#๐งฌโai-chat
1 messages ยท Page 190 of 1
its speech-to-speech only correct?
use a site like exposed.lol, get persons information, use realtime voice changer to act as their sibling or family member
say you're in distress and you've been kidnapped, the only way they'll let you go is if you text a 500$ giftcard to this number
thanks chat
?????????????????
if you want to clone voice.. you need a voice model. voice model is basically a file that stores the voice data of your character. basically a .pth and .index file
I think there are some TTS too
Hmmmm
oh sorry guys
is RVC good at TTS compared to the other TTS frameworks?
wrong discord oops
like tortoise, styletts, etc?
if you use Qubes VM on a RDP, with Mullvad VPN on multi hop (enterance sweden, exit anywhere in the world)
not even the FBI can trace you
oh oops guys sorry wrong discord
sorry chat wrong discord
mr show off just delete ur msgs
sorry i dont speak to people who play FiveM
i can't use it because it doesn't have a record button but a download
what ?
which program
answers like a bot u are
real time?
i don't know.. I never used TTS
you guys do mainly singing stuff here right?
no
Ayo? @elder willow level 3 !!! 
what do you guys do here then
we do everything
I can't send photos for some reason
including TTS? cuz most people i speak to here dont seem to know much about TTS frameworks
if ur gonna use TTS just use something like ai.key its free and has alot of custom trained models you can use for free
i need to be able to train my own models
it isnt a tts to begin with
real time dont do TTS
do they have an API?
but you can pipe any existing tts output into rvc
basically voice to voice conversion
no..? you wouldnt need a API unless you're doing development work
thats what i assumed but Dr. Jr said there were TTS models built on RVC
why would you need a API lol
RVC was speech-to-speech from my understanding
yes
why do you need a API
i am doing development work
integrating into project
whats your project

no .. I didn't said that .. but you can use it in different forks such as ilaria tts
I just need high quality TTS, doesnt need to be streamed so no need for realtime, previously used elevenlabs but im stepping away from their API
it just converts existing tts output into rvc
tts ran through voice changer
ya elevenlabs is overly expensive but
whats your use case
are your users going to be generating their own text
yes
my apologies if i misinterpreted
yeah thats what most people i see do
taking output from tortoise or styletts
and plugging it back into an RVC model
the grind for verified is unreal
Ayo? @dire moon level 2 !!! 
ur welcome jit TTS on RVC
does this support cloning/training tour own voices?
not a problem
theres at least 32 voices im gonna have to fine tune on
not gonna be fun lol
vast.ai will be eating off me
okay.
it supports RVC ur customers would need to create their on RVC models to import
if you want that I help you.. ping me or DM me
if you have 32 voices atm then they would work perfectly with edge-tts
how does the emotional range/expressiveness of edge tts compare to that of elevenlabs?
Ayo? @orchid wigeon level 3 !!! 
obviously nothing compares to 11 atm but im wondering how vastly different it will be
and just for context this is what we were looking at using:
Ayooo
thats based on ur RVC model
what RVC do you reccomend?
you can
im aiming for quality, speed isnt a priority in my usecase
shit
edge is not meant for emotion
just text
pretty flat in emotion. Your better off using openvoice
thats why ive been looking at tortoise/styletts2
Ayo? @versed loom level 4 !!! 
Create a training dataset: Use the zsxkib/create-rvc-dataset model to generate a dataset of speech audio files from a YouTube video URL.
Train your voice model: Use the replicate/train-rvc-model model to create a fine-tuned RVC model based on your dataset.
Run inference: Finally, user the zsxkib/realistic-voice-cloning model to create new speech audio (or even songs) featuring your voice.
hey
what
thats open source tech he can use/implement to his program for his customers to be able to train+create rvc models
i see, gonna look into all these
@brittle jay check your DMs
customers arent actually creating any models, just using them
creating them doesnt fit well into our use case, we handle all that
morning
Ayo? @signal cairn level 57 !!! 
Spotify usually doesn't sell songs, only gives you access to it
and that song is from a girl and old af
not sure if you might find it for purchase
Hey guys what happened to the copycat.actor ai? The website disappeared and it is no longer recommended on this server unlike before
how can i connect the voice changer to my discord input?
most websites that were removed
its bc they either became full paid
or stop working as intended
yo anyone know how to use vanity
wdym
Ayo? @elder willow level 2 !!! 
check DMs
anyone have a really good egirl voice model?
so like which one stayed free then? I wanna make some covers eventually and my brain is smooth like a turkey
This is such a beautiful song, so I really wanted to try covering it! I used a mix of Miku's dark, soft, and original voicebank. I'm really excited for the Children of the Sea movie~ (this is the movie's theme song)
Cover + VSQx + Illustration + Video: cillia/kyaami
Original Song: Kenshi Yonezu
Instrumental: ใใคใซใฉใตใฆใณใ https://www.youtube.com/wa...
Hazme el favor de ver el video en 720p HD
Me gusta mucho esta canciรณn asรญ que hice este cover para probar a Natalie en espaรฑol tambiรฉn.
Espero te guste este video.
Natalie de Dreamtonics: https://dreamtonics.com/
Canciรณn original por Los รngeles Azules y Marรญa Becerra: https://www.youtube.com/watch?v=CaSCapHZhc0&pp=ygUSZWwgYW1vciBkZSBtaSB2aW...
