#🧬│ai-chat
1 messages · Page 133 of 1
kits is ded
it's not dead.
this is taking a while
but i'll see what it will give me
oshawott?? 
Kits is not supported under Ai Hub and they have eliminated free model creation and most free model usage.
Just use voicemy.ai or find another method.
-colab @young pawn
Suggestion for @young pawn
- AICoverGen-WebUI, modded by Hina Google Colab
- AICoverGen-NoWebUI, by Ardha Google Colab
- RVC Disconnected, Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HF, by r3gf Huggingface Spaces
- AICoverGen, by r3gf Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
- Advanced RVC Inference, by neuclya Huggingface Spaces
Credits to Faze Masta and Antasma for compiling these links.
You can find more info on the #1159513888199540817 channel. If you can't find your answer, feel free to ask for help in #✨│ai-help.
I tried. Error 500.
My Colab files do work with Kits.AI's uploading.
😋
(they're .pth)
no shit
Yeah not everyone knows surprisingly
whatre we talking about
Voice training was the main topic, but as for format this person just replied no shit as if everyone knows Google colab makes pth files
hi;)
Hello
Hello, young lady
Yeah, and its poor quality because kits ruins all rvc models.
I just uploaded it to Google drive and copied the link and it worked fine for my voice model
Unless we're talking about a different thing
and they should not bother because kits is trash
Isn't that just a site to upload your zip of the files of the model anyways?
hi
hi
ai noobs
Umg keep deleting ai covers I got 2 copyright strikes
anyone know safe place to upload ai covers
how long does it take to train a model to 500 epochs with a 15 minute dataset?
depends on gpu
if im correct i have mine set at 500 epochs but have atleast 4hrs of audio, but with my 3070ti, it’s avg 40-50seconds a epoch
which would take 6.25hrs to do at 45 seconds a epoch
3070ti
with my settings atleast
what is ti?
what does this mean?
it’s the model of gpu😭
you good bro
there’s a 3070, and 3070ti
ti is better than the base model
they are usually a bit faster
Mine's 10s per epochs with a 4070 
i’m at 8 batch size rn
idk if changing that would change the time
Depends on your VRAM.
3070tis are all 8gb vram.
With 8GB, I'd try reducing it a little bit, maybe to 6.
matter fact i think all 3070s are
Mine's 12GB and I usually go with 8 and it runs like a dream ✨
i’m running it perfectly fine
Nah I mean training speeds.
yea
10s per epoch 
Sometimes too much batch can "overwhelm" the AI and it gives it too much data which it doesn't know what to do with, then inevitably causing slowdowns with training.
Give 6 a try for batch size and see if it does anything better, if not then oh well 🙂
Experimentation with training is always fun.
i’m too lazy to get out of bed rn tbh
lmao
i’m tempted to buy a 4080 or 4090
but spending one paycheck on a gpu is something i don’t want to do
I'd say just go for a used 3090 maybe.
That way it won't take everything out of your wallet.
Yeah but 24GB of VRAM 
Over here they go for about.. £650.
Which is just over 1K CAD
Some even lower.
1800 for a 4080 which isn’t all bad ig
803 CAD for another listing I saw.
If I really wanted to in that case, I could drop a month's on a 3090 and still have a bit left but.. I don't need to because 4070 is still something I am really happy with.
Massive upgrade over the 1060 3GB I used to have.
i use to run with 5700g igpu
never did ai or any of this tho
I used to have to use Colab because my GPU just didn't do shit lol.
played every game i have
So-Vits collab LMAO
never tried anything but local
I've never used Colab RVC, only local myself so I get it.
3070ti is still a solid card
Of course.
That's why I say just to try out experimenting with your batch sizes, surely you should be able to get something quicker than 45s/e.
hey guys
This is very random but do any of u here do computer science or programming
43/e
somewhat
ik ai is kinda related to that right? Anyways how do I learn how to do it
bc I’m going to college next year and I might wanna do computer science but I’m scared bc I heard it’s hard
i’m self taught somewhat
So I wanna be prepared
it’s not really
I’m just confused on like what the shit even means
I’m not good at math but I wanna do it
depends, like i found modding unity games to learn c# was better then videos cause i had to actually troubleshoot myself, learn what to do and how
but can be different for each language
like html, or css which is extremely easy to learn if you put in the time
true i am doing an app and you can’t really trouble shoot
I just read online and they said python should be FIRDT
First
python was my first
I literally know nothing abt it like I just want money and I love computers
but didn’t last long, ended up moving languages
