#🧬│ai-chat
1 messages · Page 222 of 1
merhaba
...no idea?
usually Google Colab (cloud service that lets u run code on google pc gpu) bans WEB UIs (Interfaces), but theres newers colabs that uses bypasses, you are prolly using an old one from a yt tutorial
Did u see any yt tutorial or our docs?
yours i think
you are trying to train a model with TTS right?
Yep
You could use either make a RVC AI Voice Model and then use any TTS Tools all online, or either use the GPT SO VITS (which is already a TTS AI unlike rvc which is audio to audio) Training colab that you can find here #🧬│ai-chat message
And which one would u reccomand/is easier to use?
Oh, and one guy said that the so vits needs an NVIDIA gpu
And i have the rx 6650xt
And i think its nvidia only anyway
GPT SO Vits is way better for tts quality but theres no Colab guides at the moment, while for RVC we got all the guides but it may got less quality for tts as you are just using the voice model you made (which is audio to audio) to a generated TTS Audio which is kinda different
it doesnt matter your pc specs if you are using google colabs
when you are using a cloud way like google colab, you use remotely google colab gpu pcs, nothing will happen locally in yours
Id use audio to audio, but when i tried to use one, the voice quality was awful
Google colab offers around 4 hours daily for free of gpu (T4)
unless you use too much RAM
Ayo? @dapper trail level 39 !!! 
in that case the runtime would be disconnected
use one or train one? the model quality depends on the lenght but mostly quality of the cleaned voice dataset
what? no it wouldnt gpt so vits and neither rvc disconnects u for too much ram used
they dont use that much ram
I have a freshly baked colab with integrated TTS and instant voice cloning (almost)
I say in general
I wasn't talking about those colabs
I'm just warning that should it ever happen to them
but thats another thing, he wont have this issue with the things he wants to use
they are using those colabs not coding anything soo
was prolly js scam/spam
that's why I said 😭
banned already
If id train on rvc i would have for audio audio and text audio anyway, but ivmean use, when i used ai ran on my gpu it worked awfully
Which shouldnt be so bad, i have a 6650xt
Which is equiveleant to like a 2070ti i think
usually training takes alot of resources, its better if you do it on colab as it uses a GPU made especially for AI things (T4 GPU)
amd 
Much better choice for me
you'd need an nvidia gpu because Pytorch (the essential shit for AI stuff) doesn't run well (in this case, doesn't run) on AMD cards
Yeah, that's whwt im tinna do from now on
^
Didnt plan using ai, for me using amd is much better for 90% of the case's
They have a good price to performance ratio tbf
Just... less commodities
I dont play games that use a lot of resources
I rarely play AAA games
I plan to do AI tiktoks with tts voice of idk like supermario, what's the best tts to use? smth like custom voice on eleven or other tts services?
Applio with RVC seems to be the best of both worlds rn
yo
does anyone know what that stable diffusion picture to video thing is called
i think the name of it began with a V but forgot what its called
Mmmm well if you looking for free things as we arent affiliated with 11labs tts, you could either use gpt so vits which is a very good quality free tts program, or either RVC which is audio to audio but you can use any free tts tools in our docs and then use the audio as an input and use any rvc voice model
gpt so vits way better for tts things usually, as applio tts is just Edge TTS, so basically automatically using a generated tts audio, and then putting it as input audio and using the rvc model lol
what if I'd like to add also intonations and emotions to the tts generated audio?
I think you can do that with GPT So Vits, with rvc well depends, as you are just using the output of a tts tool then ig bark tts would work as its the best emotional one even if it has less quality (like less cleaned) usually
gotchu grazie bro
Prego, cmq la nostra documentazione é https://docs.aihub.wtf
Last update: Mar 10, 2024
fair enough
stable video diffusion?
i was looking for viggle
hello
hi
oh
Umm actually i identify as they/them
Ayo? @modern lantern level 4 !!! 
I'm sorry
no no that was a joke
they
i aint they them i was kidding
ok I'll stop the joke
😭
I was meme'ing
xd
Yes, I knew it
How do u even call they them like sir ma'am
I don't know

🤔
I just wanted to use it because there's a specific model I wanted to use
@swift fox please use #1159290752195633273 or make sure what you send has context to the conversation.
sorry about that
My potato friend just finished making their new Kajiit character for skyrim.
Delete stored_settings.jsonfile the next time it ends in a loop
11labs it is, GPT So Vits you need a Mario model, but i dont think there's one made already
Hi
yes theres few on weights, the thing is that I can't find any quality long audio clips on yt of the mario voice I want
you shouldn't use youtube as a source for making models
heya
where should I search then
raw game files and original source
well I think the yt videos are taken from those
unless its a youtuber then you basically will have to use their youtube audio
its because youtube compression exists
if your just practicing making models, or want an “okay” model YT is fine, but for the highest quality stuff, it’s not good
what’s the name of that ai website ppl use to make songs with other artists? i’ve used it i just forgot the name of it😭
there's a bunch, but they'all paid these days
use AICoverGen, you can make covers for free
you just slap the model URL and youtube link
there was a free one i used, you can search up ppls voice models
Ayo? @steel berry level 1 !!! 
i’ll try that one ty
Where is the right server chat to share AI covers with the members? Made something im very proud of!
