#🧬│ai-chat
1 messages · Page 313 of 1
Okay then probably it will get taken down from Steam because of the workshop with voices and stuff and that you are charging for the program.
ok thank u then could u tell me pls
Yeah you will make no money on this because people will use You can search rvc ai voice model at:
- #1175430844685484042
- #🔍│find-models
- https://weights.gg/ (login required)
- https://huggingface.co/models (but watch out cus in hugging face there arent only rvc ai voice models)
- https://applio.org/models
- https://voice-models.com/
- https://thevoicemodels.com/ (for Turkish Models, login required with discord and level 2 on their server)
if there isnt one, you can:
- #1159289738314919936
- #1191429836321849435
- make it yourself with our docs guides https://docs.aihub.wtf/essentials/how-to-make-voice-models/
And also people prefer the fork using web browser for the performance improvements and stuff.
you are not the target market we want to make it easier to use than clicking 7 times to download and selecting the path and index for each voice you import
Ayo? @compact olive level 1 !!! 
want to help people like this #🧬│ai-chat message
Yeah but the thing is you can easily implement the guide with easy click buttons and stuff as a app with chatgpt o1 mini
but how people will know about your fork?
Yeah if I felt like it I could make that with chatgpt o1 preview
Youtube and stuff and also discord because steam is a no for legal reasons.
but unfortunately it will not help those who have 650ti for example, and there will still be problems with lag in games (0.5-08s of delay) it's not realtime
yes not for now but the bet is the algorithm will get better overtime
Yeah and also the wasting money part not knowing if you can run it or not
Can refund on steam if you have under 2 hours no questions asked
Nah rvc3 and stuff is a while away
How will you be getting the models you will be putting on the workshop? Just yoink them from #1175430844685484042?
Yeah but the thing is how would amd and nvidia support work
for gaming it's better to use a bunch of 2 pc's, one of which will handle voice, but who has 2 pc's with rtx 3070+ video cards for 1 person?
Probably he will tell them to post on workshop and stuff instead of here and the other places
probably never will
Yeah using the server mode
yeah something else will come just not rvc3
There are a lot of free voices already why would people wanna pay lol
Isn't this just pay walled wokada?
yup
well the models are pay walled but the program isnt
ease of use = $$ (apparently)
Does wok know Abt it?
i dont think so
Oh I thought he sold the program too
Yup he said it's payed for the workshop access then will be free on github if you want that way
If I remember correctly, the plan was to have rvc 3 out by winter 2023-2024, but it didn't come out and there was a message that rvc 3 probably won't come out
I mean wok is in the server
Yeah rvc3 will probably be made by someone else and not be called rvc3
Isn't this just a rewritten UI version? It's not like gonna help any performance
He said optimization in the future and integrated virtual audio cable.
not performance wise but will be easier to use for someone less technical and integrated virtual audio cable
if, of course, someone wants to develop this area further
But what will be our incentive to post models there? I doubt we will get a cut of the money he gets and if its for publicity rvc is kinda dead.
August 2025 trust the process
I don't know absolutely no reason except your model will get taken down by steam and stuff.
we plan on rewarding popular model makers
By the way, yes, MIT requires attribution
With what
by getting a take down notice from steam?
Yeah and also a achievement your Model has been removed by steam
I think there is a way to make steam happy and everyone else happy look at wallpaper engine
Steam isn't really made for ai imo
No wallpaper engine is not someone's voice look at say chatgpt getting sued for using scarlet Johansons voice.
for now maybe something they want surf the wave of in the future
It will never except for games
Steam is a game store
I personally think performance is more important than being user friendly tbh
Steam is made for games, for ai there's other different platforms like huggingface
Not all gamers want to use stable diffusion or other ais lol
Yeah and he hasn't said how amd and nvidia cross compatibility on one version would work.
100%
It wouldn't be perfect at the start probably only nvidia compatible to start
Ayo? @compact olive level 2 !!! 
And also the virtual audio cable you implement will be trash and I would always stick with VB Matrix coconut the superior one.
Then even worse when you could have it giving to versions you can install for amd or nvidia
goal is to have this but wouldn't be the case at the start
At least that's good, I hate how some programs become pay-walled asf by others just because of mit license like in the case of kits.ai
how so?
and also will you ever get popular enough to make the initial 100 usd needed back from Steam no.