Commission me!
https://ryutsuai.carrd.co/
โ
Credits:
โ
SVP - ryu
โ
Song - Nessa Barrett
โ
Mix - ryu
โ
Tuning - ryu
โ
Video - ryu
โ
Voices:
โ
- Kasane Teto
โ
- SOLARIA
Socials:
Kofi: https://ko-fi.com/nonka
Patreon: https://www.patreon.com/ryutsu
Twitter: https://twitter.com/ryutsuinthereal
Soundcloud: https://s...
I'm so thankful for all the fantastic comments on my last video even tho a long time has passed here's another jazzy cover from me, I have a lot of undone svps lying around so Imma probably gonna finish them and post but then again no promises cuz school exists. Please tell me if u like how I used Asterian here! He's a pretty nice vb but I wish ...
its legal too ๐
is new fcpe is good
Wrong chat
.fm
..And The Addled Abstraction Of Being
Psychedelic Porn Crumpets โข High Visceral, Pt. 1
did we definitively figure out if resuming a train at 50 epochs messes up a model?
no ?
expirimenting on it rn, making a model intentiooinally resuming at 25 intervals to see if artifacting occurs
local models seem to "make" better than collab imo
I don't see why it would mess anything up
it recalcs and saves over the most recent checkpoint for a resume
and the graph shjows differing results between states
tru, I am intentionally breaking this process to rule out a creation headache for a tutorial
reduction of smoothing only exemplifies my findings
does it save the models on an exact stepcount ?
or does it save in between them ?
if it's in between, you won't find the value on the graph
at chekpoint amounts chosen by user, most common to choose is 5 or 25
Ayo? @unborn tundra level 2 !!! 
epochs aren't steps that's the thing
it actualizes tensorboard every 2k steps right ?
wait no
"log_interval": 200,
does anyone have a software for ai models on mac ๐ญ ive been sent some that said they were for mac but the files arent compatible w mac
i do
use google colab
most stuff isn't compatible with MAC
ive tried and most the times it says sm is invalid
How many epoch is actually good?
https://aihubdocs.github.io/en/ check the epochs page to learn
Last update: Feb 24, 2024
should i remove reverb from inference audio too or just training
For both. Reverb or other effects impact the results in RVC
thanks!
how do i instal the ai thin
hey can youhelp me i need google colab
It keeps giving errors when I use it
tell me aswell if you found one lol
cool
hi, is it possible to lip sync an already existing video using ia ?
Yuh deepfakes
where can i buy high quality voice models
where can i buy high quality voice models
was more thinking like using an already existing video then using the same voice but making the person saying something different
deepfakes feels unnatural so idk
Find words that sound the same/similar to what they are saying
and pace it similar
its for speed and mainly for realtime
Guys, I do some concert covers, I wanted to ask if you have any advice on how to get sound from an audience, like fans screaming wildly. The Michael Jackson show, for example, had fans who fainted, but anyway, I say don't keep using the same thing.
Universal Vocal Remover should be able to seperate "other"
Ayo? @unborn ermine level 30 !!! 
Level 30 W
thanks ๐
I made Michael Jackson singing circus with Britney Spears, but the background looks very similar hehehehe
check my bio
i mean if you want a high quality one you either make it or pay someone to do it xd
they don't sound bad, but the audio samples actually made me cringe out of my mind
xD
Also I feel like ya'll trying to sell voice models would have more sucess if you had much longer samples. Like up to 10 mins of you reading something with the model, and telling people what settings you used in the voice changer, along with any filters or vst plugins you used on the voice afterwards.
As is someone could buy one of these and get ripped off pretty easily
Like every single one only has like 10 seconds as a sample at max, when you're trying to sell this shit for 20 euros or more (you're not the only one who does this. it's like every person trying to sell "egirl" models)
kinda fishy to me
sure 10 mins then someone can just clone the model lol
you can send me any audio and i can show how it sounds on the model tho
if we simply had long as fuck samples anyone would simply use it to train the model themselves
so its pretty obvious why the samples are short indeed
not a good idea. with long samples people can make the model out of it
15 secs should be the max imo
its literally "show me your product so i can copy it"
Kinda doubt that the model that came out of raw rvc output would sound as good as the thing trained on a real person
well it wouldn't sound the same but still
epochs doesnt automatically mean better right
people who dont wanna buy it would just do that
no
Bcuz like if it was that easy to get good sounding models no one would pay for that lol
like if u made a model at 500 epochs, it could outperform a 1000 epoch
too much epochs usually means overtraining
ah
yoyo,does anyone know a good free video upscaler?
https://aihubdocs.github.io/en/ check the epochs page to learn more
most people don't know how to clean them properly and wanna something done
Like your market in of itself are people too lazy to train models themselves so, y'know
long samples aren't gonna hurt
.....
you're literally the first person to complain about that lol
no
free and video its pretty difficult
not even one?
unless its something local
i have amd gpu
AMD = Shit for IA
i didnt know that 3 y ago man
hey, if it's working out for you then chase that bag
i mean most of my sales are custom ones
not everyone wants a generic voice, most end up wanting a unique one
or some wav2lip that doesnt convert your video to 720p
wav2lipHD
collab version?
AI developed very fast, and NVIDIA CUDA was already the thing for all things AI so I cant blame u
but wav2lip itself isn't that good
what are you using wav2lip for tho?
sync languages?
wdym,lip-sync
to sync lips to audio
is there anything else that i can use?
seen some services but all of them are paid
i mean the video purpose
dubbing on multiple languages?
oh
well considering there is nothing else left to do,imma try it
btw do you have the wav2liphd collab version?