I think the college I’m going to uses c+++ or whatever it’s called
c++
Like i said I literally have 0 IDEA how it works
one of the harder languages
but ik it makes some $$$
it’s not entirely hard but confusing for me atleast
even if you like find a project on github that’s broken, fix it without a youtube video
like I’m telling you I know nothing, what should I do? Like should I buy a book and go off of that
but like i said i DONT know anything…
research.
ahhh
literally have gpt, google
like with troubleshooting it’s not all too hard becuz you could just paste the error in google
Ayo? @novel moon level 5 !!! 
and stack overflow will probably have someone with something similar or what you need
stackoverflow is amazing
pth files ARE the voice models. literally EVERYONE SHOULD KNOW. and also colab doesn't "make it" btw
Then how do you explain the pth I have of my own voice FROM COLAB
I learned programming from uni and lot of programming languages from tutorial and documentations. C#, python and javascript (node) are my main.
have made some simple games using Unity, and also modded a little RVC code for better QoL for myself thanks to my python skill.
hello
HOW TO USE AI VOICES
Ayo? @elder willow level 2 !!! 
What Song are you doing lol
guys how to gett AI voices
Oh okay sorry I didnt understand. Yeah I believe the modle i used only had 10 minutes of data
Let me know when its ready to use I will try it out. The one i used had some stutters in it so I had to cut parts out of it
Ayo? @crystal pecan level 1 !!! 
ACCIDENT
Ayo? @timid blade level 1 !!! 
.
HI
hello
@hollow tangle
!ban 779118688585777182 Porn emojis spam
@waves4053 banned
Reason: Porn emojis spam
Duration: Permanent
thank you so much for the ping
np me was looking for rvc v2 link til i saw these emojis
always for free to ping me for any problems, thanks again! <3
what is the best RVC to make models? I see there are a bunch of RVC colabs
Люблю пиписьки
.
Sooo does anyone know what the index files do?
All of them have the same tech, just different ways of doing stuff. I'd recommend using RVC Disconnected or EasyGUI
-colab
- AICoverGen-WebUI, modded by Hina Google Colab
- AICoverGen-NoWebUI, by Ardha Google Colab
- RVC Disconnected, Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HF, by r3gf Huggingface Spaces
- AICoverGen, by r3gf Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
- Advanced RVC Inference, by neuclya Huggingface Spaces
Credits to Faze Masta and Antasma for compiling these links.
You can find more info on the #1159513888199540817 channel. If you can't find your answer, feel free to ask for help in #✨│ai-help.
Index is basically the mannerisms of speech being applied in the final result, like accents, how they say certain letters, etc.
I see, that makes sense as to why that would result in a larger file than a pth. Thanks! 😁
Now I just have to discover a way to apply it on my software
Кто Сможет помочь?
Meow hi
hey! i cna't wait to make Plenty of hilarious contents with this awesome program!!
im trolling my friends
Ayo? @wicked heart level 1 !!! 
just made my first masterpiece
have you changbin (skz) ai model?
Ayo? @shadow solar level 1 !!! 
@ionic pumice hi im searching changbin ai model but dont find do you have?
you could try using /search in #🔍│find-models
i did
Ayo? @tribal sequoia level 1 !!! 
colab doesn't make the pth files, rvc do. colab makes pth files then how would there be pth files on local
tysm for help <3
All I know is I used colab and boom, popped out a pth file after all the cells I clicked don't see how rvc would have anything to to with it
Do*
after reading the everything it looks like colab ran rvc and rvc is just code
Ayo? @elder willow level 1 !!! 
Hi people,
I'm looking for moderators/staff for the website
https://rvc-models.com/
If you are interested, please send me a PN @opal cosmos
I'll be happy for some people helping
@hollow tangle sei una maga , sono entrato vedo il tuo nome , non faccio in tempo ad aprire lo yt e già ci sono i tutorial su come fare grazie non ho fatto in tempo nemmeno a fare domande ❤️❤️❤️
occhio che non sono più aggiornati, mi scordo sempre di pubblicare gli aggiornamenti :(
See at least you explained why it doesn't unlike just saying "no you dumbass"
comunque è molto semplice, se hai bisogno di aiuto puoi chiedere ai nostri helper!
Grazie mille
e grazie per avermi ricordata... dopo lo pubblico
how long have the guides not been updated lol
when did So-Vits go out of style?
Does anybody need some car insurance?
u been living under a rock or smth
long time, but there are new guides like for Ilaria RVC
i was just asking i didn't know 😦
huh
no offense intended sorry lol
Sussy baka
come faccio a creare una voce
?
yeah also normal I think
okay this is cool thanks

