#🧬│ai-chat
1 messages · Page 78 of 1
should i use INFER or INFER TRAIN
Infer for covers
Training new models
That installs it
wtf 4GB
are you expecting 500mb? 😭
Bro thinks ai takes up little storage
its not the storage its my internet
4gb ain’t bad considering other ai like stable diffusion takes up like 20
do i put both .pth in da zip?
bro has a 4000 rtx GPU and yet has a shitty internet 😭
but its also a 7z which means ill have to extract it and download another 4GB
yuh i got D and G
How do I upload a file to huggingface?
You delete the 7z after extracting
so both D and G pth go with the index in a zip?
No only one pth
Which should just be the models name
give me link rvc colab
Hello everyone!
yuh i dont got that at all
i just got D_23333.pth and one in G
Did your model finish training?
yes
It’s in the weights folder
.
.
how can i continue where i left off?
like model is already done training but i need the weights folder
last time I trained with colab I connected my gdrive
idk about the colab used to train the AI now but if u connected ur gdrive check there
ya i wish, my drive has nothing in the rvc folder except the .zip
Whatsapp, happy to be part of the com
which local fix do i install
hi
okay now it says "[notice] A new release of pip is available: 23.1.2 -> 23.3.1"
but it wont let me install it
what do epochs mean
e
for the voice models
are there any creepy voice ais i can use for a voice changer?
Ayo? @velvet oracle level 1 !!! 
Training cycle, basically how much the ai looks over the dataset
oh ok
Ghostface
okay so theres like a bunch of shit on my screen after installing mangio and I dont know how to import my voice models
Ayo? @thorny scarab level 3 !!! 
what do i do
how do i make my voices not have alisp when i talk
Ayo? @elder willow level 2 !!! 
wtf its not letting me import my voices
just to ask here cause it's more active: in RVC is there anyway in i can get the audio to not glitch out when someone screams or yells?
no
how do i get rid of it
are you using the voice changer?
yes
try messing around with the index also the input
Do you have a lisp?
Hi felt
if a song has remix stems that you can find online, what part(s) would you guys recommend using for ai covers? like obviously lead vocals and adlibs, but idk about like harmonies and backing vocals
no i dont have a lisp
Try this
hi
when i say "hello everybody my name is markiplier" it says "hello ethry body my name ith markiplier"
Hm... from my experience no
I think that's just the AI lagging
does anyone have a answer for this lol
try turning it down first
then turn it up because turning it down decreases lag
it'll sound less accurate but I think you should be fine at 96 chunks
I use 96 and it works fine
or 112-128
aww man. ig i'll just adjust what i can. either that or just grab a clip from the original VA
fnaf
i turned it down to 98 and now its choppy
I tried doing the Takashi S*X screaming meme on my voices and it did not go well ahahaha, it basically broke the AI
turn it up to 128
Get higher level than me
try to find the right chunk level
no thanks, I'm sleepy
yeah i have a example in #✨│ai-help if you wanna see what happen to tails lol
Then go sleep
you can also try changing the extra
see what happens
turn it down to the lowest one which is 4096
and try it out with chunks at 256
just keep trying different chunk + extra combos till u find the right one
it takes a bit of tinkering to get the AI Voice Changer to work properly
he dies ahahaha
Friends, are there lvbel c5 and batuflex sounds?
I think the reason why this happens is the AI can't handle that sudden spike of volume and tone difference
Ayo? @vast idol level 3 !!! 
on the channel
what is extra supposed to do
that's the ram you're allocating to the application
I have half the mind to leave it like that cause it's gonna get drowned in other voices anyways but, in the case it all turns like that i'll have the solution lol
hi
hi everyone!
hi everyone, i have a question, what is Batch size per Gpu ?
hi everyone. i am now here and i casually make ai voice time to time, but rare
hi
hi
vou me mataaaa
can someone tell me how do i get a non-web version of rvc v2?
hi everyone
guys anyone here knows programming to help me rq? #✨│ai-help
Hello bois
gen chat dead huh
can you use it in calls?
not with rvc but theres other methods to make screams
what's the other methods?
Welcome to SynthV Teto! Very happy to have you here!
Wanted to test her English right away, and decided to release the SVP to comemorate her arrival. She's a real powerhouse even in SynthV!