I managed to install my 2080ti into my gaming rig and somehow it didn't break everything
Noice
I was fully expecting the drivers to just absolutely die
But nope, both my 6600xt and my 2080ti are working fine
Oh, you have both in at the same time?
Yeah
๐น NICE
I love hardware so I thought I'd see if I could get it to work... and I did.
Well one advantage is that they can run independently
They are
You could play games on amd card and then u can do ai stuff in background on the nvidia fully
That's what I'm doing
I prefer gaming on the 6600 and the 2080 isn't even outputting to a monitor, it's just connected.
Personally, I think it makes more sense to do that with the exact same card or atleast similar brand b/c the performance of one might be better
I havent looked into the newer amd cards
idk, their github would say it. probably not if its AI
The 2080Ti almost guarantees better gaming performance, but I'd rather use me 6600 for that.
Are there any rules about what voice models i can make and post
Cuz im making an austrian painter one
can someone help me with training in applio
A mod flipped out over my pfp wearing german soldier outfit (no swatzika) so I would assume they'd not be happy if u posted that here
(it was a sticker)
But also, that model already exists im pretty sure. check weights.gg
the training part stops for no reason
I'm currently alone in some hotel room like far from home rn
Ayo? @distant thorn level 2 !!! 
There is some Nickelodeon blasting on the TV in here the TV has been on for like 16 hours straight
.
Like ever since near midnight last night
Hot DAMN
Everything is working perfectly
Gaming works fine, stress tests work fine, and the second card running the ai works fine
Never have I ever thought I could get an NVIDIA and an AMD gpu working in the same rig with all the drivers installed

can anyone help me get musicgen going?
guys i have 4060 ti 16g
and voice didnt work
why
cpu core i5 12400f
16 ddr5 ram
what
what is that
this gif what can i do
can anyone tell me how to make ai voice models?
Can anyone get the delay out of the voice?
I have a RX 6600 XT i5 9400f 16GB ram
bro im 4060ti and i512400f and i dont get any voice at all
If you are running on cpu, your gonna have to have a delay
On my 3090 I can get 200/300ms tho
Gpu and Cpu is bad, this one with a super delay and cutting a lot.
Most of the time it stops picking up
only nvidia gpu is supported
While technically there is a way to get it to run on AMD GPU, its like 30% less performance and stuff so :/
and it requires linux and stuffff
The problem could be that it's AMD itself
The graphics card is not bad
Ayo? @whole verge level 1 !!! 
AMD doesnt support CUDA. Its all built on CUDA. I had an RX580 before my 3090 and it worked great for most things, most games. But it doesnt support AI b/c it doesnt support CUDA
Which is why I got a 3090
now I can generate all the anime women
@unborn ermine
Hi is there a video tutorial for this
Ayo? @tacit heath level 1 !!! 
Do u have nvidia gpu and are u wanting live voice changer or thing to make covers with
can someone help me with voice change,its not hard,i need only 5-6 minutes of your time please
@kindred ivy use #1159290752195633273
ask in #๐โhelp-w-okada
good ig
wut
For anybody still interested (I know there are none of you): The 6600xt and 2080Ti (in the same build) are having 0 driver conflicts and nothing impedes daily use/gaming
Just by default.
remember, AMD is for graphics, if you want to use AI get a real card like NVIDIA ones
I'm using AMD for gaming and just display and the 2080Ti is being used for okada
The 2080 doesn't even have a display port or hdmi coming out of it rn
It was in my old rendering build that I recently retired
My 3090 heats up my room lol
you might be better getting a 3060 12gb
for OKADA
2080ti is a good one but energy usage is pretty heavy
yea
If I already had one I'd obviously use it instead, but I already have the 2080.
not sure if the energy cost is worth it in the long run tho
fr
Meh, I'm not really going to be utilizing it 24/7 anyway, or even at max
Hell, I've had this thing running for hours and hours on end rendering
Ayo? @elder willow level 4 !!! 
Running it like this is already saving money
Yea idk, its the same argument used for electric cars
Nope
My town car gets 12mpg most of the time (except highway trips) but I havent put that many miles on my car
Electric cars are "Oh yeah I don't use gas so I'm a better person" while using power that comes from fossil fuels primarily
My argument is that since I'm not using it for as long, or as intensively, I'm using less power.
oh its not for the enviroment
i mean mostly on your energy bill xd
Well they throw a lot of arguments at u to try and get u to buy the giant iphone
fuck Apple
I'm gonna break something I stg
Im working on making a John Frusciante model
I wasn't trying to fucking make that argument, I'm costing myself LESS by not using it for rendering
itโs funny bc before the 16 dropped my 2018 SE been working fine
once it dropped shit is slow
When I said giant iphone I was talking about tesla ๐ญ
itโs inconsistently slow too
oopsie
Lol
I feel like phones arent built as good as they used to be in general
the nokia effect
#sigmarulememes , #alpha, #finance, #andrewtate, #redpill, #entrepreneur, #motivation, #fitnessmotivation, #riseandgrind, #grind, #sigmamindset, #grindset, #money, #passiveincome, #income #sigma, #motivational, #riseandgrind, #alphamale, #billionaire, #cash, #tips, #entrepreneur, #grantcardone, #tailopez, #mogulstateofmind, #garyvee, #buggati, #...