Hello everyone
Hwi
hey
hai
hey does anyone have a chief keef voice model?
hello
Hi
hai
**Method 1 🔎: ** The search accuracy is 100% 💕
Upload the <t:1691292195>
Server uploaded: AI HUB (RIP)
Channel uploaded: AI HUB (RIP) Post
Profile Owner: winwln
wlnwin
winwln
3444
EN 🇺🇸
RVC V2
171 MB
RVMPE
400
0
mb i was looking for that channel yesterday but my blind ass couldnt find it
Hi
https://youtu.be/1the1OE_Wlg Last Night by Morgan Wallen Sung by Jschlatt. Using mama leechys Model.
#🔍│find-models chief keef
damn i wish there was a chief keef model
does anyone know how to make one?
fr?
yea
damn probably coz he didnt have the chief keef ai model
that was just a warn
after he asked why it was a warn and got banned
pretty sure they banned him for were talking abt it rn on furry rp staf- cough cough i mean staff chat

hey guys
tell us the truth
u banned him to get no more flashback didnt u?
do u know ho to make a model
@wintry sinew
🤫
yh wait a moment
sorry
hey i got a question ^^
i ran start_http for the first time
does it work for mac?
it worked i shut the program down by pressing "X" then i clicked on start_http again but the process automatically stops
now i signed out of my user, thinking the process may be running in the BG, but that wasnt the case
not sure try
would u mind write this here #✨│ai-help sorry
colab works on mac its not running on your computer
could you check #✨│ai-help
well i reccomended using our #1159513888199540817 tbh
i mean the guide is prob hard for new users 
how do i use /search?
can anyone point me to good paid models?
I have a professional grade voice
Synthesizer V AI 重音テト
Vocal style:Powerを活用したボーカルサンプルです。
力強く熱い歌声が特徴。
Alive
Music:デスおはぎ
Lyrics:sforza_t
今回のためにご本人にリアレンジしていただきました!
ありがとうございました!
Synthesizer V AI 重音テト 2023.4.27 Release
製品ページ
https://www.ah-soft.com/synth-v/teto/
重音テト公式サイト
https://kasaneteto.jp/
重音テト公式ツイッター
https://twitter.com/twindrill...
yo
hey
👋
hai
hello
hai
.
hi

hai
is there anyway you are able to hear yourself with the voice changr and talk on discord
👋
just use the monitor option
ohaa
Hi
I'm Russian and I drink cola
bro can someoen find me a realistic girl model?
Searching for the perfect RVC e-girl voice model? 
oh thanks
I think I know the reason why that one model of Mutahar laughing sounds so unusual
It's because the AI used the jumpscare he was laughing at instead of the laughter itself as training data
The best meme of 2023 is..........?
Imma say Skibidi toilet
Hii
Information for new members! 
Voice Changer help:
#🔍│help-w-okada
Cover & training help:
#✨│ai-help
Searching for models:
#1175430844685484042 | #🔍│find-models
https://weights.gg/
Hi I'm new here 💜✨
hi new here, im dad
Who's dad?
I'm not a father 🥲
what does 100 Epochs mean?
100 training factors I guess
😭 😭😭??
haha
100 cycles of training, where 1 epoch means the entire dataset has been trained once
Hi! I'm new here, curious about voice models building for french language. Have a good day!
@idle light send the wiki
dicktionary entry or guide compilation?
the dick
Epoch is a unit of measuring the amount of steps the model was trained for. Usually 200-700 epochs are enough for a decent model. To know EXACTLY the best amount, use a tensorboard.
Same thing, just make sure you train on samples where someone speaks french