Original Song by Demi Lovato
SVP, Tuning & Mixing by Ariotsu:
bit.ly/41Jcyzo
vecken advertising Teto ai fr
she's hands down the best ai vocal there is
fair but i don't think this'll help me atm lmao
what are you trying to do
i wanted to get rid of the glitchy screaming thing that happens when you yell in horror. like if your shouting the ai can't pick up on that (i have a example in #✨│ai-help
rvc does struggle with extreme pitch variations
.
yarg... sploinky cat
hiç
thats just how gan works ong
people in this chat apparently dont understand that fast changes in a dataset is bad
**Method 1 🔎: ** The search accuracy is 100% 💕
Upload the <t:1688176896>
Server uploaded: AI HUB (RIP)
Channel uploaded: AI HUB (RIP) Post
Profile Owner: 0x3e9
0x3e9
0x3e9
663
TR 🇹🇷
RVC V2
139 MB
N/A
600
0
its just a limitation of the engine, its not BAD, just not designed for it.
which is based on a gan network
which is just limitations of the pretrains and the network :3
well yea i agree
its not BAD
just designed for it
is a good way to put it
its the same issue other ai have run into, it took a hyrbid method to achieve the range of a vocaloid while using realistic ai vocals
Having a conversion in general instead of varified moment
kil
if you want an open source version that's like synthV but uses similar technology to RVC there is diff singer
yes it's time for Iolite DiffSinger !!! i am so habby with the results she sounds just like her this time...
original - Kill Bill by SZA
tune+mix - Yours trulyyyy
midi - Dav-P
retweet it or smn: https://x.com/C5G4D4A3/status/1721105044962255037?s=20
no i didn't mean that 
what does DRIVER IRQL NOT LESS OR EQUAL mean
How do I get google drive links to work?
Hai
omfg
Gex
Bruh
Bro
I died cuz of u
I'll make it a bannable word
What will i refer to then
Wait you got banned from verified?
ilaria's voice
did i just do a funny?
Ayo? @stark axle level 1 !!! 
(im advertising it)
is it EN or JP
what is okada
💀
what
A real time voice inference

Does it have a huge delay
I suppose that it's very minimum if you use a wired mic with low settings in okada
I'm not familiar with it so I am not sure
Try asking on #🔍│help-w-okada
What are you going to do with it?
What gpu do you have?
RTX 4060TI 16GB VRAM
Yeah, you will be fine using it with vr lol
Last time I tried it it didnt work and when I researched it online I had to use like 2 microphone drivers or something?
Do you mean a virtual cable?
wdym cable?
You need a virtual cable to output it to other applications
huhh
-realtime
• (Local) W-Okada Voice Changer
This one has more options and is somewhat more convenient to use than Real-time RVC.
Credits: ravencutie21
• (Online) W-Okada - Colab Version
A free, online version of W-Okada that allows you to run code on Google’s powerful GPUs.
Credits: Antasma245
• (Local) Real-time conversion of RVC
More accurate than W-Okada's, but requires a Nvidia RTX card.
Credits: teas.hi
• Russian guide for voice changing
Credits: redoverflow, blackcat76it and tntmaxim
• Text to speech:
Note: If you need further help setting up/using the voice changers, you can ask for help in #🔍│help-w-okada or check the channel's pinned comments.
A virtual cable is a background process that acts like a audio cable
no i dont have that
Where out outputs go in and inputs go out
There’s one linked on the guide
which one
Can i find Bangladeshe voices model
The guide for w-okada, assuming you're using that
theres two guides
Local Okada guide
Search on #1159289700490694666 or in weights.gg
who pinged
@rapid bloom hi
@rapid bloom
No idea. Ask on #🔍│help-w-okada
is 4070 ti good for training?
level up
I hate this discord nuke
I was a model maker now I can't even submit
and here I am trying to get lv 5 for verified chat
fuck this shit
Hi
hi
Heya, someone have the doc on how to get good results with UVR? 
here me out
download models: Kim Vocals 1, MDX23C, UVR-MDX-NET Inst 3, UVR-MDX-NET Inst Main, demucs v4 htdemous_ft
i miss the best admin of aihub
download models: Kim Vocals 1, MDX23C, UVR-MDX-NET Inst 3, UVR-MDX-NET Inst Main, demucs v4 htdemous_ft
Emsample them
You can request the role in #1159289738314919936
thanks
ensemplethem?
