#🧬│ai-chat
1 messages · Page 327 of 1
what sample rate is good to use?
are you training? You need to put the same one as your dataset
ok
That's why i ask about this...
well
people just do it here anyways
some people like ai voices, some VA dont, so idrk what to tell u
it would depend by person ig
I just want to be sure about this issue, if it is authorized, then I can use the models as best as possible 🙂
Ayo? @warm pumice level 1 !!! 
i dont know what you mean by “authorized”, its not illegal
Hmmm... 🤔
i dont understand this where should i import the vocal audio?
no ai models are authorized by original people
10 minute
You mean voice actors neither EN voice or JP voice...
I mean unless made by the person the model is not authorized
Hmmmm..... 🤔 that made me think for a long time
sorry
Oh u mean if people get perms beforemaking it
no absolutely not
Hmmmm....
u just need a good 5 min cleaned dataset and boom u make it
no one here asked the person before doing it

hey im trying to finetune text to text AI and now i wanted to fine tune text to speech or speech to speech AI and anybody knows how to do it becouse everytime i tried it failed and in the internet there are no good tutorials
I got a GeForce RTX 3060 in my pc
so…
are you trying to make a speech to speech model?
or smt else?
@latent copper i think i also already replied to u but u didnt reply
Text to text and speech to speech
speech to speech, use RVC
@covert lake I replied but a hour later
Can I fine tune it there with my own dataset
wdym finetune?
like the pretrain? Or make a model
for text to text ig u mean LLMs, this server is mostly just RVC & Wokada, maybe u could check the hf docs https://huggingface.co/docs
Like use my own voice put it in a format that the LLM understands it with an index and then fine tune it with that
More like making an own model
shi i havent seen dalang pelo in a bit u a real one fo dat
it doesn’t really work like that
llm and speech to speech are 2 different things
i dont really know how to do that
I’m pretty new so yep good to know
Ayo? @latent copper level 2 !!! 
ig the only way is if u mix an llm, and a text to speech program together
rvc is not really good for text to speech, gpt so vits is better
And what if I want to create a full own model like here in the voice models tab
Based on a character
follow #🧬│ai-chat message
Okey thank you
your welcome
hello everyone
define "text to text" and there are many LLM options to fine tune, some of them like GPT only provides the API
Like you text with chat gpt
that’s an LLM, large language model, so like a chatbot
You send him a text and he responds to it how he was trained to react and I want to fine tune models like ChatGPT with my own data
you cant directly make an llm speak
Yea correct a chatbot
Chatgpt you can't
Yea I tried it the last week pretty good and local
I know you can’t fine tune chat gpt but on hugging face the gpt2 model
hello can i ask a question? wich is the best free voice model for cahnging my voice from male to female if i dont care for specific type of voice /character only care to sound realist natural and non robotic?
gpt2 is soo old and sucks
Any options that are good ?
Llama 3.1 70b
Except chat gpt2 if you say so
Livetalking can with digital human and gpt sovits but setup is not as straightforward.
Will it work on my gpu ? I mean 70b is massive i think
that should work
Got 32Gb ram so I think 70b will not work
Yea that wil work for sure
GPT-2 is so old and you could even run it in an excel spreadsheet lol
for the 70b model of ollama you need like 64gb of ram
Ayo? @sweet bone level 1 !!! 
u sure?
Yea don’t have that only 32
anithing abo 7-8b is aout of reach for reagur consumer hardware
to run teh 400b model you need a 500.000€ worksattion
they did it on linus tech tips and still was slow xD
Yea I saw that crazy haha
as far as I know you can run 70b on 4090 fairly well
What gpu were they using
but still you gonna need tons of RAM compared to GPU on 70b
especially larger parameter like 400b
so... can anyone rec oment a godo voice model for cahnging male to female that soudn natural non robotic and good nto caring about with type of voice only my voce converted to femal
Yeah then add a Lora to train it
in english
Do I need to use an 70b mode or is lower okey too
They all support loras I'm pretty sure
Lower parameter is OK. Like hell not like you want 7 words per minutes generation
Yea true
So Like 20b is okey
I think
yea even a 11b vision version
I only knew that 8B and 70B version
всем ку
hello guys
hello moto
Hello
idk max limit time i need to upgrade to colab pro
yeah, so you got 300 epochs trained
can i use my model?
did you save the model every save? or just d/g?