What if we feed a video of like a podcast where two people talks. Haven't searched yet, but wondering if there's a way to auto-detect which voice is which just by the pitch
Oh I'm not sure about that
!ban @elder willow Scam. compromised account.
@dandesz banned
Reason: Scam. compromised account.
Duration: Permanent
y'all should make it so newcomers can't send links until lvl 3 or something
nah, ppl sometimes need to send stuff for help
then lvl 1
that would be easier
or ask to send them on dms
yeah i'll ask about it
hello guys
can u
you can, just use the monitor option in the gui, set it to ur headphones
What can I use to make ai voice covers?
it depends. do you have a powerful computer?
Vote Booster: 10% (10 hours remaining)
I also want to make 🙂 but I don't how to 🥴
do you have a powerful PC? let's start with that question
Yes I have a PC but I don't know if it's powerful or not it just works well 🙃
what GPU do you have?
Sorry, I don't know about this 🥲
A GPU is a component of a computer. They play a big role on the performance of AI tools apps
In your PC's Start menu, type "Device Manager," and press Enter to launch the Control Panel. Click the drop-down arrow next to Display adapters, and it should list your GPU right there
what's the name of it?
Ah I see thanks for the info 💜
Name of my PC?
I'm not asking for the name of the computer. I'm asking the name of the GPU
I said I don't know 🥲
But still can't I make ai covers without this?
Ayo? @fickle harbor level 2 !!! 
yes
then what does it say in perfomance
my guide https://rentry.org/ilarvc_inf_guide
~ Guide made by Julia ~
~ Improved by Alex and Squishy ~
Note: It works on mobile too! If your device is slow and can't run RVC locally, don't worry. This software doesn't use your device's resources. All you need is a stable internet connection.
READ THE GUIDE TO THE END. It will be worth it!
~T...
Ok thanks I'll check this vid 💜
have fun!
Where can I download the voice model?
in the "download voice model" tab
next time you need help ping me on #✨│ai-help
Ok tysm 😊
hi
hi
hai
there is a bunch of diffsinger ai models made specifically for french
what everyone doing right now
ai hubbing
cool cool
anyone got a good corpse model there was one i had and i got a new computer now the only ones i can find are bad ones
Is there some kind of epoch calculator? It would be easy for model training
hi
NEW E-GIRL MODEL!!!
Due to some request i made a e-girl voice model with pure french accent that can easily speak french and english with pure french accent.
Grab it here now here! https://ko-fi.com/s/5723aa97ec
...or check out my other models here! (free stuff included :p) https://ko-fi.com/ilariaowo/shop

Guys, I really need help in here.
Searching for the perfect RVC e-girl voice model? 
Can someone make an Angie Yonaga AI voice model? She's from Danganronpa V3.
Did you search on weights.gg?
#1175430844685484042 | #🔍│find-models
or alternatively use weights.gg
I tried, but nothing is in here.
Ayo? @blissful imp level 1 !!! 
hi
girl model
Searching for the perfect RVC e-girl voice model? 
It worked!!! ^•^

done
Y'all should make it so you need to be lvl1 to send links
we're working on it actually
Cool then
I have found 3 results for the search leafy...
I need help
where the "voice models" channel
thx twin
Ayo? @twilit apex level 2 !!! 
i love your pfp
how do i get the bfb two voice model
if its not there #1159289738314919936
Heya guys
How do I make ai covers?
I need an e-girl voice model 😩
Searching for the perfect RVC e-girl voice model? 
Oh
Guys, please help me.

Guys, seriously, how do I make ai covers out of these models?
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
How to Make AI Covers using Ilaria RVC
Link: Documentation
Credits: Nick088, Kanav
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How to make an AI cover using Hugging Face 🤗
Link: Rentry
Credits: 👽 Julia (ailen2091)
@hollow tangle wait noo that last guide is outdated, is not ilaria rvc 😭😭
Could you please change it?
u need to ask raytracer
@turbid mulch them?
yea
Ray can I DM you?
yo
Does anyone know if it has Goku's voice pt-br?
Ayo? @stark quest level 1 !!! 
anyone got any good uhh vocals for undefeatable that ai can sing and actually sounds good
hi people i got a voice pack it's from league of legend shen how can i upload here ?
Are you talking about a skillet song?
talkin bout dez nuts in yo mouf
Got em ha
guys can i ask a question?
No me is just advanced caveman
-huggingface
Suggestions for @stuck veldt
English guide
Link: https://rentry.org/fdg_guide_new
Credits: FDG
Spanish Guide
Link: https://rentry.org/AI_Hub_guide_esp
Credits: Julia (ailen2091)
Hindi guide
Link: https://rentry.org/AI_hub_guide_hi
Credits: Enes
Turkish guide
Link: https://rentry.org/AI_hub_guide_tr
Credits: Enes
thanks mate!
Ayo? @stuck veldt level 1 !!! 
Yo I highly recommend if you have laptop is using UVR5 To get Good singable vocals
But sometimes I noticed with it that with a karaoke model if it has too much background vocals it will cut off alot of the main vocals as well
Or it's just too quiet
For that matter use 5_HP and lower the aggression setting
Leave it at like 3. But see what sounds best
I was just using weights.gg
Ayo? @gleaming mist level 2 !!! 
just needed some good audio or smth
</say gex:1185608733540225057>
Iv used that exact model or 6_hp probably need to adjust a different aggression setting
Ayo? @iron patrol level 8 !!! 
Yo wth poopmaster
dies of not knowing what any of that means
true
🤔
guys can i ask a question?
what did i do
ofc
Everything
in this server i get help with ai's right?
yes
Oops
How do I make AI voice models?
ok, so i am searching for a method to answer questions quickly, and the first thing that came to my mind was using an ai, i used bard but sometimes it gets wrong, i tried chat gpt, but sometimes the questions are about new things so he can't answer
THIS PERSON IS LYING!
Ayo? @elder willow level 1 !!! 
can anybody help me with VB
Cap
try gpt 4 or google bard for newer answers
oh wait
i didnt read fully
bard gets wrong answers
I am trying to get a clean version of undefeatable so I can have Luigi sing it, this is the best i've got (I'm allowed to post files here, right?)
THIS PERSON IS LYING!
what you asking it?
nevermind not letting me anyways
some general culture questions
Lying about what
I can link a YouTube link and Google colab to make your own if you wanna do it
try what squishy sent
Local tutorial (only if you have a powerful PC)
https://docs.aihub.wtf/guide-to-create-a-model/model-training-rvc
For low end users:
https://docs.aihub.wtf/guide-to-create-a-model/model-training-colab
For more guides:
https://rentry.org/aihubstuff
dw its just a bot auto react
stack overflow, and watch out of LLM-generated answers
@idle light next time use <>
mfer forgot about what i told em to do