-25% of users
yep we are fine with this at the start
relatable
Okay so worse then non steam version and will it need internet always.
you read my mind
will it need always online to use.
we will have a program where we pay popular model maker $ depending on the popularity from the software sales
can i get 10%
yep you are not the target market at first give us 1-2 years if everything goes well then you might change your mind
can I get 100% of the money if I posted a model.
we have 1200+ steam wishlists so far in 9 days or so
1-2 years you will be 100% irrelevant and probably rvc4 or whatever will be a thing
we plan on implementing all the up and upcoming new algorithms as they come out
Okay look what happen to the first ai voice changer on steam
Will that happen before a takedown notice
do you have the name?
you didnt answer my question
what's your question?
can i get 10%
This seems like u are just gonna re upload the server models paidly which has no sense when people can come here to get them free
depends check us out
i did
There's already over thousands of free models
we have an open mindset we don't mind cross-posting every Steamwork submission everywhere
can't find it and there is already this https://balacoon.com/revoicelive/ on steam
You did and saw a landfill
i still want money
Ayo? @granite sail level 7 !!! 
im a business man
So you don't mind knowing people can get the same things you are making paid for free here?
i conduct business
Yeah but how much is the pay if it's like a dollar that's nothing
no we just want to make it easier for people
thank you looking into it
But you are re uploading server models that are free, paidly in your steam workshop
There's literally nothing good in that lol
Okay then on steam page change it so it says where to get the stuff for free if you are fine with it
10 cents is 10 cents
But I make more from discord then that from just helping people.
I think there is value in curation really hard to find the most reacted to models here
you ever thought of getting a secretary
There is literally no reason to reupload something that is originally free but for paid 🤦♂️
No because that's money and most just thinking of youtube then and promote it on my discord and stuff and see where it leads.
yeah but you will eventually need a secretary
It takes up typing to find here a free model it's not like a complex journey
and im that secretary
people pay to save time
Yeah the most liked ones are the egirl and vtuber ones and racists ones for some reason.
The time of typing? 😭
Okay someone here can use o1 preview and make you irrelevant.
This is just trying to make some money off free models lol
Yeah making money off others work and then giving them basically none of it
we are forcing no one to pay feel free to download the software from our repo and use any other voice library
Ayo? @compact olive level 3 !!! 
all the info about this should be posted somewhere, drip feeding info in a discord convo isn't ideal 
Then make that super well known on the Steam page.
Yeah wondering the review bomb this will get
true where would the best place for this be? most people don't read
The steam page for it
People are not gonna know those are free, and why would you still reupload others work paidly?
The part about re-uploading models paidly should be totally deleted as it doesn't really help people
It's not a reupload we will improve it and make it more intuitive with the audio cable integrated
Which audio cable
You said the only improvement of the program was the UI and vb cable so no performance change, and I was talking about models
it's a steam workshop will be community made not from us directly
What Virtual audio cable are you using
You just said it will be 5$ and that it will be re uploaded choosing specific models from here to save time, what are u saying now lol
work in progress which one do you recommend?
vb cable and stuff so you can get sued by them for using it in a paid software
Are you going to include models with it? Or are you just going to leave the default ones there?
don't know where they will be from yet just said there is value in curation sorry for the missunderstanding
and it's funny the screenshot has just normal w-okada not even a rebrand.
Steam page, discord, pretty much everywhere you can find the app, will clear up confusion and drama if people are informed about what the app is, what your goal is, what your future plans are etc, intriguing people and giving a valid reason for spending $5 for the app
yep place holders for now no trailer too yet
Yeah kinda funny
we want to be like voice.ai but good and costing 1/19 with the same model as wall paper engine
So you are scaming the uninformed instead of them coming here and getting the exact help they need for free.
So you want to be like trash.ai
The exact help and a better version of w-okada
People really want easy money lol
#🧬│ai-chat message still waiting for help
He literally got the bot command with the guides wdym?
He did are you blind
It's not hard to read a detailed guide
The whole problem about trying to be easy can be literally solved by a guide, which is why there are lots of people using the program already
It is for most people unfortunatly
Guys, I was trying to train a gpt so vits model, but, it gave me a error
Yeah and if I spent time with o1 preview I could probably make the guides way better
If they can't read a simple guide then they shouldn't be using ai
Unless they have problems with reading which I can understand, how hard can it be to read a guide?
If they have the TikTok attention span that can't read more than 5 words than they should stay on tiktok
imagine his userbase he is thinking is roblox kids and old people.
It's not like it has maths or any shit 😭 literally read step and do it
If u got issues just ask in help channel
Then done
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
/guides realtime
Prolly only kids as old people wouldn't even know Abt it lol
@meager nexus, your post was blocked due to a hyperlink, I don't know what AIs allow remixes to be made
True unless they heard it from the roblox grandchildren
Either the / command or -rt
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
where do you get the voice changer
RealTime voice changer? And what's up PC gpu btw?