Found it 😄
it’s called weights ai!! finally found it
if your really cool you’ll get level 5 and post in verified :)
How do i level up? I have been a member for a year, but this is the first time i have typed anything 😄
you just send messages and be cool 😎
you can get level 5 verfied in like 2 days
What`s so special about the verified chat? Is it any different compared to the general chat?
hola tengo una duda , cual es el mejor programa de cambiador de voz en tiempo real? yo tengo el MMVCserverISO?
So I’m a little confused about how to get the models now. I think things changed.
poodoo
You mean to find them, or where to make them?
Yeah to find them.
You can find them in this server, or use weights.gg
A lot of them seem to be taken down
I prefer weights.gg
Oh I just downloaded that app
guys
ive been trying to download some ai voices.
but it just wouldnt want to upload
it just tells me the files are incompatible
so what should i do?
what website you using?
Ayo? @humble vault level 2 !!! 

well im using voice.ai
ok Ill try.
I'd advise against using it
do you have any better suggestions than the person above
Ayo? @knotty umbra level 1 !!! 
we arent affiliate with it, we use wokada which is all free instead, idk if the voices would work for voice.ai as its all paid
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
its the first guide the owkada one
what’s a really good program to train a realistic model. RVC isnt too good for the model Im trying to create. Everything is hybrid & still sounds like the og artists still.
How do I get rid of the funny crackly?
Ayo? @torn olive level 1 !!! 
use the local one if u got good pc gpu else use the google colab
make and train better quality dataset to make it realistic enough, check the guide https://docs.aihub.wtf/
Last update: Mar 10, 2024
Check out this creation I made on Weights.gg! https://weights.gg/shared/clv3zv93y0bih14lrilzlmcwa?inviteCode=16cf9
rvc is the best option, you need just a better long and clean dataset and be sure to use tensorboard
Roderick Heffley best for them clean covers
if you want better tts use gpt so vits instead which is also in our docs
hi
hi
hi
weights is mostly used for model upload, but they also added inference and training recently
so i got confused
good model/mic
might be the audio you have used
@torn olive use #1159290752195633273
its a unit of measurement of the cycle of training of the ai
So bigger is usually better?
nope, it could even make it worse, there isnt a right epoch amount
thats why we use tensorboard to see if the model is under/over training
check this out https://docs.aihub.wtf/rvc/resources/epochs--tensorboard/
Last update: Feb 10, 2024
yo
Ayo? @elder willow level 1 !!! 
how to make an ai cover?
me too
is there a solve for this?
"JSONDecodeError Traceback (most recent call last)
<ipython-input-11-320ee20c9227> in <cell line: 31>()
31 if os.path.exists(config_path):
32 # File exists, proceed with creation of creds and client
---> 33 creds = Credentials.from_service_account_file(config_path, scopes=scope)
34 client = gspread.authorize(creds)
35 else:
5 frames
/usr/lib/python3.10/json/decoder.py in raw_decode(self, s, idx)
353 obj, end = self.scan_once(s, idx)
354 except StopIteration as err:
--> 355 raise JSONDecodeError("Expecting value", s, err.value) from None
356 return obj, end
JSONDecodeError: Expecting value: line 1 column 1 (char 0)"
The datasets I have are very clean. I think for Rvc some voices just arent for the program.
no other alternatives?
you are using an outdated colabs, be sure to use the ones updated inside the cloud folder inside the rvc https://docs.aihub.wtf
Last update: Mar 10, 2024
Thank you
for audio to audio voice conversion yes, for tts use gpt so vits, for realtime voice changer wokada instead
yw
how long is your dataset and what language
I have 392 files & in english.
So on that site I can natively create covers. Is that superior to using google colab as this video instructs? https://www.youtube.com/watch?v=gtjCmJ_6d0k
Ayo? @lone nacelle level 1 !!! 
Hmm I'm getting the same error
you used tensorboard right?
yes
" # File exists, proceed with creation of creds and client"
nope, thats a very outdated video, use only the guides in our docs
"SONDecodeError: Expecting value: line 1 column 1 (char 0)"
Thank you very much
outdated colab as said
There's so much to learn 
you are using the one from our docs right ?
which pretrain?
Hi
I'm Brazilian and I'm new here
Welcome here
I will now. I thought that video had me covered
nope, 99% yt vids are outdated lmao
New to all this and feeling like a boomer
Fr
But never a better time to learn
btw for errors its better if u use help channels
I really appreciate it
-colab
- Applio, by IA Hispano Google 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
- Ilaria RVC, by thestingerx Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
Are there any resources specific to tonal languages such as Cantonese or Vietnamese?