dies of just wanted some ai-friendly audio of undefeatable vocals
oh okie
Sorry thank you
i was unsure at first but its amazing
could try that too, thanks
Go do your annual homework
dw bout it
cuz i think it uses gpt4 as a base
türk varmı lan
yes
so it's like using gpt4
fun fact: ||If you are reading this, it is already too late. You have been infected by the curse of pee pee poo poo man. If you don't copy and paste this on 5 different servers, you will face the consequences. I was a victim like you, trying to be free. ||
also it can access the internet
I am convinced bing spends more time on its ai then it's browser saftey
THIS PERSON IS LYING!
i think we have a Turkish text channel
why is it always me
Damn I like the part when it told me the fact, that hits hard
seems like they don't believe you

"is anything?"
@languid harness is this real?
is anything?

Your mom

crazy
guys how can i do one of those AI text to speech but with the voice of some celebrity etc
i dont think u can
dies of no help lol jk
use Ilaria RVC
i mean u can use in dms or #1159290666019471421
Oh yeah still needed vocals right?
yup
undefeatable, yk the sonic frontiers song
thanks
Ayo? @elder willow level 1 !!! 
a server my friend has has the bot in it somehow
Is there not any good acapellas on youtube?
well, they dont work good with the ai
the best one on youtube kinda makes Luigi sound like a bird (aka he screetches sometimes)
Maybe try pitching down the vocal in settings?
u could try asking harish on the wbb server
Cause if it hits a curtain vocal range with some ais they will screech lol
I've been trying -12, should I do lower?
Welcome to the wordband Discord bot, poopmaster_123! You can use this bot to make covers of songs using character voices. Here's how to use it:
/list List available voice models
/cover Make a cover using a voice model and a YouTube link
/register Register a new voice model from a Huggingface link
/login Connect your Discord account to a wordband account. This will let you generate more covers, and unlock new features!
/help Show this message
Use /help command:<command> to get more information about a specific command and what parameters it uses.
The wordband Discord bot works on a queueing system. There will always be a free queue, but you can buy a paid plan to increase priority and access more features.
You can join the wordband Discord server to follow new updates here: https://discord.gg/nGnmj769TJ
Ayo? @elder willow level 2 !!! 
Maybe yeah to -24 or if that fails I can try to make extract them but not sure if I would be able to have enough time rn, since the extraction takes about 15 minutes or more
Or sometimes since you said your using weights voice model it might just be the voice model too
I uploaded an existing voice model (it's not mine, I'm just using it) and i just did -24 stil doesnt sound good
basically I think the problem is too many vocal effects
Do you have a laptop?
the voice just has effects, it's all voice, but the voice has too many effects, making the ai mess up
That's what I mean is you can get a model to remove those effects
oh wahoo
that's too much, -12 is usually for female to male
Searching for the perfect RVC e-girl voice model? 
You just have to download a model to remove it
I did -12 first but it didnt sound good
Yeah usually the max I even go to is -2 if that
@tardy hare 😭😭??
@idle light can you explain to him how do download the reverb remover and de noise model for me?
It's in the settings
Click on the gear and there should be a section to download models
settings -> download center, then VR model part, or MDX part for Reverb HQ
Ayo? @gleaming mist level 3 !!! 
I almost think it's his voice model too knowing that it's a weights model but idk how good some of those models can be so idk
Dude 😭😭😭 @wintry sinew
I just know based off of a mark rober ai voice from weights it sounded garbo
bro cant catch a break 😭
😭😭😭
You win. I ran out of gifs 
luigi-dance.gif
no its not, you fool
"torch_cpu.dll", sounds like its not going to make my pc heat up
is this an embed fail or is it not?
@wintry sinew gex creature
gex
adas