Why is truth hard?
And why is royce speaking it
Its such a hard truth, its blocked by UMG
Umg moment
I see them block videos that don't have musix
This guy is hilarious fr
where can i submit my model?
how much is the BVE v2 in x-minus pro?
People: Ai will destroy the world
Me using Ai: Haha Michael Jackson is singing Smash Mouth All Star
Sleep stream come wake me up
@pearl linden use #1159290752195633273
Exactly how I feel too
Didn't some Ukrainian youtuber get her likeness stolen and used to create fake videos of a Russian person living in China
Ayo? @zinc mulch level 1 !!! 
the more epochs the better right?
how do i know which model class to use when loading the .pth file
No
How do I download the voice synthesizer?
never
and voice.ai is shit
Where's the best place to train my data, I tried on my gpu it said 3 days, i left it on for 1 day, but my PC eventually crashed
whats your GPU?
2080 ti
well you're either doing something wrong
or trying to play while training
beware of batch size too
batch size was 29
too big and it will crash your entire PC
why
29 is fine no?
well the size was from 1 to 300
Ayo? @fiery crypt level 1 !!! 
ill try one more time, but 3 days is a bit much to leave my pc on
when i was at school they let me use their SCC, shared cluster and this usually only took 3 hours
alright ill try it
are there any models, that aren't singing related here, that I can use for voice-overs on tiktok
it wont need 3 days
unless you're using like a 2 hour dataset
check DMs
@fleet ingot 
why are there two RVC's? and why only on one you can make ai models?
Ayo? @jade musk level 2 !!! 
like there's one called RVC-beta and another called RVC-GUI
rvc-gui is very old
it was made when rvc v1 was still around
dang
Murder Drones Uzi
yall like elmo
How and where do i set epochs
Its part of the model, you only need to worry about it if ur training
oh
.fm
How does one use this?
.fm
Deference For Darkness
Martin O'Donnell & Michael Salvatori โข Halo 3: ODST Original Soundtrack
yall whats a good site for doing ai covers
Ayo? @grave acorn level 1 !!! 
Don't use free AI sites
why not
They are bad
but some of them were good
there was one on here that was on hugging face that was pretty nice
Huggingface is fine, but it's better if u run it locally
any decent free ones though?
@lethal swan 
how much are the ones that cost money
@lethal swan @fleet ingot 
do you also have your .index file?
yes
Rmvpe 
i see that theres the applio on https://aihubdocs.github.io/en/rvc/local/mainline/ , but is there a code version of it? I want to add it into a bigger codespace
alright, i've trained a model with harvest and it took a while and i didnt realise that it was unsupported
lol
harvest isn't unsupported, they just don't accept harvest anymore. So whilst you can still train harvest models, don't expect they'd be accepted.
ok
Ayo? @elder willow level 1 !!! 
is there a way to TTS synthesize without using the UIUX?
Ayo? @fiery crypt level 2 !!! 
I want to add it to code
anyone expert here
i am new to this, it seems like every model i import sounds almost the same
without getting even close to the sample soundtrack
skill issue
is there a way to get multiple outputs at once? or streamline it?
are you sure they are different models, and not the same single model, or of the same person, also make sure that the indexing isn't set to a low value.

i put the indexing to 1
yall think harvest is good or nah
Ayo? @pastel bluff level 1 !!! 
Guys how do I get the jjk narrator voice
rmvpe is a way better version of it so not really compared to that
Is there a model
For the jjk narrator voice used in memes?
Please dm me if you have any idea
Ayo? @weak talon level 1 !!! 
hello
Huh?
i do have it, but it is a private model

only if you have studio vocals
Ah I see is there any methods to get it or it's just off limits
Ayo? @lusty valve level 1 !!! 

That's amazing I'm ngl
I forgot what he said
Just use crepe or mangio crepe if you have studio vocals 
Just trained my first model today. Any tips that can improve on trainning the model to re-create laughter? I've only seen one model which can do it so far.
๐ฅฒ question, which ones actually can reproduce decent laughter?
homebrewery
Bro someone came up to me and called me a d*** for using charlie morningstar's ai voice, and said I stole itโ ๏ธ
In vrchat.
I mean its ai cloning related of course people will say that
They said it caused million's of dollars and it should be a illegal, and I kindly explained to them that I'm not benefiting from it.
Does anyone have good ref base clips to test new ai model with? Like for demo.
singing wise or just talking?
I'm having trouble converting my voice to AI, the new GUI and UI have all changed from the original EasyGUI making it waaay slower and it keeps my background vocals on the output. Anyone know who to combat this, aswell as get the clean output of just the AI
help the google colab thing for the ai covers arent working, i used a model i found here, but it says error on step 2. does anyone know how to solve it
which colab notebook
ah k thanks
Good Morning :3
Fully automated blog posting system with SEO done. It will post articles that you want for your website. It'll cost you fraction of cost to manage everything that otherwise would require you to hire writers, designers etc. dm if interested
It will cost $0.3 to post an article with SEO done once you have it. you can get to 1000 visitors in 45 days.
i used another one and it worked, but the one that didnt work is this https://www.youtube.com/redirect?event=video_description&redir_token=QUFFLUhqbE5vVlBLREtXV1hPRFNEVVNqS0c4ZGVCMmpPQXxBQ3Jtc0trOVdkaXJfYXZlX1VPdjhfd1RQWDdya1V2UkdVdVRLOTdlTXdnUFltLXNyQTdocTllRWNwdm9nbk1TU0ZxZ2lqejBDVF9ZSHpNb1RJeGhVNFRscENGV1BjSko5cTgyNGtNWW4xUmkxazdHY0RGYi1Dbw&q=https%3A%2F%2Fcolab.research.google.com%2Fdrive%2F1Gj6UTf2gicndUW_tVheVhTXIIYpFTYc7%3Fusp%3Dsharing&v=gtjCmJ_6d0k
I dont remember leaving this server? Was I kicked for inactivity?