-rt first link
Finally silence
E
Not a Gtag kid that's probably under 13
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
Ayo? @warm nymph level 1 !!! 
we will see give us 2 years 🫡
are you 13 or older
Nah we will see if it gets review bombed
Also guys, I'm making a new Rvc Model of Super Mari, And Torrie
what does index even help with
2 years of rewritting UI is crazy
Yeah probably get a couple ai coders and you are set for doing stuff quick
under 2 weeks to have a free demo on steam
2 years to be good enough to be used by deep AI people
So basically no use in 2 years because we will get way more efficient models like Beatricev2 but gpu and higher quality
we have hands it will be using the latest algorithm out in 2 years
Yeah but then at that point people would be using the free open-source way not a steam way
Which will be just re-uploaded wokada basically as even the steam images are just placeholders
Doesn't steam have a 100 usd posting fee
no with virtual audio cable integrated and UI but no steam workshop probably at this time
and what kind of VAC will be used?
not sure yet work in progress do you recommend any?
Ayo? @compact olive level 4 !!! 
All current ones if used will get you sued
I asked that before
most VACs have a propietary license
there is probably a paid one right?
can UVR extract the piano of a song?
All paid VACs have a propietary license and some free ones as well, if you use them there is a chance of getting sued
Supposedly yes, but you have to download a model to extract piano sounds to do that
U dont even know what vac to use and want to release a demo in less than 2 weeks
or the sounds of a melody
apparently htdemucs_6s is used for that
only LALALai and gaudiolab have this option
I'm not the dev
got it we will find a way we don't mind paying
Who is even the dev then
I honestly don't remember what model you need for this, but it's supposed to be
he is in our disc great friend of mine
What's this trash discord
you can find it in my bio
inside the link
Created using the w-okada fork link and Notebooklm
Wrong channel stop promoting and stop asking people to try it out. #1159290752195633273
@tight jackal no need to promo ur models here else ur promo messages will be deleted
Notebooklm did a good job with just the one link
hello
hi
We do not care
WE DO NOT CARE 😭😭🔥🔥🗣️🙅🙅
why bro ping me for ts 😭
i'm doing a group project brah 😭😭🙏
You should consume a L size bra
hello chat, new here
Dont care
hallo
@timid musk
A photorealistic portrait of a 23-year-old Italian girl beams at the camera, her long, wavy brown hair flowing elegantly around her diamond-shaped face. She wears a chic, casual long dress that highlights her figure, paired with beautiful earrings that add a touch of sophistication. The setting is a quaint Italian café, bathed in the soft glow of the evening sun, casting a warm light on her expressive brown eyes and creating a friendly, inviting vibe, full body shot, 8k hdr, high detailed, lot of details, high quality.
Here you go! Let me know if there's anything else you need.
https://files.shapes.inc/bd6d582c.png
Hi
are ther any beatrice V2 voices here?

is anyone on rn
can i get help on how to use this on roblo
roblox? i dont think its working
No
it doesnt work on it?
You look cringe
me?
bro chill out💀
Hi
hi
Whats best protocol
what
Oh sorry.. ok so im trying to download voice models. But when i click into the download link, the webpage just shows me "invalid username and password" pls help? I can't download any
what webpage
the download link will bring me to a webpage https://huggingface.co/Razer112/Og2/resolve/main/Og2.zip?download=true
Ayo? @fathom slate level 1 !!! 
and it just says "Invalid username or password" nothing downloads :/
the link doesn't work
who created the model
Razer
razer
UDP
ai hub
guys
The link is privated prolly
Hello everyone
laaaame
Hey

Fr
Also pinged u here #🤖│bots message
hi
hello
Va faire caca
Hi
Hello
hi
hello
Ya
accident
No
mb 🙏
No
me me me
what?
i love kissing boys
how can i make my own ai vc model?
same
Va faire pipi chez ton père


Bruh
who can send me a good model of a female voice? Please
Pls seek help
What is wrong with this world
id do yunli if kaggle decides to actually work
Phone no
?
make the yunli model
not for realtime, phones cant use realtime voice changers as they dont have virtual audio cables like pcs
hi
hi
hi
hi
hi
hello
Where to download AI HUB

I am writing from Turkey. I know Turkish. I am writing to you in English right now. I am the character Shelly, a character from Brawl Stars, in Lets'Go! I want him to say, can you help me?
?
What
I Want Brawl Stars Character Shelly's Voice to Say Lets'Go
Ayo? @elder willow level 1 !!! 
check out those 99 pings man
What voice changer u use
Ayo? @urban hatch level 1 !!! 