Nothing shows up in #1159513888199540817 when I search tones, Vietnamese, Chinese, etc
default rvc is already good with chinese as it’s developed by chinese people
RVC being another ai?
Ayo? @lone nacelle level 2 !!! 
no its the same ai you are using
its called RVC (Retrevial-based Voice Cloning)
be sure to use the free daily gpu T4 of colab
only the colabs inside the cloud folder inside the rvc folder of our docs https://docs.aihub.wtf
Last update: Mar 10, 2024
ofc only for the RVC Models
theres also gpt so vits models (advanced tts) theres a section in our docs too
damn thanks
That's great!
I'll give this some practice tomorrow and see if I can get some accurate sounding covers.
yw
🔥🔥
I made a Vietnamese song in Suno, but would like to get a different sounding voice in there for consistency across this project I'm doing.
hi guys do you have any recommendation for bass voice?
what
you can transpose most male voices down pretty low
Hey guys, I have a problem. I use this Google's collab to convert my voice into a voice of someone like Taylor Swift and I found your AI models, but I can't integrate them because I need the following data for the model: .pth; .json; .pt
And I get this notification with every model I've put in with your models: "No clustering model found for Taylor Swift"
This is the google collab that I use: https://colab.research.google.com/drive/1128nhe0empM7u4uo5hbZx5lqjgjG1OSf#scrollTo=Bpg6Ql1QHEV6
What should I do?
with what app?
make sure you’re using
-colab
- Applio, by IA Hispano Google 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
- Ilaria RVC, by thestingerx Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
any of these
it’s built in rvc
this is sovits
the colab you are using is of so vits svc 4, its very outdated, we use rvc v2 instead which is wayy better, you can check guides abt it in our docs https://docs.aihub.wtf
Last update: Mar 10, 2024
nick your an ai

agreed
SVC models and RVC are different
and most of what you’ll find is rvc, but svc 4.0 can be good as well
top thanks
Ayo? @worthy olive level 1 !!! 
Hello
Hi.
what epoch/step does it let me hear the audio on tensorboard?
Ma ktoś jakiś inny model dawida jaspera? Strasznie chujowy ten jest
i retrained with tensorboard
Anyone have any other dawid jasper model? Terribly gaudy this one is
when you start training there should appear a graph, you need to look at the lowest point before it goes all up and like see the step, so in the weights folder u should find a modelname_epochs_steps.pth and take the one with the most similr step to the lowest point
yo is any good sukuna voice here
when u click on the folder icon, you should find some folders, there should be a weights one here
RAH
rrrah
sers
Is there an app that is like a voice changer, but text to speech? like an AI text to speeech program using RVC models and shit
Ayo? @ember finch level 3 !!! 
can people see the voice models page? it wont load for me
yes
☝️Gay👆
Yes
ilaria mainline
nm just needed to refresh
start rot13 or base64
Let me know when you've started
theres gpt so vits but thats different models so u cant use rvc ones, but its way better for tts, else u can use any tts tools in our docs and use the generated tts audio as an input for rvc
Guys how do I get models from weights gg
@covert lake bro do you know how to get models from weights gg
are you making ai covers online or locally
I made them in ai cover gen webui
@wintry sinew bro I get a error when using a model in aicovergen
then, you can download them in your device, upload it in your google drive and make it public and put that link in the colab
what error is it
I messaged you
Thanks
I think he wanted something that had both TTS and the ability to create covers
For now, Ilaria Mainline has these two functions
hello
hi
Hi, is there a filter for voices that can be used commercially?
Hi
ye thats why i said he can use any tts tools,some rvc have that implemented like ilaria yea
sorry then
❤️
also pls dont use hearts with me
DM PER FAVORE
Has anyone actually found a darwin voice?
although I'm not in the mood to help because I'm starting to cry seriously
I cry over stupid things I know
but when you become attached to things or people
This?
Hello, I am new this server. I hope you are well.
@pliant solstice #1159290752195633273
but maybe it was a scam
then they would prob post it again
and then i would take action
no need to ping the owner
what does epochs mean ?
it means how many times the model has been trained
hey can someone help me set this up perchance?
Ayo? @elder willow level 1 !!! 

guys whats the best girl voice model that can be used in any language and sound like a girl
i have 2 that are kinda okay sometimes but its iffy
mind if i try?
i need something good to fool a pedo
ama try it ig.
and this
You can find it in our docs btw
where are those docs?
wait does it have spec requirements? and if so what are they
is it easier to run than the rvc/w-okada stuff?
For doing it locally yea, u need a good GPU which do u have?
Uhh its kinda hard to do it on colab as theres no guides for doing it online even if there are ways, but for locally theres a guide for it
an RTX 3050M (laptop)
Ram?
Oh only 4 gb of ram
Thats limited
I mean you could try locally but i think it wouldnt run with such small ram
Else you could use these colabs but theres no guide for them btw https://discord.com/channels/1159260121998827560/1229627925658337360
so like, limited in terms of waht?