Emsempleing mode on UVR5
lets you use multiple models at once for better results
My combination is almost perfect of vocal seperation
Shit thank you
And what do you use to remove reverb?
@tardy hare?
after ensempleing use MDX Reverb HQ
Ayo? @half zinc level 4 !!! 
that should give you VERY good results
In my experience it's best to remove reverb and then harmonies
but I fine tuned them some settings
harmonie seperation is kinda iffy
you can use 5_HP-Karaoke-UVR on vocal track after seperation for spesificly harmonies
but 6_HP-Karaoke-UVR is better at backup vocal seperation
hi chat
bye chat
Bye chat
bye mfs
rude
/j
The VB-audio driver cable isnt working
a
hey idk if this really applies to the help channels but in yall opinion's is rvc realtime better or w okada
ah alright cool thanks
Im getting some issues with CUDA_PATH is set but CUDA wasn't able to be loaded. it worked before but this is a new install so I know my GPU is fine, I may be being dumb but its really annoying and i tried like 50 things
bro i didnt know there was a new server
old one got deleted and i thought it was done
god cuda WILL be the end of me
😭
where frisk
/j
ded
Ayo? @jade cape level 1 !!! 
.
How do I fix the voices just being default and not working
Hello!
Like none of the default voices are working now they’re all at zero and 0.1
This happened when I tried, adding a new model voice
Ayo? @hybrid hollow level 1 !!! 
i installed cuda, double checked cuda install, set cuda path
nothing fixed it
Ok sersouy
I looked it up, i tried reinstalling
theres no guides that i can find on this that fix anything
fuck kits
THEY COULD NOT
IM SORRY I HAD TO SAY IT
did you run as admin
sometimes cuda isn't added to path
Yes
Also i added cuda to path
Il figure it out later need some time to cool down before i throw me computer into the next universe
Can someone pls help me
Hello guys 🙂
Hi boys
Yo
Hi
So is rvc and okada generally the best live voice changers ?
hello
hello
hi

asdf
greetings
Hai
h
I feel so old
who the fuck gave birth in vc
The hell
look at the annoncement
Just did
its a new rule apperantly
The actual fuck
yo is there an fast way to apply ai models on vocals? old web rvc isnt working anymore lol
that was my reaction
yoo
look at the annoncement
Apperantly some sick fuck thinked its a good idea to give birth while on voice chat.
and its a rule now
Ayo? @half zinc level 5 !!! 
Don't give birth on vc
wait
Ayo? @gloomy tundra level 1 !!! 
i think i already saw it some weeks ago
it seems to be a repost
so are there 2 births given on this server ??? 💀
ohhh shiittt hey boys
didnt know we were back
can i get model master back w/o filling out an something again lol
No you have to fill it out again
more than 6 lmfao
You need 15 for master
Alguém do BR?
opa
Aeee pensei que tava sozinho
Bitch
lmfao
double trolled ong
you're so predictable 
what’s your GPU?
3060 i think
Ok
laptop, might be a 3050
So a 3050 mobile?
ethier 3050 or 3060 but yea
Ayo? @eager heron level 1 !!! 
welll like i just wanna be able to run the voices, not trynna make them
Yep you can do that too
Installing RVC Crepe Fork Locally (EASY METHOD) Editor's note, 6/11/2023: Both of my colabs and this guide are up to date now. Keep in mind the ‘crepe’ I refer to throughout this guide is the ‘mangio-crepe’ option, which seems to be usually superior to the official RVC crepe implementation. The ...
Open source
hi
So you can check the code and stuff if you feel like it’s not safe
the runner is at least?
All is safe
bettt tsm
Vencord moment
oh no
hello im using rvc dissconnected and when i ran feature extraction but index training isnt working
Hi using rvc dissconected
Also hi my name is Okada
is there like a best RVC model database?
Ayo? @jade cape level 2 !!! 
ModelAI bot or weights.gg
thanks
hi anyone can help me
I had Model Master in the previous server, how do I get it back?