-rt first link
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
damn
What
Are you there? Is there anyone who can answer me?
ai hub is not a program...
if u need help ask in help channels
I Want to Make a Shelly Ai Cover
whats ur pc gpu?
Do you want an ai cover with a pre recorded audio or text to speech ?
Computer GPU AMD XR 480 8GB 256 BIT
whats ur pc gpu
I Am Shelly Character of Lets'Go! I want him to say
im not really sure if u can do it with that gpu, but For AMD GPU Locally on Windows you can read https://docs.applio.org/getting-started/installation#amd-gpu-support-windows
Or else if u don't wanna do it locally, u can use Ilaria RVC Zero
Also, both of those have Text To Speech
Table Of Contents Introduction (with website link) Model Loader (Download & Upload) Inference (use RVC AI Voice Models) Ilaria TTS Settings (Inference) Vocal Separator (UVR) Troubleshooting “No gpu is available for you for 60s” Introduction (with website link) Ilaria RVC Zero, is an RVC (Re...
Documentation for a high-quality, open-source speech conversion ecosystem designed for simplicity and optimized performance
is more stuff malware than safe
stuff we suggest here is safe dw
Thanks
thats the discord warning for literally any download links lol
hm ok
ye dw all safe here, we are a big community
you're welcome
wheres the http_start_bat
Ayo? @hearty sun level 1 !!! 
i only see http_start
yea
because 99% youtube vids are old
You can check your pc gpu via:
ctrl+shift+esc (task manager) -> Performance tab -> GPU
how much vram?
how do i see that
are you using a laptop ?
dedicated
?
is it 4gb of memory gpu?
Hello
0.6/2.0 GB
yea im not 100% sure it will be good, but you could try the wokada fork which has better optimization
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
its the first guide link
its a modified version to help performance like for low gpus
Where i get ai voices :>
not really sure tho
You can search rvc ai voice model at:
- #1175430844685484042
- #🔍│find-models
- https://weights.gg/ (login required)
- https://huggingface.co/models (but watch out cus in hugging face there arent only rvc ai voice models)
- https://applio.org/models
- https://voice-models.com/
- https://thevoicemodels.com/ (for Turkish Models, login required with discord and level 2 on their server)
if there isnt one, you can:
- #1159289738314919936
- #1191429836321849435
- make it yourself with our docs guides https://docs.aihub.wtf/essentials/how-to-make-voice-models/
Oh ok...
Ayo? @mild veldt level 1 !!! 
if u want u could try locally,
Else i can give you the cloud options (it runs the program on a remote good pc)
Ayo? @hearty sun level 2 !!! 
I just want girl voice to troll people >:)
For Realtime Voice Changing for Calls online (YOU CANT DO THIS ON MOBILE btw):
- Google Colabs (4 hours daily of free T4 gpu, easy to use, require only a google account) :
- How to use Original W-Okada's Voice Changer Google Colab
- Modified W-Okada's Voice Changer Google Colab
- Kaggles (30 hours weekly of better GPUs, T4x2 & P100, harder to use, requires an account and a phone number)
- Original W-Okada's Voice ChangerKaggle
- Modified W-Okada's Voice Changer Kaggle
Lol
Those are the cloud ways, i'd personally suggest the kaggle ways
kk
prob next time when i get more gb....
i dont wanna waste my last precious 13 gb

dw, with the cloud ways u don't have to download anything
it runs it on a remote good pc, doesnt install or run anything on urs
what does dw mean-
'don't worry'
ok
I dont know get thes voices
yea... im just gonna delete the voice modulator for now...
i sent you all the ways to find voice models, most good egirl models are paid tho
Does anyone know a good audio transcript in Arabic?
No paid :c
delete some games or files
Free ones pls
I'm trying to track a song in Arabic, tried to translate the English text but didn't work
Ayo? @mortal ingot level 1 !!! 
Im searching 2 hours for find great voice changers =/
search it urself in the sites i sent u
Thes?
yes one of the sites
where is?
Dataset - 21min
Pretrain - Snowie v3
u mean where is the pretrain ?
is this a voice version or a customization?
no, its a thing used for training (making models)
Table Of Contents Table Of Contents Introduction Types of Pretrains Where can i find pretrains? Index of the most famous public pretrains: Where to find and share other Pretrains How to use them locally: Non Applio/Other RVCs Users : Applio Users: How to use them online (Google Colab/Kaggle): RV...
Thank you
yw
m
hellooo A question, if I download a model to my cell phone, what program do I need?
I don't have a PC

Nothing because there is nothing for phone realtime
And if I get a PC, what program do I need?
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
And you need a good gpu
How much minimum ram???