thats a thing??
getting kicked for inactivity?
Gm :3
???
when was the last time you were here ?
how come when i download something it just shows a paper icon?
Idk a couple of months ago
before october ?
Download what ?
I joined around early fall time i believe
I think so yeaj
how you can get it ๐???
If it was before october , yeah the old server got nuked by discord lol
Ooooh
Ayo? @abstract wyvern level 1 !!! 
shouldnt it be built in ?
You have an nvidia GPU ?
Thats weird
If there's nothing available to read said file, yeah it will do that
Oh no. I remember when the server crashed and then it was back up again later. So discord deleted it for some reason?
Oh that was even earlier
how do i solve the problem
bruh
The kaloscream incident
Discord had internal issues when someone decided to add a soundboard to the server
What are you downloading plugins for ?
I can't guess
music
Ayo? @elder spire level 2 !!! 
fl studios
Yep thats the one. Playing a certain sound crashed the darn thing
You unfortunately downloaded the AMD build for voice-changer then
Direct link :
https://huggingface.co/wok000/vcclient000/blob/main/MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.17b.zip
(from the downloads listed at https://huggingface.co/wok000/vcclient000/tree/main)
Are all the voice models back or are most missing after the nuke?
Most of them are, but there's been some losses
Ouch sorry to hear that :/
I don't use FL, but I think you're supposed to import them directly from FL studio
yeah thats right but the file cant be looking like paper
Ayo? @dapper edge level 5 !!! 
It can
you sure
If FL doesn't tell windows it handles the file format, windows just doesn't know what to do with them and shows a paper as icon
submited my first model
im on mac
alright thank you immagive it a try
Which it probably doesn't
what kind of file exactly? a vst plugin for FL?
guys if im a female and i used female voice how much tune i use?
its a vst plugin
yup for fl
Hello, I did a training session for the model, it doesnโt look very similar, can I do something to make it look like a human voice?
it should be installed through an installer or extracted from a zip file to the usual C:\Program Files\Common Files\VST3 or whatever for another kind of VST (according to some instructions for the plugin), which should be detected by FL and most DAW softwares
thnak you - do you know what installer to use?
refer to the site where you downloaded it for some instructions
make sure every voice in the dataset sounds same, in the same usual pitch range
Do you guys know a good website to train a voice model, for ex if I want to copy my voice or smth like that?
And also a program that modifies my voice in real time but a free one, because I keep finding some that have only paid voices
hi guys
its kinda cracked
but cant find any instuctions
we won't support the kind of that here, so take your own responsibility 
how do i update my app to the latest one
yo can anyone help me. idk what is the best when it comes to training my model?
Which app ?
what is better between pm, harvest, dio, rmvpe or rmvpe_gpu when it comes to training you voice model?
That would be rmvpe(_gpu)
oh ok
Ayo? @gentle pebble level 1 !!! 
they're both the same, but gpu... well runs on GPU so processing time is lower
oh ok
because im trying to make my voice model sound as realistic as possible
what do they mean by Enter the GPU index(es) separated by '-', e.g., 0-0-1 to use 2 processes in GPU0 and 1 process in GPU1
Is there a way to use ai voice using my phone for calls, discord and Xbox VC?
just put 0
ok
that's only useful if you have multiple cards
0 to make it use the first card, 1 to use the second and so on. Can also use multiple at the same time
ok because its defult its set to 0-0
so i just let it be like that?
or delete it and set a 0
me yaw
Ayo? @hexed geode level 2 !!! 
yo when do i know my feature extraction is doen?
terrible site that uses RVC
yo when do i know my feature extraction is done?
"all-feature-done" on logs
i mean i wouldnt say any of these AI sites are good, unless they cost a ton, which in that case theyre only fast but still prob gonna lack alot of features u can get for free if u run rvc on ur own
its just overpriced
on mine it says no-feature-todo
you screwed up on something then
thats true
what are ya using
for the training thing?
yeh
if you're using RVC disconnected then you screwed up on the dataset formatting part ig
im using gradio.live somthing
oh wait are you using it locally
i saw i video tut
yeah
Ayo? @gentle pebble level 2 !!! 