@lament pasture are you sure you had 15?
i wanna change my language category cmon someone tell me
.
it shows only english voice contents
I mean I didnt have 15, but I did a KSI model which he reacted to and a mod gave me model master role
I have 14154 Voice Models. ❤️
- EN 🇺🇸: 6810 Models.
- ES 🇪🇸: 1907 Models.
- Unknown 🇺🇳: 1720 Models.
- TR 🇹🇷: 1089 Models.
- FR 🇫🇷: 583 Models.
- BR 🇧🇷: 551 Models.
- JP 🇯🇵: 410 Models.
- IT 🇮🇹: 344 Models.
- RU 🇷🇺: 331 Models.
- DE 🇩🇪: 91 Models.
- PT 🇵🇹: 86 Models.
- PL 🇵🇱: 80 Models.
- KR 🇰🇷: 77 Models.
- RO 🇷🇴: 37 Models.
- CN 🇨🇳: 15 Models.
- AE 🇸🇦: 8 Models.
Many models were not classified correctly since their publication. AI was attempted to be used to organize them. If you have a model, you will soon be able to modify its record.
That’s not really how it works now, I would be a master if that was the case now
dang
Maybe try searching that character in their translated name
ill try
Ayo? @normal zodiac level 1 !!! 
which model?
It was an among us model, before the server was deleted
So reactions are gone now
Upload the <t:1692910559>
Server uploaded: AI HUB (RIP)
Channel uploaded: AI HUB (RIP) Post
Profile Owner: timbovill
TIMBOVILL
timbovill
6056
ES 🇪🇸
RVC V2
51 MB
Harvest
50
0
h
weights.gg is nottt working for me lol [models,getAll]: Unexpected token '<', "<!DOCTYPE "... is not valid JSON 💀
yeah weights.gg messed up for abt an hour or so now
Helo
hey y'all im pretty new to this stuff but I wanted to make a voice model to use either as a voice changer or for making voice clips but idrk where to start. I tried installing Tortoise on my pc but I think I've either fucked up my directories or am missing a bunch of dependencies for anaconda/python. Not rly a big coding guy so im kind of walking in the blind here. basically, do you guys know of any other software I can use to make something like that?
Ive been planning a dnd campaign and I want to make Balthasar's voice from warhammer to use for one of the characters
Rvc v2 is good if your on a slower device depending
But there is a command with other options I forgor
yeah I just got linked to a google collab version of it but uh i already used up all my runtime on tortoise on collab trying to make clips
Ayo? @warm prairie level 1 !!! 
I was actually able to make 3 pretty good sounding clips but it took a lot of time
so was just looking for alternatives since spending all that time for 3 10 second clips is a little wack
You can try ila tts
is there a general favorite webui for training on github or smtn
Most use mangio I think
Hiii I'm new here and just wanted to say hi to everyone :DDD
Yep i think
sup
Hai

helloo
Hai
hai
Hai
hey
Hi
is rvc2 google colab also working slowly for you?
it needs so much time to generate the audios
what could be the problem if not
Aguante SOAD lpm
Real
a bugs bunny model exists right
has someone recorded a clip of them saying those exact captions
and put them into the ai
that would be great

H
Laugh at this user!
HEHEHAHA
I love this 😭
howdy
I'm getting errors all the time while trying to get output. Which UI is the most stable in that sense?
Ayo? @elder willow level 1 !!! 
Yep. I don't think my computer is powerful enough for local generation
Does it really matter if I'm doing generation online?
Anyway, you mean RAM specs, right?
Ok, give me a sec
I have 16 GB of RAM, so it should be fine then
Which UI do you generally use?
Could you send the link please?
To the UI
of ram or vram?
Idk if there's any Samsung Galaxy Owner with Bixby on it
But i wanna know if there's even any mod that allows you to install .pth AI Voices to it
Ayo? @sharp flicker level 2 !!! 
Hi
Hello
hi
Hiii encanta tu banner ^-^
How do i even extract the .pth file from an already uploaded AI Voice from kits
Es mejor si no usás kits en primer lugar, es terrible
Con RVC de manera local o los espacios de huggingface tenés libertad absoluta y no tenés que pagar nada
Usá Ilaria RVC. Está en #1159513888199540817
Lo se
Era bacan hace meses hasta que te empezaron a cobrar por todo
Solo quiero saber como recuperar los archivos .pth/index de una voz que subi hace meses
Que el tipo que tenia el link ya no le sirve
Uff que mal
Buscá su cuenta o el nombre del modelo en hugginface.co
A lo mejor lo subió ahí
how upload model here?