In terms of power of your pc
Like low ram
yeah I know, but if I try to like run the program
what would happen?
(wait were you asking for vram?)
if it's normal ram then I have 16 gigabytes, 4 gigabytes of vram
Ohh
Well honestly you could try to run it yea dunno tho how much it needs as i cant do things locally
hello now i have my voice in pth how can i upload it to be a link?
Do y'all know how can i use udio with a real artist voice?
hi
hey guys question, am I the only one having trouble with step two "Download The Model" ? I'm using the link I found in "Voice-Models" and it seems to be running error each time I try to do it
forgot to say it was on the google colab
send the colab link
Ayo? @primal idol level 1 !!! 
you are using an outdated colab thats why
oh
use only the updated ones inside the cloud folder, inside the rvc folder of our docs https://docs.aihub.wtf
Last update: Mar 10, 2024
thank you !
Yw

wheres suggested execturos
Dawg who
@vernal ibex
Hey
HI
Hi
what's the best tts to use voice models on?
does anyone know how to use voice to ai with the ai models here?
when i convert i get error
its not necessary anymore to use UVR to separate the voice from the song ?
ye it is necessary wdym
can somebody help out with making a gpt sovits model?
it converted just using the voice I choose and the song from a youtube link
oh yea aicovergen uses uvr5 automatically lol
well need to do a few adjustment cause its not bad but its not clear enough
I got a question
What model would you recommend for using content creation without facing any allegations, kinda like vtubing but the ai would handle the talking
Ayo? @hoary wren level 1 !!! 
Like i don't wanna use a real person voice but then yeah every voice should come out of a human any ideas??
Create multiple voices, merge them, boom new voice
Okey this is literally genuis
Do you recommend any models tho
Like i am aiming for feminine voice
Some models are good at text to speech but ama be doing live pass through kinda thing like a voice changer
The best is creating multiple models yourself then merging them for most consistency, since your skills of creating a model will be similar
Else you can try to find any girl model tbh and just try around tbh
Aight then
Well, voice models that are made for realtime specifically are hard to come around. If its for infering (so you record everything with your normal voice, then convert that audio) is much more common to find
especially if youre going to create videos this is the easy path. If youre going to stream then yeahh best off creating your own voice models
Dm me if you wanna talk
Ahh okey i will try learning about training and see what i can do
Thanks a bunch
dm me
gotta be one the most obvious cat fishing accounts I've ever seen, ngl
lmao
What
real
new account, generic anime girl profile, she/her, "hi i'm 14"
I don’t know what to put
do you know what server you're on
Yeah
Ayo? @compact elbow level 1 !!! 
try elsewhere honey
I didnt do nothing wrong

please promo only in #1159290752195633273
hello
hi
istg so vits svc and kalo colabs gave me traumas
most errors are caused by people watching yt tuts
was it shitty or complicated
BOTH LOL
it was so much effort i gave up after a week of failing
dang 
lucky i didnt do any svc shit
if im not wrong theres svc 5 now
or still 4?
i saw someone talking abt svc using whisper and asking if it could make rvc better lol
hı
hı
ı dont know
turkey decided to be more special than the rest
oh lmao
Do you know the "brother ewww" meme?
LIKE THE VIDEO
Merch:
https://my-store-e85bb5.creator-spring.com
Business Email:
sponsormemezee@gmail.com
Twitter ► https://twitter.com/MemeZee_YT
Instagram ► https://www.instagram.com/memezee.yt/
TikTok► https://www.tiktok.com/@memezee_
Second Channel! ►
https://www.youtube.com/channel/UCa60-kUD9ppI46oaODMxIig
Anyway why im asking is
the way he pronounces ew is how you pronounce the i without the head
theres a meme video on that "i" letter but couldnt find it
lmao
someone give me a number, 1, 2, or 3
2
hello
myil from india
help someone knows why it sticks a lot and takes a long time to reproduce the voice.? help
lj
"Throughout Heaven and Earth, I alone am the honored one" Gojo Satoru declares, staring down the fraudxians. The goo furries sink back in fear as he utters the last words they'll ever hear... "Guess I'll be a little rough. Cursed Technique Amplification: Blue. Cursed Technique Reversal: Red. Hollow... Purple!" screamed The Strongest as he unleashed his strongest cursed technique upon them, instantly erasing all of them. The Gootraxians cornered the injured Nanami, Maki, Naobito and Megumi. As the furries tried to move in for the kill, Megumi opened his domain to attempt and create an escape route. As the hole in the domain was opened, a pair of hands reached out through it. "Those who inherited the curse of the Zen'in family, and the one who couldn't fully leave behind that curse. They would all bear witness to the bare flesh of the one who is free. To the one who left it all behind and his overwhelming intensity!" The Narrator states boldly as The Sorcerer Killer pulls up through the domain barrier, speed-blitzing Maki and stealing her Playful Cloud. The Gootraxians whimper in fear as they bear witness to the true Honored One as he slowly strikes them down with the cursed tool. As The Nightcrawler attempts Pursuit of the Sorcerer Killer, it gets struck on the back of the head. As the Nightcrawler looks behind to see the attacker, it was Toji Fushiguro, the Sorcerer Killer! The vessel of carnage struck the kobold one last time with Playful cloud, instantly killing it.