.wk
๐โโ
โฆpharmaโฉ - 453 plays
โ
2.โโ
โฆcharlโฉ - 314 plays
โ
3.โโ
โฆnepโฉ - 263 plays
โ
4.โโ
โฆuh1cโฉ - 237 plays
โ
5.โโ
โฆauthenโฉ - 196 plays
โ
6.โโ
โฆyeresinโฉ - 187 plays
โ
7.โโโฆagentโฉ - 186 plays
โ
8.โโ
โฆfantaโฉ - 179 plays
โ
9.โโโฆwraithโฉ - 153 plays
โ
10.โโฆkaygreenzโฉ - 145 plays
**โ
11.**โโ
โฆegmond Jr.โฉ - 134 plays
โ
12.โโ
โฆLรผh Shmรผnkyโฉ - 125 plays
โ
13.โโ
โฆprisjโฉ - 98 plays
โ
14.โโ
โฆM4โฉ - 72 plays
.wk lazer dim 700
๐โโ
โฆegmond Jr.โฉ - 265 plays
โ
2.โโ
โฆyeresinโฉ - 251 plays
โ
3.โโ
โฆIsaacโฉ - 194 plays
โ
4.โโ
โฆuh1cโฉ - 59 plays
โ
5.โโ
โฆfantaโฉ - 48 plays
โ
6.โโ
โฆcaosโฉ - 37 plays
โ
7.โโโฆtittylicker47โฉ - 22 plays
โ
8.โโ
โฆkaygreenzโฉ - 11 plays
โ
9.โโโฆHelp Me Rhonda/Funky Prettyโฉ - 7 plays
โ
10.โโฆcharlโฉ - 5 plays
โ
11.โโ
โฆtakingmyswagโฉ - 4 plays
โ
12.โโ
โฆagentโฉ - 4 plays
โ
13.โโ
โฆ๐ด Pluggyโฉ - 2 plays
โ
14.โโ
โฆAdiioseโฉ - 2 plays
Crown stolen by egmond Jr. with 265 plays!
Previous owner: nigga14-88 with 265 plays.
hm...
that can happen due to spaces on folder names or the audio just screws up
i can send a picture of what my presets on
Fmbot
yea, send on #โจโai-help
How to use the models/ Which program do I need to talk?
heyo
hai
hey guys, can someone tell i we can train models on mac m1 using mangio rvc fork?
please use the channels with the correct purpose, your honor.
ok sorry, your honour
it's ok
ok
why hasnt someone made alice from murder drones yet
search there:
- #1175430844685484042
- #๐โfind-models
- https://weights.gg
- https://huggingface.co/models (but watch out there arent only rvc models here)
If there arent any, #1159289738314919936 or make it yourself #๐โmaking-models checking our docs https://aihubdocs.github.io/en
Last update: Feb 24, 2024
wazup yall
google colabs isn't working right or am i doing something wrong?
Is there a japanese voice for gojo satoru?
same happening with me
well did you get written permission from the voice actor?
yeah and im not getting an answer...lol just trying to make a funny video for my wife birthday
!howtoask
How To Troubleshoot 
- Don't simply mention your issue, like "
my rvc is not working". - Describe the step you are on, what you're trying to do, the RVC you're using, a screenshot, etc.
- The more context, the better.
- Don't be desperate. You can ping a Helper, but if they ignore, they aren't available/don't know the answer.
- It's okay if you're frustrated, but don't take it into this server.
- Don't DM without prior consent.
- Don't ask for every little instruction. Put your own effort & test things by yourself.
- Don't ask to ask.
- Check if your answer is a Google search away/on our guides website.
pleas five more information
lol no one is complaining or causing any issues here,we are just wondering why something isnt working as intended, mind you text has no tone.
im just asking for more information
thing is, google colab could be anything, it's not exclusive to any software whatsoever
let me try and remake it and tell u what it says when i do it
๐
JSONDecodeError Traceback (most recent call last)
<ipython-input-5-d4621d05d8da> in <cell line: 31>()
31 if os.path.exists(config_path):
32 # File exists, proceed with creation of creds and client
---> 33 creds = Credentials.from_service_account_file(config_path, scopes=scope)
34 client = gspread.authorize(creds)
35 else:
5 frames
/usr/lib/python3.10/json/decoder.py in raw_decode(self, s, idx)
353 obj, end = self.scan_once(s, idx)
354 except StopIteration as err:
--> 355 raise JSONDecodeError("Expecting value", s, err.value) from None
356 return obj, end
JSONDecodeError: Expecting value: line 1 column 1 (char 0)
Ayo? @inland spade level 1 !!! 
what are you even running ?
im trying to run a singer for an ai cover but when i put any artists model it dosent do it and says that
@wintry sinew its saying this when i try to do step 2 in google colabs JSONDecodeError Traceback (most recent call last)
<ipython-input-6-6006d452c649> in <cell line: 31>()
31 if os.path.exists(config_path):
32 # File exists, proceed with creation of creds and client
---> 33 creds = Credentials.from_service_account_file(config_path, scopes=scope)
34 client = gspread.authorize(creds)
35 else:
5 frames
/usr/lib/python3.10/json/decoder.py in raw_decode(self, s, idx)
353 obj, end = self.scan_once(s, idx)
354 except StopIteration as err:
--> 355 raise JSONDecodeError("Expecting value", s, err.value) from None
356 return obj, end
JSONDecodeError: Expecting value: line 1 column 1 (char 0)
can you share the link of the colab you're using ?
same there
hes using the old kalomaze download one
what's up with that one all of a sudden
if you see JSONDecodeError its always that
look at the download and you will see why
๐ญ
whats the link to the new software?