128tb
Xd
16gb but 32gb or more recommended
Thank you very much, I'll try to get one :p
Ayo? @elder willow level 1 !!! 
-rvc
- How to use RVC Mainline Colab by Cauthess
- AICoverGen Colab Guide by Eddy (Spanish Helper)
- Create a model with RVC disconnected (colab) by Angetyde
Where is French Friarrhea
Halo i'm Noah
hello
what is MDX HQ4 on x-minus
that's probably an instrumental model
Does anyone here have a good V from murder drones voice model I can use
When I search for it I get nothing
https://www.weights.gg/
search "v murder drones"
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.
@graceful loom
Thanks g
Ayo? @graceful loom level 1 !!! 
True
Ayo? @rich palm level 1 !!! 
😢
😫
have questions hoe do you crete ai songs
@rich palm Please don't spam too much
Ai covers u mean?
yes
kk srry
i di
yessss
What's ur PC GPU?
im immobile
Then u have to use cloud ways (runs on a remote good PC)
Use Ilaria rvc zero
Ilaria RVC: CLICK HERE 🤗
Guide on how to use it: CLICK HERE 📝
Don't forget to thank Ilaria if you find it useful! 💖
is there any bangali voice models?
hello
Hello
Hey if i have a question regarding hardware andpython do i ask on this channel or somewhere else?
What's the question
Is there some workaround to use AMD GPU acceleration in PyTorch if the gpu doesnt support ROCm?
On what software because it depends on the software
u mean IDE?
Ayo? @dreamy temple level 1 !!! 
No software with pytorch do you want to use amd with.
i want to run LLM
Then if the llm doesn't support amd then you are out of luck
i mean torch itself seems to not support majority of AMD GPU's
Pytorch supports amd with pytorch dml unless something else is added that stops it from.
Never heard of pytorch dml, i'll check it out
What
Cheese
Cheese
Woah
Hallo meine fruendins.
Ayo? @thick rain level 1 !!! 
Hallo meine Freundind,
I recently got muted for 24 hours,
For something that was not right. I have read the rules clearly and why I got muted wasn't there.
Cheers
I wasn't disturbing the peace, being disrespectful, or harassing someone.
You should check on who and why you mute people.
Then why did you get mutted and don't say no reason
yeah shouldn't it have a note saying why you got muted or somethin
Yeah it should that's why I'm asking him to not lie
That's why I got muted
yeah if you did it more then once and stuff makes sense
Ayo? @earnest dragon level 27 !!! 
I did not do it more than once
yall what should i make an ai model of
I'm not sure but that seems annoying (though should have prob just warned instead)
your missing father
It's not possible for him because he probably doesn't have enough audio
🤪
how can i use the ai models? is there any collab link or smth
That's what I thought too! Not only that I am new to this server.
don’t send the paragraph but what did it consist of perhaps
The reason didn't state
Ayo? @thick rain level 2 !!! 
It was simply just writing a long paragraph
should’ve been a warm instead of a mute
I have the exact reasoning here
I cannot post my screenshot of the reason
More info on a forum I made.
any1 that can help me setup a vscode from unknowncheats
hey!
No not helping you with cheats.
need to be level 5 for embed perms and stuff.
@opaque root, I have found 5 results that match your search!
Hola
is there a ai that will extend my songs for free?
cause suno wont work with my sample
@paper blaze
Maybe udio, idk, don't know anything outside rvc 
dang, ok bro. thanks
your dog fused with your cat
made a whole album with my friends voice after I trained a model off of a discord call
Ayo? @minor raptor level 1 !!! 
I'm showing it to him tmrw lmao
feel free to share it in #1159290752195633273
I shall not because I wish to respect his privacy as I would wish would be done for me
well I don't force it if it won't be public tho
Hello i have a adcreative free trial will i be billed after the trial ends?
I'm not saying you are lol
I just wanna keep his likeness private
he can share it around if he wants but that's up to him
not a right place to ask such random paid services
i don't know where to ask. I don't wanna be charged i can't cancel it
nvm i think i figured it out
Wrong channel #1159290752195633273
Ayo? @worthy coyote level 9 !!! 