A recomposition made with Mario Paint Composer of the AI song "Heart on My Sleeve" by @officialghostwriter977, which was made originally with AI-generated voices of Drake and The Weeknd.
No copyright infringement intended, this video was made only for entertainment purposes. I don't own this, all rights reserved to their original owners.
Copyr...
Hi:')
I said stroke 40 guys
.
I really tried to get w okada running, followed tutorials on how to install PyTorch and Python and all that, but yet when I run the start_http all I get is just a command prompt
Please, if anyone has a tutorial from beginning to end or knows what the issue is, let me know
can anyone help me geta girl voice changer will pay, thanks
whats the best realtime voice changer????
hola me sale erro al abrir el w okada
Welp, I noticed it does do something when I keep the command prompt open, but then it shows me something for like a second before it just closes on me and nothing happens
can anyone help me with the voicechanger?
hola gente tengo un problema es que antes el w okada me anda bien pero ahora que actualize los drivers de mi tarjeta , luego al abrir el programa me salta un error y eso no me pasaba antes ;(
ahora pasare el problema que sale, podrian ayudarme bros
Ayo? @inner wraith level 1 !!! 
hello
how can I use this in a game for example mw3 I try playing with the voice changer in game chat it plays my game audio
I extracted a zip folder of program files for an rvc voice training thing and now my screen randomly goes black for a second. Thanks, AI Hub.
Ngl FNF trickys vocals are absolutely fantastic for heavy screams
a
I accept Paypal thanks
There should be settings
Ayo? @elder willow level 1 !!! 
Depending on what you use it’ll rather tell you or you’ll have to figure them out messing w em
Or that’s how it is for me atleast
yours work on discord? or any other app
mm
but the app likes, it gives you a 'microphone' like voicemod?
can isee your settings maybe?
It gives me a setting made for my app and then it tells you what to adjust it
can i dm you?
I don’t hv them rn I’m in my phone sorry
oh my bad
What are you using?
Alr
Anyone have any idea how people use AI to create covers of the instrumentals? Like this video here https://www.youtube.com/watch?v=n0K_jN72YaM ?
does anyone know how to make an ai peter griffin singing and make it sound like him?
all the ones i tried sound weird
anyone knows how the new model of yolo works?
Ayo? @near wing level 1 !!! 
yolov9
.
Hi
Hello
-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)
check our docs https://docs.aihub.wtf
Last update: Mar 10, 2024
https://rentry.co/VoiceChangerGuide
Anything else ask in #🔍│help-w-okada after having read through the guide
That sounds unrelated... I think thats caused by something else.
If you never unplugged your monitor before, then unplug your monitor cables for 10 minutes and plug it back in to reset it.
Any issues ask in #🔍│help-w-okada
guide:
https://rentry.co/VoiceChangerGuide
Its straight forward. If your command window opens then closes randomly, then write that in the help channel, ill send you the fix soon if you confirm its that
What are epochs
Last update: Mar 10, 2024
Oops lol not here 
someone join my online class
.wk
👑 takingmyswag - 2700 plays
2. enoch - 848 plays
3. osc - 696 plays
4. roybee - 337 plays
** 5.** gilbert - 244 plays
6. crows - 242 plays
7. praestone - 233 plays
8. adam - 211 plays
9. epilogue - 200 plays
10. proxy - 174 plays
11. Balakayy - 154 plays
12. uh1c - 132 plays
13. Baxter - 129 plays
14. qvsn - 119 plays
🤯😭
helloo
"Sayori’s best girl"
"Only reason why Monika isn’t first is because she has a mild case of Eldritch Madness, and I don’t quite want someone with that in my life."
"Especially since she ||ends the world at some point. Which kinda sucks.||"
wat
I need ai admires my foolz10 got disabled
?
you need 
just chat for a little more
I'm back!!! :3 
Boykisser is back
best voice changer?
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
First guide
e
.
e
e
I tried using voice changer on games, like Valorant, Lethal Company, etc, but they couldnt hear me. It was laggy or something, idk. It works in discord well tho. Im on 1060 6Gb, i7-6700k, anyone else having this?
graphics card won’t be able to handle playing games and realtime as the same time
unless you crank your delay up to like 5 seconds which is bad
chunk size up ⬆️
Yeah i thought it might be the issue tho
could always w-okada on colab but that’d still have delay
hi
hi
yeah it's not gonna be a pleasant experience tbh
I'd just recommend Colab in that case
1060 is already showing its age sadly 
hey
hi
can anyone provide me with gojo sensei's voice?
https://www.youtube.com/watch?v=tCTRwlHOyBg
just search on #1175430844685484042
Create with our AI tools for free. Generate AI voice covers, text-to-speech, and more. Join our community of creators sharing RVC and AI voice models.