-colab
- Applio, by IA Hispano Google Colab
- AICoverGen-WebUI, modded by Hina Google Colab
- AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
- AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
- Ilaria RVC, by thestingerx Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
- GPT-SoVITS (demo), by litagin Huggingface Spaces
everyone's using it all of a sudden for some reason lol
which one do i use?
thank you @wintry sinew
other servers and yt tutorials ig
i need good 1050 ti settings man ๐ i don got no money and no my pc was built by my uncle soo ye is there any good settings for 1050 tis ๐
voicechanger?
try 128 chunk +8192 extra
king bingus
That stuff you posted is just some stupid made up stuff
...just don't
i downloaded applio but i dont know where to put what
@waxen grove can you help me please
i dont think that will do anything that much but ill try
Ayo? @frank phoenix level 1 !!! 
the hell is monitor
what i have to do if I want to become a model master?
okay .. then..
? really...?
Idfk
I assume those who are master have been recognized for their skills and models
at level 20 what will I get like now I'm an Ai chatterbox
Ayo? @queen kernel level 19 !!! 
Level 1: <@&1159345504442925056>
Level 5: <@&1159520205395984475>
Level 10: <@&1159537820352917516>
Level 20: <@&1159537862115598396>
Level 30: <@&1159537896471138345>
Level 40: <@&1159537927278309477>
Level 50: <@&1159537986053099542>
Level 60: <@&1159538018206625932>
Level 70: <@&1159538074439663666>
Level 80: <@&1159538136087527504>
Level 90: <@&1159538241897242664>
Level 100: <@&1159538281512439920>
Level 110: <@&1159538318183243777>
Level 120: <@&1159538343986614317>
Level 130: <@&1159538478669901844>
ohh thanks..
can you tell me about all commands for all bots ?
I don't even know all them lol
But u should get a lastfm account and run .fm regularly
To flex ur music
And u can find people with similar music taste
ohh .. that's pretty amazing..
Vote Booster: Vote now for a 10% boost. https://arcane.bot/vote
โจ Tip: Use /card to modify your rank card
there are a lot of things which I want to know about this server. but I can't.. I need someone who can tell me.
its useless
what useless?
Feet emoji next to your name
Commission me!
https://ryutsuai.carrd.co/
โ
Credits:
โ
SVP - ryu
โ
Song - Nessa Barrett
โ
Mix - ryu
โ
Tuning - ryu
โ
Video - ryu
โ
Voices:
โ
- Kasane Teto
โ
- SOLARIA
Socials:
Kofi: https://ko-fi.com/nonka
Patreon: https://www.patreon.com/ryutsu
Twitter: https://twitter.com/ryutsuinthereal
Soundcloud: https://s...
what is this ??
the best ai voice on the market
ใ้้ณใใใใใฉใฏใซใใฅใผใธใซใซ โ Labyrinth by Miracle Musical Cover Ft. Kasane Teto
Special thanks: https://youtube.com/@mr.fiffles
https://www.youtube.com/watch?v=MdAzl3sOwmY
but I was asking for something else
you dont need anything else except for teto
I think we have a different section for promotions ?
teto owns ai hub
Teto .. who is he/she ? the owner?
hi
it does not have a gender.
what ?
its a chimera
we making it out of UTAU with this one
alright...
what ๐ญ
tbh the only times i hear teto is from you ๐ญ
teto propaganda is real
some people want to make ai cover
did the update for uvr do anything crazy
we made several

Ayo? @elder willow level 16 !!! 
@quartz roost 
its weird al, ive met him. he dont give a fuck
is there somthing to generate music like suno but local
remember we get new voice models first
and ours is huggingface links instead of weights links so it can be downloaded on colabs/downloaders without needing to be reuploaded
Sup guys, can someone recommend me a good voice changer that is compatible to the voice models we have here?
Ayo? @lofty wave level 1 !!! 
w-okada here is the guide
ty so much!
hello
im sure they exist , just didnt have luck finding a v flower model
what is epochs
number of training cycles
You dont really need to worry about it unless ur training ur own model
ah ok
what was the site where you could upload you own model
https://aihubdocs.github.io/en/rvc-resources/epochs-overtraining--tensorboard/ here's everything you need to know =3
Last update: Feb 10, 2024
ๆๆ็ฉบๆณ่ฏ่ฏ่ฎญ็ปstable diffusion็ๆจกๅ ไฝๆฏๆ็ต่ไผ็ธ
i mean like upload it and then also use it to make a song
not the google thing
https://aihubdocs.github.io/en/ Theres a few different options. If u have a nvidia gpu you can run it locally. If not, collab ig
Last update: Feb 24, 2024
Theres also a live voice changer one (local only)
okay
Ayo? @void mason level 1 !!! 
theres also the wokada colab
but not for mobile
.fm
I appreciate it fr
-colab
- Applio, by IA Hispano Google Colab
- AICoverGen-WebUI, modded by Hina Google Colab
- AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
- AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
- Ilaria RVC, by thestingerx Google Colab
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
- GPT-SoVITS (demo), by litagin Huggingface Spaces
i wanna edate so bad
quack
denied
denied by a kitten
ok bud
lmao
the responses are funny asf
fay-ger-bak-ke
Fey-ger-Baak
ok
if I got a better graphics card would running stuff locally be faster
https://aihubdocs.github.io/en/ here are our RVC guides =3
man ๐ what is that
most defenitely ๐
what gpu do you have currently?
why is google colab 100x faster
why is this link not working ๐ญ https://colab.research.google.com/drive/1Gj6UTf2gicndUW_tVheVhTXIIYpFTYc7?usp=sharing
maybe because colab is running on a powerful gpu on cloud ๐ :/?