First you gotta get your model maker role on #1159289738314919936
do i ask to get the role or
Oh my bad, I meant #1159514067187277865
oh okay
I wish to make a FNAF 1 Song AI cover.. How do you guys do it? Is their a acapella version for the vocals?
Ayo? @elder willow level 1 !!! 
on the Installing TensorBoard Locally file, it says to put TBI and TBL on my RVC folder.
where the hell is that folder
hey guys what is the best sample-length for ai recording? like ~15 minutes maybe?
or ai making I mean
why it say no interface is running right now
how can I ai my own voice
i bless the people of this server that deal with these questions
2 to 4 hours
oh wonderful
Ayo? @elder willow level 1 !!! 
💀
for professional grade
Oh
for a rvc you can do like 20 min
teto has over 2 hours of data
ahhh
asterian gave her the pass
ah alright
Edit: omg, idk what to say, I'm absolutely flooored right now at seeing the amount of views and likes this got in 5 days. I'm honestly crying out of happiness right now. 10k views and almost 2k likes...holy moly. just....Thank you all so much for the support!
AAAAAAAAAAH THE FRENCH BREAD CHIMERA ARRIVED ON SYNTH V AAAAAAAAAAAAAAAAAAH!!!!!!!
*...
we needed a big black male vocaloid so bad...
Look into GrayJay app, it allows you to have platforms like youtube and twitch in one app with a lot of better features.
is Applio better then Mangio
no idea but i personally use applio atm
Applio > Mangio > mainline imo because the accuracy is better on applio and mangio, but the stability and ease for mainline is best
Applio has the best UI and new features though
tts broken on applio.. i mean i dont really use it but still
you can just disable that iirc
ngl i just use applio since thats the only thing i managed to run on colab
Eyeball
Ayo? @digital horizon level 2 !!! 
Heyyaa y'all!
hey one guy who use mango, do you know what this error is ImportError: cannot import name 'FFmpeg' from 'ffmpy'
Is there a program that lets you use a model as a text to speech? Not for making song covers but for saying things you type
Ayo? @spare ocean level 1 !!! 
The problem came from a bug in the Visual Studio compilation tools and the redistributable package, so you will have to reinstall or repair them.
Ayo? @forest crypt level 3 !!! 
a
im p sure u can use any model for that
try elevenlabs, its not free though
ayo for sorisori does anyone know how to upload your own model to make covers with? it only has one option and thats their sample male voice
There's PlayHT that's free. The only downside is that it doesn't emote anger or highlight tone as well compared to elevenlabs because you can just emphasize words with * or "___"
using colab so thats not the problem there
I've been using Audio Webui, which uses the Bark model. It's pretty easy to setup and use, but still getting used to it. I'm super new to AI voices, so not sure on where to begin, but it has text-to-speech on it
skibidi toilet
hello
こにちは
konichiwassup
mb
Ayo? @void wadi level 1 !!! 
/j
basically edge tts port but eh
.
does anybody still have a link to the ai sans model where it sounds like him when hes talking in game
The de de de de de?
nvm I found it I just had to search through my drive
Ayo? @ember bramble level 1 !!! 
Hi hi
🕵️
turkish song ??
hi
hi
do you use the app?
chara from undertaale?!?!?
Hello im new here
x2
Ayo? @ruby vessel level 1 !!! 
I wanna make ai cover so can anyone teach me how to do it?
u just do
Can anyone help me?
Ayo? @cloud nimbus level 1 !!! 
Itsa me mario
Dude
All the rvc's sound like robot for anime girls or girls
Is there a single decent one?
Like they all sound like actual siri
ayoooo
maybe the ones on here u have to train more? shoudn't sound that way
or maybe the audio ur trying to use doesn't translate well to the ai voice, thats usually the reason.
try using a high pitch voice that matches the ai
Get better audio, get better dataset, make better model
Or find a highly popular model
Where's the guide for vc2
Ayo? @deft sapphire level 1 !!! 
helo is there any open source text to image genration or text to video genration we can use it in our post without getting sued.