-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
- 🆕 EasyGUI, by Rejekts 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.
anyway, you can extract a model from g
how to save a model?
there'a checkbox when you train
the moment when a 70b model can beat GPT-4o and Claude 3.5
https://www.techpowerup.com/forums/threads/nvidia-fine-tunes-llama3-1-model-to-beat-gpt-4o-and-claude-3-5-sonnet-with-only-70-billion-parameters.327796/
yea it means you finished your daily free gpu, you can:
- Use an alt google account
- Use kaggle which gives more gpu time but its harder
- Wait until tomorrow
- Pay for colab pro
I HIGHLY SUGGEST KAGGLE
i thought i had that set to auto save on colab
i have to train that again?
yup
i waited so long...
use kaggle next time
oh ig that could work too
how do i use g/d?
still better to do that in kaggle tho
As you dont got a good PC, its better you use cloud (remote good pc) 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):
Kaggle is way better than google colab
hey hey!
Ayo? @exotic bear level 1 !!! 
hey hey
cant i just model now? i dont care if its 300 epochs
that's a lot of hardware just to run e-girl eRP chat
doesn't really work like that, google colab free tier got AROUND 4 hours of gpu that IS NOT GRANTED, meaning that u can randomly get disconnected on colab
Kaggle gives u 30 hours weekly of better gpu granted
i hate how such hardware is just used for that
"eGiRl VoIceS"
i saved theese
i need to wait 2 or 3 days to get verified
for kaggle? u'd need to only verify your phone number in some mins
unless u got a problem with that and have to wait for support
i acidently useed lighning ai...
can i do anything with theese pictures i send
hello all its me barb
Ayo? @normal current level 1 !!! 
what?i get 1 thousand?
i mean, lightning.ai is good for not having restrictions, but kaggle gives more gpu
just use kaggle
u can still use the same number
Am i still muted?
is ai reaL?
i didnt ever saw it in real life and im scared it mighyt be a democrat conspiracy to stop trump from winning
0/10 ragebait
there's your model with e300
whether it is the best model or not, you can check on the tensorboard
he would still be able to check full tensorboard even of the epochs before if he uses G/D as 'pretrains'?
ask in #1159290752195633273 , THIS IS AN AI SERVER
maybe someone in promo is a designer if ur looking to hire, idk
nope
yea as i thought
tensorboard uses that tfevents file
what is the best free realtime voice changer now?
tensorboard can be installed on local RVC (whether it's mainline or applio), and no need GPU to run
wokada that uses rvc models, especially the fork that has better performance
what's ur pc gpu?
(or wthiout any rvc)
rtx 3080
just pip install tensorboard
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
1st link
yea, but i don't think he would be able to see the tensorboard of the previous 300 epochs he trained
imagine the model already overtrained in those 300 lol
you can just run tensorboard on that events file
he could as long he doesn't lose tfevents file
accidentally deletes the tfevents file 
docter martin scab wants to give me 1000 epochs if i call him
and cash app
HE IS GOING TO GRAND ME 3 WISHES
guys he want 10$ for 1000 epochs
isnt there any youtuber voice that is good as fuck and doesnt make long s sounds very robotic
like the allu
Where did you download the model from? If you've downloaded it from an unauthorized source, there's a high chance you'd get some virus or malware alongside with it.
wtf?? don't follow that guy
i banned him for scam before
i searched up DIO
what? Could u share where exactly u found it?
the first one i clicked then my laptop suddenly started randomly moving and searching random shit
in the #1175430844685484042
https://discord.com/channels/1159260121998827560/1257826077095755798 this is the first one that came up for me
yes
Ayo? @prisma cradle level 1 !!! 
dont use it..
where did you get your rvc?
ohhhh
from here??
im so glad i have a windows defender
did you download RVC (the program, not model) from https://docs.ai-hub.wtf or from a yt tut?
Last update: Mar 10, 2024
yt tut
never follow yt tuts about RVC. could u share which one did you follow?
and if you ever had any issues using that rvc with other models before?
i just tried it today
it wasnt the same model i was searching how to get a vc so i can like prank my friends..
cus technically HuggingFace detects if there are problems, and here there are none reported
bruh tf
brainrot 😭 🙏🏻
did you try any other model before that one?
no
Would W-Okada give even better results if I first filtered my own voice through OBS through with another virtual audio cable?
yea i feel like the issue is from where you got the program, could you share the youtube tutorial you followed?
oh that's not the RVC program, that's voice.ai, which is a scam as its paid and you can just use Wokada that is free and open source for realtime voice changing for calls with rvc models
other than being a stupid paywalled program, this is very weird as that shouldn't be a virus
could you tell me exactly what happened after u tried to import the model in voice.ai?
it started changing like my pc was offing and oning like becoming blank out of no where
what's offing and oning?
all good mate, do you still have the program or uninstalled it?