discord sucks
fr
reloaded and the embed is here
the double gif is real
teach me
I can stream if it'll help
but still
teach me!!!
You just need to download it in UVR5 and use it in the program, I can't do it atm cause I'm going to headed off to work soon
THIS PERSON IS LYING!
Imma dropkick you
lmfao
whats the name of it
lol
Just search for a model that is titled in the download section that says de noise or reverb remover
There's different models in 4 or 5 different categories in the download model section
I think it's the very first category
would reverb HQ do anything?
Maybe idk iv never used it
wow all of the de-echo/reverb are made by FOXJOY
dunno why i said foxjoy in all caps
Lol, well I hope I helped in some way I still haven't fully familiarized myself with the program so I couldn't outright say the exact right things to go to
But yeah that model should do the job
i love how got rejected 2 min after posting,but when i upload again,there is no reaction 😂
Thank you
hey, does someone have a good girl ai voice
Searching for the perfect RVC e-girl voice model? 
@modern carbon
ty
that website looks sketchy ngl
its rentry 💀
if i try to download it says "🥰🥰click here🥰🥰"
its a rickroll if u want actual models go #1175430844685484042
oh😭 aight ty
Ayo? @modern carbon level 1 !!! 
Hi!
It looks like UMG can now copyright strike ai covers that used their artist now, not just songs licensed to them anymore. I knew I should've took down all of my Rihanna videos from the other day. 😞
hi
i want to use text to speech of a character, so i need samples from that character, which i will cut myself, but then what am i gonna do? what website or program am i gonna use? and will it suck if i dont know anything about? is it difficult? because i wanna give the voice some emotions like anger, yelling, accents, sad, happy etc
can anyone guide me
hi whats the best rvc software right now for making covers? I have a 3060
Information for new members!
Voice Changer help
#🔍│help-w-okada
Cover & training help#✨│ai-help
Searching for models#1175430844685484042/#🔍│find-models
https://weights.gg/
what is a model?
Ayo? @elder willow level 1 !!! 
thanks
voice models
so its a character
yeah i mean a person can be called a character
would you call a smoke detector a character?
can you help me too Squishy
wat u need?
i wanna make ai text to speech for some cartoon characters, so first i need to make a compilation of them talking for a few minutes, then what? what is next?
and what do i need to download, are there any free ai voice cloning programs?
its called voice cloning, right?
and how do i give the voice emotions? like, yelling, whispering, anger, sad, etc
anyone can help me to install ai voice changer
use Ilaria RVC
is illaria local?
isn't that just under dutch law?
are you sure or you are trolling?
what?
no
use it
it's good
i've used it
theres so many buttons and stuff
and it didnt create the voice that i uploaded so idk how is that supposed to work
Guide made by Julia
Works on mobile too! Just make sure you have a stable Internet connection.
How To Use
Start by making an account in Hugging Face here.
Go to the Hugging Face space of Ilaria RVC. It's recommended to duplicate the space to make it faster and to avoid errors in the system. For t...
Is it? Maybe I'm wrong, a lot of my Rihanna videos were removed this pass week but they weren't copyright striked, but today UMGP copyright striked Rihanna covering woman by doja cat, a song that I thought wasn't licensed to them in anyway
what is the best kanye west model rn?
is there any uvr model that supports getting vocals with an already existing instrumental of the song?
Do you have a voice model for Bridge Carson? 
Did you search on weights.gg?
I didn't know he existed, let me see
Ayo? @elder willow level 1 !!! 
Uhm yeah, MDX23C
It's a website :3
It's cloud-based. You can even do it on mobile
Then you can make it yourself or ask someone else to do it in #1159289738314919936 for some money
what order should i input the files in uvr5, original or instrumental first or it doesnt matter?
. . . OK
Instrumental first
thanks, i hope this works because ive been unironically bashing my head against my computer for the last 6 hours trying to get a good isolation and i have around 50 unsuccessful attempts
Ayo? @pallid osprey level 1 !!! 
ok wait its just running interference on each file individually
a
Can someone please do me a favor and make a Pinkie Pie AI cover of the song "Birthday" by Katy Perry? I'd do it myself, but it's too hard.
Ayo? @sturdy violet level 1 !!! 
Why can't you do it? It doesn't sound hard
It is hard.
It'll take forever for me to download the stuff I need to do an AI cover.
Please???
What stuff?
@sturdy violet you know nowadays you can even do it on mobile in just a few minutes, right? I can guide you through :3
yeah uh how do i make uvr5 actually get both files as an insturmental and source and not just 2 seperate inputs
For free too
that literally explains nothing
it doesnt do anything with the mp3 file i upload
then idk how to help go check #✨│ai-help
You can fine tune all those parametets
im eepy
What app?
https://rentry.org/ilarvc_inf_guide if you need help ping me on #✨│ai-help
~ Guide made by Julia ~
~ Improved by Alex and Squishy ~
Note: It works on mobile too! If your device is slow and can't run RVC locally, don't worry. This software doesn't use your device's resources. All you need is a stable internet connection.
READ THE GUIDE TO THE END. It will be worth it!
~T...
it just makes the same voice
Go eep then !!!
idk how am i supposed to use this, also, what do i need to learn before trying to do this?
oki
its worthless
Ayo? @elder willow level 2 !!! 
sorry for bumping this for the third time but i really need help with this
Alright so, explain everything from scratch. What is your goal? How does your audio sample sound like?
I'll help you
basically i have a song where the voice has quite a lot of reverb, which ends up introducing a lot of noise when i just feed it into isolation models. but i have an instrumental track of the song and am just looking to somehow isolate the vocals with the help of the instrumental track
Speaking of UVR, is there a method to extract the harmonies only? I know Karaoke 5 removes it so I want to do the opposite
pretty sure karaoke models extract the harmonies alongside the harmony-less track
Neat, i'll take a look
Did you try using DeNoise on the isolated, de-reverbered vocals?
i did, it didnt work
Ayo? @pallid osprey level 2 !!! 
Can you send me the link of the song?
Btw if you are using MP3, you are in the wrong track
i was using ogg, should i convert it to wav?
where can i use the rvc modals anyone got the co;ab link?
was good
That won't do anything. OGG is a lossy format, which means it loses the original quality. Converting it to WAV won't do anything since you can't magically revive the lost data
Check DMs
yeah i dont think i can get the song in non-lossy, thats the problem
What song is it?
@empty flicker
@empty flicker
Ayo? @old spire level 1 !!! 
@old spire
Ayo? @empty flicker level 1 !!! 
@empty flicker
KaiCenat ft KaiCenat - It Wasn't Me - Lyrics (AI cover) I create this song 😂
Ayo? @unborn berry level 1 !!! 
can somebody make ai model for me?
does anyone know how to remove backing vocals from an acapella?
With UVR. Use 5_HP algorithm
thank you!
Ayo? @pale nimbus level 1 !!! 
is there a model to remove the lead vocals?
The app will extract both the backing and lead vocals. You'll get both things
alright, thanks!
Yw!
hhtht