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
better (and free) alt to voiceai
is there girl voice?
dunno, look for them

Nah not one of you
:c
people make e-girl voices, look for them in #1175430844685484042 or weights.gg
people also do comms for them
check #1159290752195633273
ik get them weights.gg but i need place to use thes voices
^
better use this https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit
Table Of Contents Introduction (with website link) Model Loader (Download & Upload) Inference (use RVC AI Voice Models) Ilaria TTS Settings (Inference) Vocal Separator (UVR) Troubleshooting “No gpu is available for you for 60s” Introduction (with website link) Ilaria RVC Zero, is an RVC (Re...
its rickroll right on photo
Ayo? @mild veldt level 2 !!! 
it's legit seriously
anyone wanna try some ai vids?
can u just help me by downloading it im dumb
\
we made a small tool to generate short story based on genshin impact character, but my lv is too low to share img/vids
you can DM me if you're interested in
we dont care
that's online one, you can find the local ones in: https://docs.aihub.wtf/
Last update: Mar 10, 2024
thats not for realtime which is what i think he's looking for btw
i had to add that to the guide
i wonder how many people got rickrolled in that guide lol
not sure that I thought assumed rvc in general
still better than people falling into ridiculuously paywalled services and old youtube videos 
fr
theres alot of clickbaits too
yesterday someone showed me a fake yt vid of 'wokada on android' which is impossible
and the guy restricted out comments that say its fake
ыщ
so
aaaaaaaaaaaammmmmmmmmmmm duuuuuummmmmmbbbbbbbb
can someon add me to freinds and i on dm on call and someon says what to do?
i dowenloaded voice so where do i use it? 
Can someone help me? I installed everything correctly, both apps, but my voice doesn't come out normally or in DC. I activated the Cable but it doesn't work.
-Colab
- Applio, by IA Hispano Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- RVC Mainline, by Hina Google Colab
- AICoverGen-WebUI, by Hina Google Colab
- AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
- AICoverGen-NoWebUI [Spanish], by Eddy, Hina and Gdr Google Colab
- UVR5 NO UI, by Eddy Google Colab
- UVR5 UI, by Eddy Google Colab
- Modified W-Okada's Voice Changer, Google Colab
- 🆕 FaceFusion UI, by Nick088 Google Colab
- 🆕 FaceFusion NO UI, by Nick088 Google Colab
While the Colab free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.
-local
- 🍏 Applio, 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
- GPT-SoVITS, by RVC-Boss GitHub
Credits to Faze Masta and Antasma for compiling these links.
chat
we see many sites of generative music models, is there any that can be locally ran?
Nope.
CRAWLING IN MY SKINNNNN
hi
hi
hi
@golden brook Estou usando um tradutor. Excluí sua postagem em #1159516963014451302 porque esse é um canal para sugestões reais de servidores, não para modelos.
Para solicitar um modelo, use #1159289738314919936 ou #1191429836321849435
What software do you use to change the voices? the models I have already downloaded
man i lost 100 epochs of training because i was greedy
"oh i only need x more epochs then i'll save" then bam, gpu runtime disconnect
You lost it because you finished ur gpu quota right?
if u want, u can use kaggle instead which is harder but gives 30 hours weekly
realtime for calls?
and whats ur pc gpu
for calls, I want one that looks very realistic, I downloaded one but the voice is very robotic
Rtx 3070
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
use the first link, the wokada fork
that also can depend by voice model, u can try a different one and see which is good
do you have a link to that?
All training cloud options are:
- Google Colabs (4 hours of daily gpu for free, not much hours, but easy to use):
- Applio (ui)
- Mainline (UI)
- RVCDISCONNECTED (no ui)
- Kaggles (a bit harder to use and needs phone number but gives 30 hours weekly of better gpus):
- Mainline (UI)
- Applio by Vidal (UI)
- Applio by Shirou (UI, no guide as of right now)
- Lightning.ai (Kinda hard, needs login, no issue with web uis or anything, but only free 15 credits monthly):
- Mainline (UI, No guide as of right now)
- Applio (UI, No guide as of right now)
I'd suggest one of the kaggles
very grateful
Ayo? @empty oasis level 1 !!! 
👌
I DOWENLOADED VOICE WHERE DO I USE IT!?

guys!

hi everyone
"Hey everyone, my account was hacked a few minutes ago, but I've sorted it out now. If you received any links from me in this server or through DMs, please don't click on them as I’m unsure where they lead or what they might do. I apologize for any inconvenience and hope you all have a great rest of your day!"