Ayo? @cobalt wren level 1 !!! 
i was tryna delete it then my defender deleted
Ayo? @prisma cradle level 2 !!! 
.
Oh? It seems like when you update a GPU driver on your device, your screen will flash for several seconds.
Did you let a chatbot to control your account?
just banned him, too much fast and replies in a good way randomly to people, feels like ai
💀
I think he was hacked by some random hacker weirdos who uses Linux. 
hol on imma delete voice ai it sounds so bad when u use a random voice
very weird, @chilly lake is there any way to check better a model safety? cus the model https://huggingface.co/Coolwowsocoolwow/Dio_Brando/resolve/main/Dio_Brando_v2.zip?download=true has been flagged as safe but there is someone saying their pc was acting sus after using it in voice.ai (following this), this sounds very weird ngl
Hey pals! Today I will show you how to use Voice.AI free AI Voice Changer in Discord Realtime with RVC AI models. Transform your voice into any Celebrity, VTuber, Anime Character or Cartoon hero.
Download Voice.ai: https://voice.ai/
Most popular RVC models: https://linktr.ee/rvc_bestvoices
#discord #voicechanger #aivoices #rvc #ai
so, using preset voices worked with no issues right? the only time ur pc did weird things was when using that model?
could you tell all the things it did?
did it search something specifically?
wdym?
you said after using the model, your pc turned off turned on then searched smt up
searched up?
The only time when your screen flickering is when your GPU driver is updating. I'm not sure what kind of malware would cause the screen to flicker like that.
i mean this
You meant Google?
flickered it randomly flickered when i was put it in the voice ai
Ok, so your screen was flickering. And while your screen was flickering, it randomly searching something on your browser, right?
Was your laptop pressing random keys or something?
Ayo? @solar torrent level 9 !!! 
Could you check and tell me what was the virus name when it was detected by Microsoft Defender? Because each virus and malware has its own payloads and targets, so I could make sure if the virus was related to all of this. 
Sorry for the long wait
i dont really know but i couldnt even type or do anything i was panicking
guys
it went blank n then 1 second later back
can someone helo me train model?
dio_brandov2zip.
an AI model have index file and pth file right
why i uploaded on gg drive and copy link and download it then it didn't show?
im trying to use a vc can u recommend me sum that isnt potentially dangerous
it is a regular model
No no, that's the infected file name, not the virus name. The virus name should starts with something like Win32/.
yea which is why i don't really understand of this virus
idk if it was a virus or my power going out but my windows defender said there is a virus detected
yes
are you following a youtube tutorial?
yes
because that's the worst thing you could do
:((
did you already train it btw?
Could u share me the yt tut u followed?
can you test my AI model
99% of tuts are 1 year old which are outdated
Here's how to train RVC voice models in Google Colab for FREE. No disconnect message! Clone anyone's voice, as long as you have their audio.
0:00 Intro to RVC
1:30 Colab disconnect message
1:54 RVC v2 Disconnected
5:15 Preparing the training audio
8:53 Feature extraction and training
15:09 Resuming training
16:12 Using the voice model in RVC
C...
i don't use that...ugh...magio rvc...whatever
i use applio colab
I don't take random tutorial videos from YouTube, I already have my own knowledge. 
but i think it was my trained model problem
that's what the youtube colab notebook shows uses
it uses a good notebook but its an old one showed in the video
this is regular weights
The AI Hub is one of the only place you can trust. 
WAIT
IS THE BOTS ONE SAFE sounds like a very stupid question but yeah
bro can you help me train AI model
this is potentially unsafe pickle
there are some functions and stuff
What do you mean by the bots one safe?
Last update: Apr 01, 2024
like the vc the bots create
but i feel like its just gonna be not so good as its trained without the tensorboard
i mean, time to test, u need to upload it first tho
wait
i sent u the link to upload ur model so i can test it
google drive link
https://drive.google.com/file/d/13Q4JvmT-d2djqO-n2Nmvh78vPfF1HV_H/view?usp=sharing
oh
btw, what's ur pc gpu? i don't want u to be one of those that have a good pc and use services like google colab that are cloud (remote good pc) computing
just...meh 1050ti
laptop
ah nvm then lol
As you dont got a good PC, its better you use cloud (remote good pc) 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):
I'd highly suggest kaggle mainline next time u train
PickleTensor has also been used for Stable Diffusion models, which there's a high chance it will be infected with some malicious code. Even CivitAI recommends you to switch up for Safetensors. 
it got wayyyyyyyyyyyyyyyyyyyyyy more gpu time than google colab
(ofc quality is the same)
lemme test rq
yea its safe
help me T_T
Ayo? @jagged geyser level 6 !!! 
spills diet coke on keyboard, wonders why random stuff gets pressed
no idea
problem between chair and keyboard
If Google Colab is on free tier, I now only use it to separate stems with demucs. 
are u testing
If your laptop is so dirty, I wouldn't be surprised.
uh how do u use wokada..