So iv been procrastinating on getting in on this stuff but now that I got the audios i need to make a model I see people saying if you use the free rvc colab stuff google will just ban your account on colab what do
Suggestions for @spring drift
- AICoverGen-NoWebUI, by Ardha Google Colab
- RVC Disconnected, Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- 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
You can find more info on the #1159513888199540817 channel. If you can't find your answer, feel free to ask for help in #✨│ai-help. Credits to Faze Masta and Antasma for compiling these links.
thx for the quick reply, im def a beginner at this. I was gonna rely on the few yt vids that explained how to use this stuff p straightforward but now they're updated to say the methods they used dont work to train models anymore. No idea what stuff like tensorboard is but ill learn
hi
thanks
Hey lads i want to use my voice to make a cover anyone knows how to ?
where i can get beatrice voice changer type?
Hello how are yall doin today??
fine!
How do I infer Google Colab's RVC without warning?
Ayo? @kindred rock level 2 !!! 
How can I make a sound
@feral halo @feral halo

did you check weights.gg?
Hi
nvm I found it
Ayo? @rapid monolith level 1 !!! 
hihi ppl
Monica Maggioni voice model?
is there any prefectly trained female voices ?
-rvc
RVC Guides (How to Make AI Cover)
Translation by country
hello
Hi
do i need rvc installed to use rvc on voice changer?
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. ||
it does actually
its really fast on cpu
Yeah I'm really confused
i havent tried it yet tho
No code whatsoever
its literally in the 16 build
i found websit
But the source code doesn't have anything
- the beatrice DLLs may or may not fail on some machines
Overall I'm not too sure what to think of it
Anyone here know of an ai harmony isolater? I don’t want to remove them, I want to isolate them
help
i was training an ai RVC v2 Disconnected
i can't import the dataset even i have put the thing in the right directory
if i could've been able to upload images it's best to understand what issues i have
MedleyVox can isolate doubles and harmonies but it's not out to the public yet (at least the harmony version)
#✨│ai-help, send a screenshot there
Great! Thanks
wooo
:3
you gotta be model master or part of the staff team
how many models should i upload
to get model master
i was a model maker then the old server got bombed
5 hq models or 15 models
i dont think i have uploaded 5 models
but i have uploaded at least 1 model so i need model maker
anyone know of an ai plugin or service that lets you feed your own midi chord data into it and it generates a melody out of it?
Can someone just give me the Ayaka rvc model JAPANESE ONE!

just search in weights.gg or make ur own
hiii everyone, ive been wanting to make my own ai cover but 1. i dont know how to get the voices i want 2. i literally dont know how... can someone help me get started with the basics please? thank you!
we have the #1159513888199540817
thank you!!!
Hi I'm Abhishek from India. I'm looking for Shahrukh Khan voice model for a paid project. Please message if you can help
Ayo? @sweet raptor level 1 !!! 
Hi
Was anyone else removed from the server randomly?
Spent 10 minutes looking for it in my tabs
Read the pins that's how I got to know
'Poop' say no more
Ah damn, lots of the models have gone
Ayo? @stray wraith level 1 !!! 
Not really check out #🔍│find-models and weights.gg
Thanks!
is the AICoverGen google collab by Ardha broken rn? it always failed when i try it
Go in DMs if you want!
But remember to search in weights.gg to see if the model is available there
So you won't pay for something that can be taken for free
Yeah, got restricted
Check #✨│ai-help
for the pinned AICoverGen Colab

yo where can i look for the spooky event's ai arts
Ayo? @restive prairie level 26 !!! 
Hi
Sa turk var mi
hi
we need a section for tortoise/bark on this server
did some models get deleted?
hi
How do you make models?
There's a guide on #1159513888199540817
hey
hi
hi
hi
Hi
hey
hi
Ayo? @elder willow level 2 !!! 
LETS FUCKING GOOOOOOO
a model maker lvl 2
wow
never saw something like that lol
dimple is writing...
hi, im looking for a website where i can record upload movement of my arms and an image and get same movement by that image is there something like that.
Ayo? @barren wyvern level 2 !!! 
