can also find FREE voice models! 
If you DM me and pay with PayPal you will get a 20% discount! So do not esistate! 
I wonder how to use "/search" in search-models channel and what the key word for chinese voice model ,thanks.
So-Vits-VC is obsolete
There will be
ai voice audio is so weird it doesnt sound nothing like the artist i selected
yo
anyone know what could be the reason
-svc
Yeah I still use so-vits-SVC
-rvc
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
Model training with Mainline RVC
Link: Rentry
credits: Raven (ravencutie21)
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How To Make an AI Cover With Ilaria RVC
Link: Rentry
Credits: 👽 Julia (ailen2091)
-svc
Yeah I still use so-vits-SVC
-svc -sv
Yeah I still use so-vits-SVC
-svc -svc -svc
Yeah I still use so-vits-SVC
-svc
Yeah I still use so-vits-SVC
й
i have swaggersoulss voice saved
i want to make it into an ai model
how do i do that
all i have is the mp3
hi
either the tone or quality of the model ig
like original voice being too deep or high,or quality of the model
so i have to look for a cleaner audio to convert it without problems yeah?
Ayo? @jade valley level 2 !!! 
can u guys help me i wanna make my mp3 into a model
its a 30 second clip of swagger talking
How do I continue training a model after my collab will close?
gpu
are there any white ssds our gpus out there
im trying to make a fully white pc build
prob not
there’s already models of swagger
i mean you could just paint the cases white for them if youre not a dumbass
otherwise just like cover them with something white
oh i found a gpu
i’ve seen white gpus
Fr?
Ayo? @plucky lark level 1 !!! 
is 32 gbs of ram enough
What are you doing on the pc
gaming
Then yea
Hello! I'm trying to make a deep fake video but have a post-processing issue. The source face has lighter skin than the target, so the output skin color is way too dark. Does anyone know a technique or colab that could fix this? unsure if it's okay to ask this here as its not related to voice models :/
. If you know where else I could ask this or have advice, could you DM me? Would be willing to pay someone to do it also
anyone recomend a trump model for live
is 6tb storage enough
yes
i really dont know ive used my laptop for almost 15 years
i never ever has seen or touched a computer ever
Hi guy's does anyone know how to create Ed Sheeran model?
anyone know a good trump model recomendation??
Ayo? @keen ember level 1 !!! 
If anyone know pls tag me or dm my friend is sick and he loves Ed Sheeran and I want to give him some positive messages with his voice
What if the audio has slight background noise, or background music
@meager arch I can help you
hello
hello
remove it with UVR
Anyone have a Lady Gaga model?
Check https://weights.gg
is background noise affected by the voice changer also? coz idk if my mic is picking like fan noises or smtg but i was using the anime voices and kinda heard crying and when i used the peter griffin i faintly heard "help me"
yes
kinda funny doe but scary
Thx :}
use sup1 or sup2
Pon

Anyone want to help me with a good trump model? Thanks
The ai wants out
Check the website weights.gg it should have some trump models on there
Thanks 🫡
Sorry I was in your walls and got stuck

how can one get to the hq voice model channel for a way better result?
Channel goes unused
Just use different settings or try different vocals or even models

Hi, do you know how I can use the AI voice modifier on discord and games? Thanks in advance.
Ayo? @elder willow level 1 !!! 
RVMPE
Hello guys. I just learned to make a model. How long should I input the voice example to make my model perfect?
Do you know how to put the file on the website on the iPhone? I only know on Android, on iPhone there is no option to put the file, whether audio or video, help
yes
neither, if those are your only options your using an outdated version
trump models already exist
for example, I've already chosen the voice of who I want, but when I enter the website, they ask me to add a file, but mine doesn't give this option, how do I fix this? I've already converted it to video and audio, but neither of those are working correctly.
what are you using
Drag and drop
Drag the file from the files app onto the box
Look, to do the cover, you need a voice and a music file or voice to transform, right? My problem is that when selecting a file, it doesn't give me an option, I've already saved it in audio and also in video, but none of them are working... the files are left without color and without the option to click, you know?
Ayo? @neon nexus level 1 !!! 
I just joined lol
Send a screenshot in #✨│ai-help
ok, thanks
Information for new members! 