Ayo? @prisma cradle level 3 !!! 

why do you still use voice.ai if it's flickering lol
i already unistalled it
😭
yes wait a bit
kk
||sorry for the seizure-triggering gif
||
reported, banned.
💀
what's ur pc gpu
wait
throwed me an error
if you wonder, that my friend's voice he asked me to make his AI cover
might be bc of the special characters
howw
can you train for me T_T
Nope, takes time
if u want to do it urself for free i suggest u to take a look at #🧬│ai-chat message , and use kaggle mainline
48 GPU? What kind of GPU even is that? 
48 gpu??
yessir
You can check your pc gpu via:
ctrl+shift+esc (task manager) -> Performance tab -> GPU
48 gpu does not exist
There's only Geforce RTX 4080.
msi lap?
top
what? do you want to train hundred billion param LLM using 48 gpus?
i have a razer gaming lap
could u check the name at the top right of the gpu tab
48 GPUs are insane for some workstation tasks like machine learning and mining coins. 🔥
in the gpu tab u should have a name at the top right
um
unless u got a macbook, which i don't think u have tho
If your laptop runs Windows, in Task Manager, in Performance tab, there it should tell you something like:
GPU 1 Nvidia GeForce R... 1% (30 °C)
idk
you know what ignore what i just said
what i said was stupid as fuck and irrelevant
okay thats no good
NR-word as Cyrillic letters? Wow, that's too racist.
brain damage
Just go to a racism server just for that lil bro. 🥱
just make a circle jerk there you go
Ayo? @tacit falcon level 14 !!! 
yo guys
wierd laigh
im so sigma im not like you
spit your shit 🔥🗣️🗣️
Ideas Only Spread Because People Like Them, the "Free Market of Ideas" is not a Meritocracy, it's a Set of Unhealthy Coping Skills and a Corporate Oligarchy, Wake Up, Man! Don't Let the Lies Dim Your Light!
sex
HI everyone help me to make .pth and .index file from a database file (have 2 .wav files), please i not have good gpu and!
dataset*
and if your pc gpu is weak meaning u can't do it locally (on ur pc), u can just use cloud (remote good pc) instead
As you dont got a good PC, its better you use cloud (remote good pc) 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):
Google colab = ez but low gpu time
kaggle = harder but wayyyyy more gpu time, SUGGESTED
Yeah, but it sucks, I've already sent a message there, but then I have to wait for another 5 hours to send another one
yea you need to wait
no promotion is allowed outside of that channel
You know, you wanted to promote your channel. But please understand that any channel other than promo isn't where you promote your things. 
command prompt quickly opens and closes then calculator opens 
Yeah, I know, I have nothing to do so I'm bored, so I just sent that
censor the link in case someone have the genius idea to open it
sending malware because youre bored 😔
It's not even malware, I just coded it myself
you coded real esrgen? 🤝
One night, I dreamed of myself in a Ye West's listen party. There was long bass going on, but then I rapped N in Paris and the music suddenly changed into N in Paris. 

might be better
it seems pretty sus an .exe installer of realesrgan
No, I have made the program itself that uses realesrgan as a backbone of the code
LOL I had my first dream about Kanye West too
when you can run it on a web browser without installation
it was weird
he was with a bunch of other artists and they all looked up to me in a way
Well, I wanted a UI version
If I want an AI to go through the possible solutions for a situation with complex logic, what AI would be the smartest?
That was crazy. Was it a party or something?
yes
It was a small party, invite only
Took this image a few days ago
anybody know?
Ayo? @bronze lava level 1 !!! 
That's interesting, and of course, it wouldn't be a Diddy party anyway. 
foggy
Beautiful. 
hi ai hub
Is there a way to, perhaphs, combine the power of multiple AI for this ?
does anyone know how to download aristois in tlauncher
ill have to dm someone who speciallizes in AI
aint that shit a virus lol
Where is the voice model of the character Lyn from Fire Emblem and was it made?