Could have more fun doing MMD with a kinect for motion tracking
my brain is about to explode
THANK GOD TETO FAN N1 EXIST
im simply on a different level
that reminds me i forgot to put the model in the actual channel for them lmaooo
I'm nowhere close to the biggest fan of teto... do you know who is?
eeh did i got mentioned ?
hi
Tf going on in this server in my absence
Who has Dio's voice model???
thanks
Your hot
hello master
animation - Original choreography by Natuse Ao (@natuse_sea) https://www.nicovideo.jp/watch/sm40564982
audio - Teto AI & LALA // UTAU Tuned by suzie // Mix by LALA // VSQx by DEUXKO // UST by the other Anna https://youtu.be/yA1EMsXxgV8
original - Giga https://youtu.be/LbO4pg_LHUI
model - arue式!
Tda/WYKP/iRon0129 Hair: Tda/Pu/vroid Cloth: Yama...
hello
probably in Weights.gg
а
Cover or an original song
either one
Thank you so much
Np ;)
did something happen to this server i swear i was in the old one and it just disappeared one day and i just realized
or am i just schizophrenic
Yes, our previous server got taken down because some people plotted to take it down
i see
And posted copyrighted content...
Ayo? @forest bone level 1 !!! 
crazy
yep
Maki Ligon did the base file. But you can take a midi, or play it with a piano
Type in lyrics. Then draw the pitch by hand
Because autotune sounds bad
Wdym take a midi? You mean use pitch correct sample in FL studio and then export it as midi?
You can do that right in synthV without a daw
Wait HOW?
But you can also make a midi with fl studio and it plugs right in
The record button, once configured to a midi keyboard you can do it live
Hi, I'm new to playing with AI voice, and have only puttered around with RVC a little bit but I'd love to learn more. I plan on reading my face off for a few days. Thanks for having me.
Hi
Hey can anyone help me with ai voice I wanted to make konig mw2 idol ai cover but not sure what website can do it
Please help me
guys can you help me, I finished training my AI model, but somehow it says index file not found
guys what does the index thing do under tune
Hi
And this lets you use a custom voice model and not a default one pre installed right?
makes character accent stronger but isn't good for performance
what accent
american???
the accent of the character
What do these 3 things mean?
we can't see the image
Try in help rvc
ok
Ayo? @keen kite level 1 !!! 
index rate
voice normalisation
consonant protection
index rate - how much of the accent and stuff the index does the conversion will use
consonant protection is exactly what the name says
voice normalisation just keep as is
ok but what does consonant protection do?
ok so if i set the index rate closer to 1, it will be more efficient towards the voice?
You are welcome 
it'll be closer accent wise
...it protects the consonants?
any drawbacks?
Use Ilaria RVC. There's a guide on #1159513888199540817
might be a little worse in quality
but what do you mean by protecting the consonants? 💀
Try asking on #📑│making-models or #✨│ai-help
😭😭😭 it just does it man, thats all i know
makes Index accuracy worse tho
@storm meadow explain pleas
Can I use in mobile?
I honestly forgot about what this parameter does or what it was called but I think this is to make the vocal less breathy(?) but may cause phasing/tearing
ooh ok
You may have to play with this parameter because if you set it too high I think it will cause some glitching but if you set it too low the vocal may start to sound robotic and start phasing
ooh let me try it out then
Hi
What if I disable it though?
Depends. If you have a clean studio acapella then the vocal might just sound normal but you might hear mouth noises and voiceless consonants like plosives, and the vocal might sound a bit breathy but that's why there is a median filter I think it's called?
You can remove harsh plosives with iZotope's RX but it's paid software but it's AMAZING
Oh okay so, what if I have low quality vocals? At around which level should I set it up?