5_HP kareoke?
just curious cuz I was wondering the same thing
Yes. Same thing
guys what is an epoch
An epoch is basically how many times your dataset has been trained fully.
E.g. 200 epochs
Means the dataset has been trained fully 200 times.
Why cant i make models anymore?
ok so its not something i have to setup in the rvc
uhh it only extracted the backing vocals
in uvr you can tweak it so you'll get both outputs
i think you forgot to set that up
how do i do it?
in the main page there should be two options that you can tick. i dont remember how they were called
@wintry sinew do you?
huh
they wanted to extract both lead and backing vocals in uvr. they only got backing as output. how do they make it so they get both files?
let me check
i meant uvr
what model
5_hp
tetotetoeteotetoetoetoetoeototetoetoetoetooetoetoeto
dang I said toe
is it possible to train a model to whisper?
mhm. just make sure the audio is clean and use wav
oh ok thanks
it would be here
@pale nimbus
morning everyone
Make sure you set Windows Size to the maximum value. The bigger the more output quality
Uncheck both
Morning mx
how r u
how do i make a model
where is ilaria rvc
it doesnt work for me
Error: Failed to load logs: Not Found. Logs are persisted for 30 days after the Space stops running.
If you're going to get in legal trouble with AI voice models what's the point of keeping the server public?
There is legal ai voices
True
However
how do i converrt a mp3 to a path file
https://m.youtube.com/shorts/Bosor4s7FLI
I got this dude its awesome
Have you seen what happened to Überdruck ai?
Ayo? @half wind level 1 !!! 
Its been done for decades before it was a trend without issue.
Utau, vocaloid, neutrino, cevio, ace studio, diffsinger, synthesizer V and more all have ai voices
Good evening. Anyone here that knows how to train voices in RVC 2?
does anyone here do free request
im trying to find a fix for an error that has been preventing me from training voices. Error number 38, i think it's pretty common
Ayo? @solemn dune level 1 !!! 
Ayo? @tender finch level 1 !!! 
Hello !
No results found for the search dovah...
hello
Anime women arent real
Anime women are real
true
Ayo? @austere rampart level 9 !!! 
oh