You can search rvc ai voice models at:
- #1175430844685484042
- In #🔍│find-models :
- Send " @atomic vector search (name of the model)", without the ()
- Do /find with @hidden grotto
- 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/
IDC RAAAGh
i need to download it so i can destroy tha opps
the mine-opps
mood
dyk how to install it in tlauncher?
last time i used it was a year ago lmfao idk
like- broke minecraft
TLauncher contains some allegedly malwares. The Legacy Launcher, also known as the OG TLauncher, is simple and safe. I have them both installed on my Windows 11 hard drive. 
seek meteor client instead brother
this is an ai server
- sklauncher is safer
i mean i already installed it
but like- idk how to make it work
right shift doesnt work
yea ofc if u have a bomb, why throw it when u can keep it 
IDK HOW TO DELETE IT
people asking for advice in how to hack in lego game
Oh, this server is AI Hub, not Minecraft Hub lol. 
theres no alive minecraft server
and this is the closest there is to something computer related.. that wont kill me for talking about mc cheats
go in a roblox cheating server you'll probably get better LUCK LOl
also omg.. fumo pfp on tha big october... ainnoway
nuuuh im alr getting good luck here
how do i uistall sklauncher..
ai hub in 2025 bro lmao
can someone fill me in on the typhlosion meme rq
cant believe this shit is almost 2 years old now
Hawk tour (that thing). 
Here is the WAY to Delete TLauncher from your COMPUTER ! Easily !
Oil up. 
jawg.. nvm..
it doesnt support other versions...
im gonna hack on a modded smp..
2020
ooh my godd bro lmao
i need to destroy my classmates PLEASe
no you DONT LMAO JUST LEAVE THEM BE 
PLEASe
alr no hacking here
ai hub in 2025
GIVE ME THIS ONE PLEASE
ded and filled
which client to use for other minecraft versions
don't talk about hacking there
you do not need to destroy your classmates you need to get your gpa up lil bro
why would u ever ruin someone's else experience
they killed me and stole my shi
i need to get my winter arc+still water
that's good, skill issue honestly
nope, no hacking here
aww man
Cheat on your friend or girlfriend for revenge doesn't sound too good. 
Ayo? @solar torrent level 10 !!! 
NOO THEY RE MY OPPS
Ayo? @elder willow level 3 !!! 
then thats a skill issue 
why dont you focus on school or something other than cheating in lego game
focus on getting that gpa up lil bro
Yoo
wokada fork is the best
whats ur pc gpu?
did u figure out yet 😭
How to use AI Voice? How does it work? I just started using it for the first time🗿
dont use ai voice its VERY bad
Fr?
So....is there any program you recommend?
im asking the mods also 💀 🙏🏻
👽
-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
ai hub doc link is pbroken use https://docs.ai-hub.wtf/
Last update: Mar 10, 2024
are u talking about voice.ai?
don't use it
are you talking about realtime voice changing for calls or for pre-recorded audios
wtf? it doesnt appear if u ctrl+shift+esc?
it does
yea then go to the gpu tab
how do i do reailtime voice changer
what's ur pc gpu
idk
lemme see if i can show u one sec
i have a good pc tho
You can check your pc gpu via:
ctrl+shift+esc (task manager) -> Performance tab -> GPU
Yes
Ayo? @prisma cradle level 4 !!! 
are u on windows 11 or 10
11
gtx 1660 ti
gtx 1660 ti
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
1st link
tys mbro
do i get the nvidia download
Ayo? @unkempt ice level 1 !!! 
tell the names of both of them
yes
ok
i mean the name
of the gpu of both
the name should be top right, as u seen in the example "Intel(R).."
the other one? and also is 8.0gb the gpu memory of both?
wha
oh ig thats the other one
well, with a rtx 2050 u should be all good lol
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
1st link, the wokada fork
its the gpu 0
yea the rtx 2050 is good
in the guide, u have to read it
yes ofc
is that sarcasm
yes..
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
skibidi
'
hey frnd
anyone here to help me
can i run this modle in dubbing ai or other software
What is epoch is?
Epoch is the number of iterations performed to complete one full cycle of the dataset during training. You can learn more about it in the Applio Docs
Documentation for a high-quality, open-source speech conversion ecosystem designed for simplicity and optimized performance
But i cant find epoch in menu
Howw do I delete a voice model on w-okada though
PLS BRO SOMEONE HELP ME ITS BEEN fKING 1H ALREADY
😭 🙏🏻
@covert lake pls help 

@covert lake
What hpnd
Nvedia if you are using Nvedia gpu
it's literally in the guide 😭
Ayo? @prisma cradle level 5 !!! 
that's worse
there is no yt tut for wokada fork, + all normal wokada yt tuts are outdated
its better to just read the guide, it takes 10 mins
yes ik ive not even seen it
After downloading just extract it. And ask in #🔍│help-w-okada if you are unable do install it properly
It's simple
man i dont want to press a wrong button
Just press download button
.
i didnt see it
....................................................................................................................
wtf 2.5 gb
crazy
Maybe 3.31
Just see everything carefully. Do not rush
Read the guide and you will find everything
BTW we have different channel for it. #🔍│help-w-okada
@prisma cradle
It's totally depends on your specs
Ayo? @queen kernel level 26 !!! 