I think is fixed
so if you get a link message or it pops up here pls dont tap it
hello
😎👍
whats ur pc gpu?
yes, ai requires good computing
you can't train (make) locally (on ur pc) with that gpu
You have to do it on cloud (using a remote good pc)
#usefullinks
As you dont got a good PC, its better you use cloud for training an RVC Voice Model:
- Google Colabs (4 hours of daily gpu for free, not much hours, but easy to use):
- Applio (ui)
- Mainline (UI)
- RVCDISCONNECTED (no ui)
- Kaggles (a bit harder to use and needs phone number but gives 30 hours weekly of better gpus):
- Mainline (UI)
- Applio by Vidal (UI)
- Applio by Shirou (UI, no guide as of right now)
- Lightning.ai (Kinda hard, needs login, no issue with web uis or anything, but only free 15 credits monthly):
- Mainline (UI, No guide as of right now)
- Applio (UI, No guide as of right now)
Personally i suggest kaggle
#usefulinks
ai
yes, i explained u ur pc is not good enough so u have to do it on cloud, the best service is kaggle which gives more time of gpu for free but its a bit harder to use
so like i wanna make a ty dolla sign
how can i do it
like the voice and that shit
- Make a good dataset
- Use one of the cloud services i told u, most of them have guides, i'd suggest you to use kaggle mainline, and be sure to be using the tensorboard
Last update: Mar 8, 2024
hi
where can i put the dataset
first of all make one, then it depends on which cloud service u are using, its all written in the guides
it can still actually run, but 4 GB vram, plus lack of tensor cores would mostly bottleneck the performance, like playing games in 12 fps all the time
what is the name of the voice changer app ?
what would i need to do to resume training my model from colab on kaggle
Ayo? @gusty vessel level 2 !!! 
I have 10min of audio that i want to train how many epoch should i do and should i increase the batchsize of 8?
Hi guys, I need some help, and I was wondering if you could help me out...
1/3 cup cocoa powder
1 1/2 tsp instant coffee granules
1/2 cup + 2 tbsp boiling water
2 oz unsweetened chocolate
1 stick melted butter
1/2 cup + 2 tbsp vegetable oil
2 eggs + 2 yolks
2 tsp vanilla extract
2 cups granulated sugar
1 3/4 cup all-purpose flour
Salt (to taste)
Cacao nibs (optional)
Chocolate chips (optional)
thanks helped me out alot
Ayo? @exotic thorn level 1 !!! 
@heady venture beni neden disladin ki 😭
What's the link to the hugging face voice changer
Hey Y'all Andy from London here newbie to this server
hi chat
Guessing that you are using the mainline kaggle https://rentry.co/RVC-Mainline-Kaggle#resuming-your-training-session
This guide for Mainline Kaggle is an alternative option to the Mainline Colab notebook for training voice models
It is complete and should walk you through every step of the way since Kaggle has a difficult learning curve. However, it will be updated constantly to go over parts that need more cla...
im kinda outdated on the topic, but whats the best tutorial for making a model?
and also which rcv is better now?
should i do mangio-crepes or not?
Yo chat
what
is it that much worth it?
mine is 44 what do i do?
40k
ok thanks
sorry if im annoying u but what do u suggest?
v2
newer architecture
rvc v3 wen
is this right?
rmvpe better
okk
I could ask someone else to train the voice model for me, but who in here would be able to train a 🇫🇮 Finnish voice model?
And preferably for voice-to-voice purposes, not tt
this is what i got while installing tensorboard
Sup
i didn't have to format. it just worked
shadow fight 2 save game is still good guys, some day i will finish the game
yw ❤️
im having trouble installing tensorboard could i have some help?
1
before making the cover, you have to train the voice (make it), whats ur pc gpu btw
Ayo? @elder willow level 1 !!! 
训练结束, 您可查看控制台训练日志或实验文件夹下的train.log
Ayo? @unkempt bluff level 6 !!! 
what is this error? i just started the training
that depends, training could be an hour or more, u need to check with the tensorboard
try asking #✨│ai-help
tensorboard is a tool to check if ur training is going well
RuntimeError: CUDA error: an illegal instruction was encountered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
u can see it in the guide
does someone know the fix to this?
try asking in #✨│ai-help or #1192011222023950368
i did but nobody answers tbh
I dunno about your error but if u are using mangio rvc i'd suggest u more updated things like applio or mainline
no im using rmvpe
i mean the program not pitch extraction
ah mb
Hello chat
do u happen to know how to make the gpu use less memory cuz i think the problem is that mine is running out of memory
what gpu do you have
4070 laptop
it says 8 gb dedicated and 16 shared
RuntimeError: CUDA error: unspecified launch failure
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
i keep getting this error
What gpu and what ai software
is it during traning?
4070 laptop and using mangio rvc
yes it starts makes the epoch 1 but doesnt make progress it seems
and that should solve my problem too?
think so
No one use Mangio just use applio
mainline is better than applio
what's the difference between rvmpe+ and fcpe
How is it better
faster, more reliable
What more reliable with it
less problems
right?