gtx 1060 ti 3gb, what graphcis card should I get in order to run shi faster
i mean it depends on your budget
best ai voice on the market
3060 iirc
altho you need a card with a good amount of vram
oof it's really outdated
i'd say 6gb minimum
ngl just gonna upgrade my whole shit lol
alr
and no longer in maintenance
tho still, whats your budget?
do you have the latest link could you send it to me plzzz
1k
https://aihubdocs.github.io/en/rvc/cloud/rvc-disconnected/ yup here's a usage guide =3
Last update: Feb 14, 2024
yeah you could build a pretty good pc with that budget
thanks
Ayo? @cunning merlin level 1 !!! 
yw! any questions ask in #โจโai-help
thank you so much๐ญ๐ญ
my dataset is 20 minutes long and some samples don't have really clear audio
do you think it'll still do good
Dataset must be clear in every part
You can try, but i recommend to cut the unclear parts for a better result
is it not available for Mac?
for it to be faster than colab? 4070ti
nah like similar
Ayo? @glossy jungle level 2 !!! 
dont wanna overkill it just for some ai voices
Which thing exactly?
i mean its a fucking T4 with 15GB VRAM
Hm
so you either go overkill or go nowhere xd
for training?
hi maki
inference wise you really dont need that much
yes only for inference
3gb version is rip
facts
https://youtu.be/V0MTaZaEQ_k?si=AeA27Yfby7aWoAyE this works for mac
the yes
Download Monone: https://lunaiproject.uwu.ai/#monone
More singers for DiffSinger on LUNAI Project: https://lunaiproject.uwu.ai/
ใปsong: Ariana Grande - yes, and?
ใปvocal: Monone v105 for DiffSinger
ใปcover: keirokeer
ใปillustration: uriyana_yuku
ใปabout me / contacts: https://keirokeer.carrd.co/
midi, ust, tuning, mixing commissions: https...
how do i find turkish voices?
TTS? or RVC models?
Guys i need help
i was using w-okada voice changer and i setup everything and its all like the vid but i click start and my voice dont change
dont use the vids
read this: https://rentry.co/VoiceChangerGuide
Other Links
Antasma's Local Error Fixes
Antasma's Colab guide
Sushi's useful Links - You need to follow this if you are on an AMD or INTEL ARC graphics card
Frequently Asked Questions
Jump To The Important Sections
Other Links
Download NVIDIA
Download AMD/INTEL GPUs
Download Mac
Online Hosted Alt...
Reminder: this is not a help channel. Use #โจโai-help instead
what is S.Thresh
can anyone help me with W okada in the #๐โhelp-w-okada channel
Does anybody know what's the best ai photo generator? For free/crack?
Guys how do i get more coins? Exept buying vip
#๐โhelp-ai-art stable diffusion is the best prob
stable diffusion
Yuh, theres different forks of stablediffusion. I use Automatic1111
sdxl better
It has sdxl support too
yes
I guess but I use colab ๐
I use colab because it won't work on my laptop T-T
there's also stable diffusion.cpp which is great in terms of performance
literally running on a CPU faster than some GPUs
obviously using ggml
@~@?!
but it's kinda missing features
I don't understand smart words, sorry
still amazing
sorry
last time i checked automatic1111 is the best 

did you check anything else?

no ๐ฅ

my Linux distro is the best because it is the best
my IDE is the best because it is simply superior
mid
I have more
I have 600 gigs of models alone 
You're creeping me out T-T
:(
comfy ui
I'm not here
I'd rather drown in my tears
I use Foooocus, comfy, sdcpp and a1111 depending on what I need
Idk if it's the right place but I hope some of you understand in PCs
I have a fairly new laptop (Asus vivobook if it's matters) and recently from just one moment my mouspad clicks just broke
It's like the mousepad sank inside and the clicks don't even exist, I can just push it deeper in the computer
Does anyone ever stambeled across something like that?
does anyone here know how to make python checkers for netflix,spotify?
i could give you some of the working accs
or if you know someone that could help me
Ayo? @glass dust level 7 !!! 
tbh I feel like if you try to login to a bunch of accounts back to back, successful or not, itll get rate limited or flagged
Anyone know how I can clone a voice for text to speech
-tts
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
-audio
- Perfecting Audio Isolation on Low-End Rigs: A Practical Guide, by Litsa The Dancer and Faze Masta Google Docs
- Gathering and Isolating Audio, by SCRFilms Google Docs
- Creating Datasets for RVC using iZotope RX, by LUSBERT
Rentry - Vocal Mixing Tutorial, by Roomie YouTube
Audio Separation/Isolation
e
g
how are u guys
im finee, thanks :>
thats nice
nice

yo bro i need help
the google colab has this error when i use the hugging face link
if u know what im talking about
this the error
JSONDecodeError Traceback (most recent call last)
<ipython-input-6-d237e45f065a> in <cell line: 31>()
31 if os.path.exists(config_path):
32 # File exists, proceed with creation of creds and client
---> 33 creds = Credentials.from_service_account_file(config_path, scopes=scope)
34 client = gspread.authorize(creds)
35 else:
5 frames
/usr/lib/python3.10/json/decoder.py in raw_decode(self, s, idx)
353 obj, end = self.scan_once(s, idx)
354 except StopIteration as err:
--> 355 raise JSONDecodeError("Expecting value", s, err.value) from None
356 return obj, end
JSONDecodeError: Expecting value: line 1 column 1 (char 0)
what the hell
use a newer colab