Your CPU, your GPU and Your RAM
oh
Yes.
is RTX2050 good?
The sound model I am looking for is not available, is there anyone who can do this?
⠀
Settings for Nvidia GPUs :nvidiagpu:
F0 Det.: rmvpe (suggested for all series)
RTX 40-series: 80-96 chunk | +16384 extra
RTX 30-series: 96-112 chunk | +16384 extra
RTX 20-series: 112-128 chunk | +16384 extra
GTX 16-series: 128-192 chunk | +8192 extra
GTX 10-series: 128-192 chunk | +8192 extra
Advanced Settings
Protocol : Sio or Rest
Crossfade: 4096 start 0.2 end 0.8
Trancate: 300
Silencefront: Off
Protect: 0.5
RVC Quality: Low
⠀
@prisma cradle
?
You. Can do it. It is very easy
how
The settings for your gpu
you can make it urself or #1191429836321849435 #1159289738314919936
-rvc
Suggestions for @white sky
- How to use RVC Mainline Colab by Cauthess
- AICoverGen Colab Guide by Eddy (Spanish Helper)
- Create a model with RVC disconnected (colab) by Angetyde
Read Ai hub docs
5 seconds left
@white sky
Okay. #🔍│help-w-okada
Msg here if you need any assistance. @icy pendant will help you.
done u
Hey anybody knows how i can create an index file and and own model onapplio?
Yes
is it a ai
-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
could you tell me how please
Read applio docs
A helper real person for Realtime voice changer
Just click on three lines and you will find everything you want
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
theres no zip
Ask in #🔍│help-w-okada
What you have Downloaded then
**Who should i ping for requesting models? **
either u #1159289738314919936 and wait, or contact model masters for paid commissions #1191429836321849435
@arctic tinsel Can you dm me the 48k and 40k beatzforge files?
im in the download center but what model should i use to remove backround noise there are options like:
vr arch, mdx-net, demucs
Ayo? @latent copper level 3 !!! 
im in ultimate vocal remover
like the doc said
?
im in the applio docs
how to create datasets and there it says in a sentence:
Now we just need to separate the vocals from the instrumental. If you don't know how to isolate audios, check the audio insolation guide.
and im in the audio insolation guide and there it says to use ultimate vocal remover
@river verge What is the best drum seperation model?
A model is the result of training on a dataset. You can learn more about it in the Applio Docs
Documentation for a high-quality, open-source speech conversion ecosystem designed for simplicity and optimized performance
umm idk im pretty new
hi chat
does anyone know prism launcher? i tried using it
and my fps is so shit
it goes 120 in tlauncher but only 20 in prism, why?
Render distance?
prism launcher shouldnt make your fps worse unless in tlauncher you had 30 billion optimization mods then that would explain why your fps is terrible in prism because you likely forgot to install the mods
Just install fabric and sodium and lithium (and iris if you use shaders), or just optifine
Anyone know an old senior woman voice model I can use? Im using it for a video
Ayo? @elder willow level 1 !!! 
its a forge server
Oh, idk then
Ayo? @real sentinel level 1 !!! 
when i played in forge servers i had really bad fps no matter what lol
yeah what with probably 30 mods and ridiculous ram usage lol
you can still try download optifine for forge tho, if its still being updated
I guess I can just use a old man and crank the pitch up a tad
Also maybe u have more ram allocated to the game in tlauncher
yeah but it can be insanely bad if you didnt allocate enough ram because then it has insane garbage collection times
and that causes ultra stutter or lag spike
Hi everyone !
hi
Ayo? @willow pasture level 1 !!! 
Tlauncher is considering Spyware by most people
There was a better way that I forgot the name of that used paid minecraft accounts but they had banned servers because other people used the accounts
that used paid minecraft accounts
Yeah
that seems sus
Yeah it was a minecraft account sharing launcher that had account system so you couldn't steal the accounts and had to verify every so often so someone didn't snatch the account
Hey idk why but my Appolo webinterface is not creating an index i checked it and its not there but its says in the webinterface that it has been succesfully created and i get this error then and im pretty new so idk if its okeys or bad:
Starting preprocess with 7 processes...
100%|████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:07<00:00, 3.82s/it]
Preprocess completed in 7.64 seconds on 00:03:45 seconds of audio.