If it's low quality then I suggest you can set it lower, but not too low because for me the vowels start sounding a bit robotic if set it too low. If your low quality vocal has reverb artifacts you can try a de-reverb VST
Ah yeah no I never have clean studio vocals, I always rip them from their og song in flac
And after I remove the instrumental, reverb and harmonies it ends up losing some quality it seems
I don't know how to get them really clean
If there is no studio quality acapella available in a stem pack or something then your best bet is to sing over it yourself or manually mix the vocal, I might make a tutorial on how to make a separated vocal sound more clear because I have done it before but you really have to tinker around with the settings because every song and vocal is different
It's always nice to have the lead vocal only because if your vocal also has ad-libs it may cause 'glitching' issues
Yes please do it omg!!! Admins might link your video to #1159513888199540817
Yeah it took me a long time to figure that out 😭
hi
dumb Q how can you tell if they are low Q or not. i have a voice setup but im mostly tone deaf (its for voice not singing) real time voice changer
If you're using a real time voice changer for talking you should be alright, just make sure you don't have a Michael P microphone and can generally be heard and you speak in a clear tone
well when i train it sounds like trash bu ty
it works with other ppls models tho so cant be mic
If you have like an audio file of the vocals you want to process, it can be a bit low quality if there's reverb artifacts, harsh plosives and mouth clicking and other mouth noises, instrumental artifacts and so forth
Oh you can tell by hearing if the vocals either sound detailed and clear or like a 2007 youtube video
where is the colab links?
Ayo? @tacit edge level 7 !!! 
thanks
Are you going to make the tutorial one day?
I will! I don't have anything planned today so I might make it later on today on how to make RVC vocals sound better. I hope to make it as free as possible but there's a lot of paid VSTs out there but they are very good too
where is the ai cover colab????
it came off a real movie tho somewhat around 2016 i think
it is just making datasets i guess
sounds ok but im the worst person to ask
Oh awesome then! Thank you!!
Then make sure it's on good quality
Sonically
@hollow tangle yooo project is fire
I'm putting all my personal data in that link rn!
Hi everyone!!
Hi!
hello
Hey question about making my own ai voice, I can speak more then 1 language (I speak 4 languages to be exact) could I put this in one big ai model I'm making or do I have to make different models for each language, new to this so im sorry
Rvc models speaks by of any verbal
So you could speak any language you want with a ai model if that’s what your talking about
wait let me break this down
Ayo? @lusty dust level 1 !!! 
I speak 4 languages
I put 5m of english talking and 5 more singing
then I do the same thing with the other 3
then I train the model of those 4 languages
Could that be possible
Yes you can. RVC doesn't recognize any human accent, it just mimicks the sound of an audio. There are AI covers of an electric toothbrush and Minecraft zombies 😭
Air ai model
Yeah exactly 😭😭😭 that's actually possible I believe
alright cool, just making sure because it would be weird if me speaking spanish and then it makes it sound japanses and polish
Restaurant background noise ai model
Spirit Halloween ai model
Just screams and yells from the animatronics
Has anyone made a car ai model
Like a actual car
Don't worry about that either. I used an ariana grande model to make it sing a spanish song she did it perfectly.
It's insane
ok well how do I make the model since I recorded like around an hour of each language
speaking and singing
You could use something called instant dataset maker
32 HOURS WTF BRUHH
-rvc
Hey, @lusty dust!
👇 Here are some useful resources to help you learn how to make ai covers
Translation by country
Oops I read that as dataset
whats that one fnad fan made thing where a woman dies bc of a blue bunny and then she is put into a sheep suit and also she has no head or sm
the bunny rips her limbs appart
apart*
somoen tellof
You can make a fart sing Any language
Hi
someone help me
Also my accent mildly changes between languages to accommodate for the diffrent words
When RVC3
tomorrow 
RVC4 when
Have you checked on #1159289700490694666 or weights.gg?
Pls
what channel is that
no i dont wanna create one, i meant what program am i supposed to use them in
im new to this
do u have a site where i can clone?
like i have an mp3 and i want to make a text with the voice from the mp3
any romanians?
I recommend Ilaria RVC
Or wordband
Discord died
hello
Hi
Discord issues
Can I use rvc for real time voice changing ?
Yes
Hi
You would use the realtime version of RVC or use Okada
Guy's can you tell me how much time gets in Ilaria AI cover maker
@wintry sinew you know how when you order food you can say "we are gonna order some Chinese"
it depends
Same in Dutch
on how much long is your audio/if u duplicated the space

And I wish we can put YouTube links too
justs download the youtube video
u can use ai-cover-gen-nowebui for that
never tried it tho

Yo
reminded me i need to test somthing
Hi is there like a pinkie pie and rainbow dash voice model here?
i feel useful now
lemme see
733727362727737283747
Hey discord what song should I have donald trump singing?
what are ppl using to do the song stuff
RVC