E
Heyy
ello
I just wanted to ask if anyone can offer me some help
yeah sure, just head to the help channels
uh what
idk what i mean
yeah I can't understand either
elo monkeys
how can i download rvc gui?
hallo
hallo
any benson boone fans?
Hello my names Caylie
Last update: Mar 10, 2024
Hello guys!
Next to the Tyrannosaurus Rex we also have the mysterious fossil known as Frieren. It is believed that this ancient creature lived so long that the universe looked like a baby. It is known for its impeccable desire to convince everyone that believe it is only teenage years old, but with little success.
whats the best out of the three: applio, mainline or mangio
yo how do i use this on discord or roblox vc it doesnt show up
applio is most updated
everything works and works the same way, but applio is the only one keeping up to date
Hi chat
whenever i use virtual line 1 it doesnt pick up anything
what to do if its taking too long to process
Wait and hope
Chat
Which pretrained model should i use?
hi
Depends, what language u training and how long is the dataset?
JP and about 10 mins
Yes sir
Didn't test it myself but looks like the only pretrain for japanese
Yw
hi
What's up, everyone?
the sky
I don't get it.
Its a joke saying the sky is up lmao
taylor dearden ai where
it does not exist yet the model
beglua???
it's you???
Hey, I have a perfect original instrumental, but the Acapellas I want to replace have some chours effect (like doubler). What do u guys do?
(irony)
did you use uvr?
yes
model?
try cutting out the parts with the backing vocals
if you are trying to make a model
Ensemble BSROFORMER 97 + MelRoformerViperx + insth4
Nono, thers no backing vocals, it's the chours effect (like stacked vocals) I tried just getting the center but it really messes the quality
can someone pls gift me clickteam fusion on steam!!1
I'm doing a ban speedrun someone please ban me!
can someone pls gift me clickteam fusion on steam!!1
PLEASE BAN ME
PLEASE GIFT ME CLICKTEAM FUSION 2.5
I NEED CLICKTEAM FUSION 2.5
I NEED BAN
BAN ME
okay
I WON'T STOP UNTIL I'M BANNED
no i think they were talking to me
BAN ME
Ayo? @grim nymph level 1 !!! 
Why?
@covert lake 😭
wtf
who banned them before meee
@covert lake hey, im trying to use the minecraft skin ai locally but i get the ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt' error
ui or no ui?
how do i make it sound more real and less robot like
this happens when i run the install.bat
did you download the whole folder?
i clicked on code and then zip
you unzipped all of it right?
mhm
clickteam fusion sucks ass, use a real game engine
can you show me a screenshot of whats inside your Python_Scripts folder? there should be a requirements.txt
yeah but it's empty for me

lmao
wait
alright xD
the error still happens, huh
when i run it as an admin it says "python: can't open file 'C:\Windows\system32\Python_Scripts\Install.py': [Errno 2] No such file or directory" and when i open it as a normal user it says "
Traceback (most recent call last):
File "C:\Users\*****\Desktop\ai\Stable_Diffusion_Finetuned_Minecraft_Skin_Generator-main\Stable_Diffusion_Finetuned_Minecraft_Skin_Generator-main\Python_Scripts\Install.py", line 3, in <module>
subprocess.run(["pip", "install", "-r", "requirements.txt"], check=True)
File "C:\Users\******\AppData\Local\Programs\Python\Python312\Lib\subprocess.py", line 571, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['pip', 'install', '-r', 'requirements.txt']' returned non-zero exit status 1.
Press any key to continue . . .```
huh? did you download the new .zip and unzip it?
yeah
How do you do ai voice with a model downloaded?
"Yuri needs to be hospitalized.
Sayori just needs medication, therapy, and someone to give her lots of hugs.
I'll take the later."
fixed
there was a problem with finding it but now its in the right folder sorry for the issue and try to download the new zip
How do you do ai voice with a model downloaded?
Have the audio file of your song ready, & let's extract the vocals from it with an audio isolation software.
Ayo? @fallen ruin level 6 !!! 
yeah it works np
Hey is there a good (and possibly) safe way to get songs as mp3 files?
9convert
iTunes or Amazon Music 😁
@covert lake another error awmygad, i ran the ui script and put everything in, and then it crashes with the error
File "C:\Users\****\Desktop\ai\Stable_Diffusion_Finetuned_Minecraft_Skin_Generator-main\Stable_Diffusion_Finetuned_Minecraft_Skin_Generator-main\Python_Scripts\minecraft-skins-sdxl.py", line 2, in <module>
import diffusers
ModuleNotFoundError: No module named 'diffusers'```
Hey, do you guys know whats the discord command to put all the discords server on "Seen"?