Traceback (most recent call last):
File "D:\ApplioV3.2.6\rvc\train\train.py", line 84, in <module>
with open(config_save_path, "r") as f:
FileNotFoundError: [Errno 2] No such file or directory: 'D:\ApplioV3.2.6\logs\my-project\config.json'
An error occurred extracting the index: [WinError 3] Das System kann den angegebenen Pfad nicht finden: 'D:\ApplioV3.2.6\logs\my-project\v2_extracted'
If you are running this code in a virtual environment, make sure you have enough GPU available to generate the Index file.
An error occurred extracting the index: [WinError 3] Das System kann den angegebenen Pfad nicht finden: 'D:\ApplioV3.2.6\logs\my-project\v2_extracted'
If you are running this code in a virtual environment, make sure you have enough GPU available to generate the Index file.
An error occurred extracting the index: [WinError 3] Das System kann den angegebenen Pfad nicht finden: 'D:\ApplioV3.2.6\logs\my-project\v2_extracted'
If you are running this code in a virtual environment, make sure you have enough GPU available to generate the Index file.
An error occurred extracting the index: [WinError 3] Das System kann den angegebenen Pfad nicht finden: 'D:\ApplioV3.2.6\logs\my-project\v2_extracted'
If you are running this code in a virtual environment, make sure you have enough GPU available to generate the Index file.
i put my own dataset in and everything followed the docs fully but its not working
open model info .json
you did not run feature extraction dude
hm the docs didnt say anything about that where is that?
oh my fault thanks for the help 🙂
Dataset Processing - slices audio, does normalization, some filtering, creates two sliced audio folders
could you give me the link to the doc site where it is located please?
Documentation for a high-quality, open-source speech conversion ecosystem designed for simplicity and optimized performance
Local applio or what
is the extraction in the webui in the Train tab and then in Extract?
yea in the webUi
So local
yes
hello
Did you do pre compiled version
Ayo? @midnight pelican level 1 !!! 
idk this version im using this version currently
ApplioV3.2.6.zip
4.21 GB
LFS
3.2.6
20 days ago
the latest
Are you sure its not missing any files from the version on you download from huggingface
yea im shure
wait im unzipping the backup i got to check if i delelted something
is the Feature Extraction this?
https://imgur.com/a/y0WcVFa
Should I buy ChatGPT + ?
if you need it to go yes but if you just want to use it locally and private then just use a spare pc with an nvidea gpu and ollama and an ssd not an hdd
Ayo? @latent copper level 4 !!! 
Yeah if you actually would benefit from the better coding and stuff o1 preview has
o1 is still limited i think with 25messages a week but yea if he needs it to go
50 per day
So wait I can store it locally seriously but is ollama good as ChatGPT & can u train it further
No as good but does have Lora support
ahh okey crazy
yea you can finetune it with your own data then
and your family could use it like idk teach it with your monthly income and in what you currently give the money out and he can help you to save money or tell you what is nonsense and whats more important
and becouse its local nobody can look at it except you
@earnest dragon ?
Is there a way to get the entirety of quora locally? I know you can do so with Wiki but I was just wondering cause I think quora solves more questions than wiki
Ayo? @simple rain level 1 !!! 
i think yes if you finetune it with it then it should work
if you want to finetune it use trusted sources like: www.huggingface.com
thats one website i know thats trusted
poggggg
real
i was using it like 6+ months ago, then i started to get strange emails about my microsoft account, reinstalled my windows to win11 and everything is good, no one from brazil,china,turkey trying to log in into my microsoft account
I don't trust Gates so I un-installed windows
.
Why your phone does way more then windows does
are most of you guys here programmers?
i only understand JavaScript and TypeScript lol
yea that's called masochism
prolly not most, average user prolly not
no im just gay
Ayo? @real sentinel level 2 !!! 
damn I got some questions/ideas around some AI stuff but I feel like I can only get so far without a programmer explaining shit in more detail or telling me where things can and can’t be done
hey what if my model gets overtrained what should i do?
i read i should give the model then a new dataset so that the other one is away and is not boring anymore is that right?
whats up
hi
why doesnt it let me search for specific ai voices
you switch to rmvpe
wdym
Heyo
Crepe not as good
@timid musk can we get s mod in voice channel MAX 5 please
hyper_teodor4o and night_fury777 are doing bad stuff
crepe is slow. crepe-tiny is faster, but you lose precision, rmvpe is like 100x faster
how do you get rmvpe
Ayo? @hollow crescent level 1 !!! 
what specifically
nvm
Yup
Ayo? @pliant pollen level 1 !!! 
i only have dio pm harvest crepe and crepe-tiny
Thanks to mods!