But the thing is less problems means nothing if problems are not public.
yes you can use either of them, normal runs on CPU, GPU runs on................. GPU
couldsome one remove vocals from a song for me my laptop is too trash
What's the difference between rvmpe+ and fcpe
everyone with a bit of knowledge knows mainline is still unbeaten
rmvpe+ is just a slightly modification of fcpe, fcpe should be used only for realtime but rmvpe is still better quality-wise
fcpe is just a tiny bit faster
how do i remove vocals
UVR locally or online
my laptop too trash for uvr
Basically the same except for hybrid and the audiosr it has
use the online one
mvsep
?
?
hybrid trash + audiosr is garbage
it doesnt let me train the features index
did you put the dataset into the dataset folder?
How is hybrid trash
me no likey
can someone do a uvr extraction for me
no we cant otherwise we will be struck by copyright
its not a copyright song
what is it then
its a underground rapper
Yeah I wish the better hybrid was a thing but the fork is not update to date that has it.
still copyright 
hows it copyright if theyre not signed to a label\
fuck im so dumb
still mainline is the go-to for training and stuff
Copyright it themselves
Okay whatever people prefer
because otherwise i can simply take it and claim it as mine
meh.. fcpe should NOT be used for training
No they copyrighted it themselves most likely
Files in Queue: 378
Your order: 349
No I know for non realtime conversions not training.
RuntimeError: CUDA error: an illegal instruction was encountered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
i still got an error from mainline
wtf
my ai voice in discord doesnt work and the ai voice lags alot
hello
Which voice changer.
Voice changer client demo
hi guys i want to ask a question about ai security and data governace where can i go to do that
its been 8 minutes and im still at epoch 1, is this normal?
you mean w-okada or what.
The ai voice changer client
Ayo? @worthy ledge level 1 !!! 
show me in #🔍│help-w-okada
is it normal that it takes 8.30 minutes for 1 epoch with a database of 33 minutes and a 4070? i think something is not working, also tensorboard doesnt work so when should i stop approx?
sounds like a normal time for 33 mins methinks
ah alr didnt know
but tensorboard makes the whole training crash, u know a decent stopping time ?
My vb audio virtual cable doesn’t work in discord
theres a RVC V2 now ?
did you restart your pc?
Yes
did you followed the guide?
Yes
Send screenshot here #🔍│help-w-okada
@frosty sail, I have found 1 results that match your search!
AW HELL NAH
I need to ask someone if the CEO/founder of a cybersecurity company which is based and operates in the us can live abroad
Anyone know what lawyer I can ask this question to
We avoid lawyers here in ai hub
i cant send but voice changer client demo or something MMVCServerSIO
Follow that guide I linked, it is the best and most up to date version of mmvcserver
What GPU you using btw?
Yeah you shouldn't have issues
That guide is the best, most information packed thing we have
thats a differenct ai voice changer tho
Ayo? @unborn lily level 1 !!! 
Its a lil overwhelming but spend some time with it and youll get it running
Its still MMVCServer (w-okada)
Just an updated version from someone else. Its a fork
Works with all the RVC models found in here
perhaps wrong gpu? lets continue in #🔍│help-w-okada
Ayo? @vast sluice level 1 !!! 
NO WAY :0
auh~
what's the point of this? making a model for website rating?
it should be fine to run, and still better than the 4 gb and 6 gb variant lol
sound dogsghit thoo
no this doc I checked it's for ranking ai discords and stuff and it's trash no ai hub and stuff on there
prob for his school/uni assignment 
but isn't chipsandcheese.com a hardware site, while most others are ai related?
it's not his and rankings are not acurate.
because of people's bias and subjectivity
help
Ayo? @latent valve level 2 !!! 
whats that website to strip vocals i forgot the name
Mvsep?
omg no but thats even better thankss
yuh so i was using that but it said my ip made too many requests weird af
@tight jackal
thing is you can train them and test to see what sounds better but you don’t have to post every single model you make
regarding the character
and training method
Okay
Also, I made a rvc model of Fand (Fand Leonard) which is my best friend I met at school
dawg 😭
What's wrong?
Yeah he did it again and also that model he just said is a model of a kid so that is weird.
figured
She likes that I made a model of her voice
Okay
not even surprised atp
also what’s the difference between these two
like genuinely what’s the difference
I basically retrained them from the original episodes and from the Shorts
I feel like he should get his model maker role removed because he keeps posting about them in general.
lmao
But, I'm a model maker now
But do you really deserve it
Yes
Why if your models are a kid and then 2 of the same one.
i think i like cheese ngl
Ayo? @surreal cave level 1 !!! 
ok thanks
shut
np
yea but why do you keep posting models of the same voice? that seems sus
I say we ban his ass
Ayo? @wide cargo level 14 !!! 
AI HUB Docs