Because i remember that exists, or existed
HOW
right clicking discord logo at top right
Ayo? @open helm level 1 !!! 
says mark all as read
It doesnt do anything
This?
Its an external plugin, Vencord
A
one sec ima make a venv version so there will not be errors
when its done ima tell u (5 mins prolly)
kay
whoever made Omni-Man, thank you!
give it a try with the new version
eh idk, depends on what u want
try out all of em
bruh
Ayo? @harsh verge level 1 !!! 
how can i make udio ai with kanye west's voice?
i heard a couple of people that did that
Ohhh you're so edgy based sigma, aren't you???!/1/11/??!?!?!
banned
bruh moment
bruh
no way, he was banned
i should have banned the second he said he was mark
whoever could've seen this coming
whos mark
markiwrench
who is that
prolly a 12 year old kid
Ayo? @elder willow level 1 !!! 
lets be frl
banned user
wdym kyc
so first off i'm very sorry the layout here's confuseing were can i search for voices for Voice ai i tried to find it on my own but to stupid/confused
it works but it's slow af
Create with our AI tools for free. Generate AI voice covers, text-to-speech, and more. Join our community of creators sharing RVC and AI voice models.
do you have a link
all i'm trying to do is just find a glamrock chica voice that's free.
beacuse i'm broke.
Anyone got the duckus egirl voice model?
Ayo? @hollow pewter level 1 !!! 
he just gave you one
No worries, search glamrock and a couple models are there
Does anyone know like a really good model maker or model that’s already made im trying to egirl voice troll in dahood and I was tryna sound like how duckus did in his video
can sombody make ronnie mcnutt voice
yea the venv makes it slower on the first start up as its a virtual enviroment where it fresh downloads all the requirements, it should we be faster booting the next times, btw whats your pc gpu ?
a GTX 1650 with 4 gigs of ram (but it has a great personality)
vram*
W great personality
but yea thats also the reason its slow while making the skin
if u want u can do it online for free btw
yea ik
Aye 1650 gang! ✋5️⃣
Where should I apply to make a module with voice samples. Because I can't do it myself
like requesting a model? you can use #1159289738314919936 or #1191429836321849435
hey all, does anyone know how to set the background of a project?
hey
hello
Any RVC working? i can't use any of them
The mainline version works perfectly
what lmao
Even for a technologically illiterate fool such as myself
- Applio, by IA Hispano Google 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
- Ilaria RVC, by thestingerx Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
any of these works
the most dummy proof one is https://huggingface.co/spaces/TheStinger/Ilaria_RVC
they always process infinitely for me
they don't
it takes a while for it to work first time
just give it a few
oka
you are a POOPY HEAD ❗❗ 🗣️ 😤 ❗❗ 🗣️ 😤
Ayo? @elder willow level 1 !!! 
no fetty wap AI model 😦
No fetty wap ai model...yet. be the change you wish to see in the world
I know you can make one. You have the ability inside of you. I believe in you
I'm trying rn lmaooo never done this before tho so it's a little confusing
Oh yeah I've never done it either. But if you follow the instructions in the guide you WILL succeed
you're right. i WILL. i got this.
Ayo? @pseudo heron level 1 !!! 
Plus you've got tons of experienced helpers here who are always willing to point you in the right direction if you get stuck
If you look at the server member list it will show you them. They'll be listed under the role "helpers". Also the #✨│ai-help channel will be a great asset for you
And a few of the members here have written some awesome guides in the #1159513888199540817 channel if you haven't already discovered them
There's truly no better to learn than right here in AI Hub!
Appreciate the help bro!!
Of course! I'm always happy to help out. I was a beginner once too! Feel free to ping me if you need anything.
StIcKiNg OuT Ur GyAtT FoR NeRiZzLeR
@wispy fulcrum free cover 🔥
bro is converting people into model makers 🔥
Fr lmao
Indomitable model maker spirit
Can never have too many of em!
He can do what others can not do
bout to make the greatest fetty wap ai of all time 🙏🏼🙏🏼🙏🏼
Tea?
Ay bro I got a qs
can I make a model from a diff lang?
TEA?
OMG TEAAAA
Ayo? @fair anvil level 1 !!! 
YES
Please, fix Applio Google Colab!
I don't see why not. I've seen lots of models in other languages.
DANG U GOT A BROTHER
Tea vs tea vs coffee
Whos the next one
Cola?
😭
If u got any issues pls ask in help channels
Allow me to introduce myself
Yo did u fight pepsi
Plot twist,, betrayal, you are going to feel the very, very, sharp feel of the most biggest, strongest, powerful ever tool created on discord.. not only.. THE WHOLE INTERNET.. DA BAN HAMMER
/j
Whos going to carry on my legacy of no life-ing the wokada help channel?