how do i turn it on
they’re gone other way dw
let me fire it up on my end and look it up
can u send me tut in dms
depends on the program you're using
this too
Ayo? @mighty coral level 1 !!! 
rvc
which rvc? realtime voice converter?
yes
can u send me the link
so i can straight get it downloaded
what the heck is fcpe
it is in english
the screenshots in the readme are just examples
downloading it from the exact link i sent will give you the app in english
where do i dowlnoad it
updated the link
just extract it and run go-web.bat
assuming you want it for training and have a nvidia gpu
okie dokes
i want it to convert songs into ai characters
ah
it would be the same then
you just download your model from anywhere and put the .pth file in assets whereas you will find it's respective folder
and the index file in logs
I looked at how you did it, it seems so long and difficult that it would take days and it's beyond my ability to do it.
is there a voice changer better than rvc , i speak arabic and i find that it doesnt produce the letters voices as clearly as in english for example?
Thats not a voice changer problem, thats a voice model problem
Arabic is too diverse. Until someone makes an arabic pretrain, it wont be the best
Can someone help me setup the voice changer?
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
First is modified, second is og
which one is easier
2nd
hi
Guys can you tell me how to add voice models in applio??
Ayo? @hollow crescent level 2 !!! 
it was easier but laggy asf
unzip .pth and .index into logs folder, click refresh on UI
can you send me tutorial in dms
Cheese
guys is this safe or can it dmg my computer
it is safe
Ayo? @mighty coral level 2 !!! 
there's no reason this could damage your computer
it will put a lot of hard work on it, but as long as it's tidy clean and well ventilated for the components to take a big breath of air, it would be all good
ok ty
Still waiting for rvc3
Never it will be something randomly released by the Chinese ai community
How can I create a parody using the same voice and same music. I have my parody lyrics already written out
Best ai to leave insturmmetal best like you cnt almost not notice it is ai?
remember in the past if you had a bad pc but wanted to swap vocals of a song with an AI voice but you used Google notebooks? How do I do that locally through my GPU? I have a 3060 now and back then I had a GT 710
Ayo? @lusty storm level 4 !!! 
I can't find anything on Google except those nasty subscription based garbage websites
Hi, could someone help me with generating TTS with a custom model i made?
Omg
Does anyone want to verify my linkedin for $10?
How do I create a voice model with my voice to use in ai cover? Please anyone tell me the steps much appreciated
What
-rvc
Suggestions for @limpid silo
- How to use RVC Mainline Colab by Cauthess
- AICoverGen Colab Guide by Eddy (Spanish Helper)
- Create a model with RVC disconnected (colab) by Angetyde
Im not tea im FERN
can’t be it’s gotta be tea
Anime.
Isnt that girl a lolli?
i have my own parody lyrcis, is there an ai/website that uses the same singer and music to sing the song for me
WSG BOYS
oh my god dude
How many audio files do I have to add and how long?
android emojis
Ayo? @elder willow level 1 !!! 
guys gen question to the ai voice users here.. which is better Voice.ai or rvc client? (okada)
i think when you do your voices in real time, it sounds like garbage even in 4090.. i think prerecorded files does it much better
Try to change settings
already optimized
Yeah set settings to recommended and use a good model
Are you using 4090 ?
4080 Super
Ayo? @exotic ice level 1 !!! 
You said 4090 here
yeah because Ishowspeed got an rtx 4090. still sounds like ass when he tried it
Try to use another model and try to pitch up or down your voice. And use noice reduction
why blame gpu for ass quality models
no ass models. i download the ones with more positive reviews
hi
Hi
ooo ai pfp beautiful what model did u use
just one i got didn't use a ai
Ok
how do i use the voice models channel
Yeah because most streamers like him are not smart at setting stuff up they just have someone do it for them or they try and it sounds bad
It is not that bad as you are saying. It converts in realtime maybe it provides less accuracy. But it works fine
from what i remember, he watched this to set it up https://youtu.be/dZ_2HELnWJU
Remember when he created a beat in FL studio
Yeah YouTube instantly out of date I was using w-okada since w-okada added amd because I kept asking for it and also was the only one asking.
Yeah not that good
share yall rvc setups lemme copy..
lmao outdated videos keep screwing ppl
How do I create a voice model with my voice to use in ai cover? Please anyone tell me the steps much appreciated, and how much audio files you need and how long?
Ayo? @limpid silo level 1 !!! 
i have my own parody lyrcis, is there an ai/website that uses the same singer and music to sing the song for me
Ayo? @wraith cedar level 1 !!! 
Nah it's just streamers not being smart
also what version u guys use
prob both
Suno or udio
can i get a link


AI HUB Docs