haha
hey poopmaster
Yea
Move it?
get it out of the compressed file
Yea
Where did u get the link to it?
a bot in the server
Whats the best version of rcv to use now, is there even a difference
Then it's safe
I use applio since the old google collab died
Depends on the person what fork ig
Mainline is faster but has less options
Applio can be a pain to install
All i care about is the funny voices sounding good
But i dont really know how that works since there are so many forks
Thats what i thought
Where can i check on the progress of rvc2 itself, whos making it, is it being updated to compete with the likes of elevenlabs
It's not being constantly updated
But is it in development or "sorry this is it" kinda like what happened to local image AI generation who now gets mogged by bing
Ayo? @tidal echo level 1 !!! 
I dont really know whats going on with the development 
-rvc
Suggestion for @subtle fulcrum
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
How to Make AI Covers using Ilaria RVC
Link: Documentation
Credits: Nick088, Kanav
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How to make an AI cover using Hugging Face 🤗
Link: Rentry
Credits: 👽 Julia (ailen2091)
what will i click though
i downloaded it and get it out of the compressed file
but theres a lot of files inside the folder so
which one am i gonna click to use the app
embedn't

ily
i like rin penrose
GG
Hello!
hi
How's it going. This is an amazing Server!
im tired
but i cant sleep because i dont wanna miss the season finale of rick and morty
THIS PERSON IS LYING!
if i sleep now i will miss it tomorrow
because i will sleep tomorrow too
but if i skip one sleep then its balanced and i wont miss anything
-rvc
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
How to Make AI Covers using Ilaria RVC
Link: Documentation
Credits: Nick088, Kanav
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How to make an AI cover using Hugging Face 🤗
Link: Rentry
Credits: 👽 Julia (ailen2091)
no ice spice voice changer
hi
gi
Ayo? @south obsidian level 1 !!! 
i mean hi

what is a stable diffusion
/duplicate
how do I erase duplicate models?
im trying to use the /duplicate command but idk what channel to do it in
hello , im sorry to make this question but how does the voice models work ive been awake to many hours and i just want to do a private collaboration with my AI artists

basically a voice model contain the files with the data about the voice. to use a voice model you need to put it in a software called RVC
you wanna learn how to make an AI cover?
ping me on #✨│ai-help if you need help
yuh
so true
Hi
hai
give me this cat right now!!

@languid harness
.
Hello
give me this cat right now!!
give me this cat right now!!
hello
No
Who makes the most realistic models
hello
1 hundred %
hi
does anybody know how to make whispers not so glitchy in Mangio
jugboy try using rvc disconnected here is the link so you can try to make models https://colab.research.google.com/drive/1XIPCP9ken63S7M6b5ui1b36Cs17sP-NS#scrollTo=Yj-_npuA_HRB
Will training a model with whispers be better?
all my queries run for like 10 minutes
is this normal? should i just let it run, or is something wrong?
Reduce the search feature ratio value and try using mangio-crepe
15 minutes
What should I put for "protect voiceless consonants" and "median filtering"
Ayo? @dreamy temple level 1 !!! 
.
Most people don't use that option
Hello
Hello there
hi
hi
hi
Ayo? @frosty spindle level 1 !!! 
guys look what i made 👀 https://www.youtube.com/watch?v=zEWnWabcHz8
hi
so after downloading a rvc model how do I actaully use it like assuming im on a collab drive or something etc. also what are the best settings to get highest quality for covers? should you upload vocals unproccesed with some tune and compressor for cleaning up purposes etc?