Smh ima hire poopmaster
Im gonna give him this
🦧
Real
Any opinions on Mangio RVC vs Applio RVC vs Original Github RVC? Like what's the gold standard
in regards to creating ai models
Mangio isnt really updated anymore tbh, you could either try original or applio which has some good features
Ooo yeah, I was wondering because I'm somewhat new to this, and the mangio fork looks like it has a easier to learn interface
Damn, what happened?
Ayo? @zinc spear level 1 !!! 
did creator discontinue it or did it get shut down by laws or?
The creator left that's all
No laws involved
I see
I'm looking for the highest quality fork, I'm assuming you would lean towards applio?
swsleri nerden buluyoz
With all honesty i dont know much of applio but they got alot of features like tts and overtraining detector if im not wrong
Btw you could also try Ilaria rvc Mainline if u want
I see, yeah don't worry about it, It probably doesn't matter too much in all essence, but i'll def check the sponser for ilaria's 
Just that ilaria mainline will prolly be discontinued soon but it got good things too like the training helper
I personally cant do things locally
When i used to do rvc things i used ilaria google colab
Ye fr
Its amazing for developers too
For sure
ilaria is an amazing person!
All people are amazing in their own unique way. Humanity is a truly beautiful and wonderful thing
🎙️ Realtime Voice Changer (Local)
Description: This one has more options and is somewhat more convenient to use than Real-time RVC.
Link: Rentry
Credits: ravencutie21
Description: More accurate than W-Okada's, but requires a decent GPU.
Link: Google Docs
Credits: teas.hi
Description: Text to speech with RVC
Link: Rentry
Hello
Let me breath air jesus 😵 Fr lmao.
touching time for anyone who pings me
yoooo!
if anyone needs beats i got u
for free
anyone knows how to see the spanish voices?
Nope, I've never made a single model. Never even tried lol. I've made a ton of song covers and I plan to eventually make a model of myself someday. But as of today I have yet to actually make a model
okay i was just wondering about something but if you've never done one before then i guess you probably wouldn't have an answer for me
i'm training it rn but idk if i did it right ig it's trial and error 😭
@fathom palm ASSALAMU ALEIKUM GUY
.wk track
where is zec
1189692177853452309
[9:50 PM]
972168218842116126
@wooden jolt
plz ban them
there trolling
they have invis user and pfp
Snitch
f u i got tired of them mic peaking
tyy my bro
i was jsut looking at suno, and is there any way to put my ai modles into suno?
Hello, I have a suggestion/request... Would it be possible to add a "gender" (or similar) filter to the #1175430844685484042 ?
Sometimes I need "male voices" (low register voices/deep voices) and other times I need "female voices" (high register voices/high-pitched voices), regardless of who owns the voices.
The "E-girl" filter has helped me a lot, but it doesn't cover all voices and there isn't a filter like "male voices". 😃
.
we have a tag limit
О
someone once told me a person made a supposed program to "clean" images contaminated by Nightshade. but it turns out it does not clean the poison, but just state it has or hasnt
is there other way to depoison images contaminated by nightshade?
is there a way besides the promo channel to see other people's covers on weights?
Ayo? @torn olive level 2 !!! 
hello there i have something to ask
how many samples do i need to make a voice model sound good, or the best it can. Can i request a model of a real person that is my friend, and is it free or do i need to pay the person whom makes my model for me. i just wanted to ask this, i hope someone can reply soon with a answer thank you.
Found out about viggle.ai , can you download and use locally for it or is it strictly online through browser?
why do you give a shit. they clearly didnt want you using their artwork. this makes you look like the asshole
LOOL
'guys this person doesent want their artwork stolen to train AI, how do i steal their art to train AI'
you can absolutely train a visual AI model, just use data that's morally correct and that you have permission to own!
Ayo? @uncut olive level 1 !!! 
usually paid requests are faster
photoshop denoiser
thank you, i will think about doing a request, thanks
hmmm, should try
you may also consider paying attention to copyright and moral guidelines 
Why does anyone care if their art gets used to train ai models? It's not like the ai model is breaking into their home, pistol whipping their children and running off into the night with the only copy. Idk maybe since I'm not an artist I just don't get it. Someone please enlighten me
because visual ai's only spit out existing data, they're incapable of creating any true data. they work by using a TON of data, and learning from that. but they can't create anything actually new
ai art that was trained illegally, is by definition plagiarism lmao
Is it printing out an exact 1:1 copy or is using it in conjunction with other pieces of data? If it's the former id agree it's wrong but if it's the latter then I don't think it's any different than sampling older songs or cutting out pictures from magazines to make a collage

sampling songs is fine because you're using actual creative processes to use them + the original song owner didnt make whatever you're sampling(usually)
artist's images posting on websites for FREE that people can access for FREE and right click and save for FREE. then its content that can be downloaded for free.
and people share it everywhere for free
made by the owner, but can be viewed and downloaded for free
and you still have to get it legally cleared lmao. you can't sample stuff and then release it on paid streaming platforms lmao


