#š§¬āai-chat
1 messages Ā· Page 167 of 1
Ayo? @brittle nova level 1 !!! 
Wdym way worse
super staticy and keeps cutting in and out
hi
Make sure your sample rate in the same for all audio devices your running audio through
If its massively different, itll be choppy like that
I had similar issue when trying to record + listen to vinyl at the same time via VoiceMeeter
ended up just coping with virtual audio cable instead
But my bitrates were wrong
how do i check?
using virtual output on everything
but discord sounds trash
while obs sounds good
anime women are real
@hollow tangle i have successfully runned gpt sovits locally
yoo
jjl
hi
hai
heyhowareya, I'm looking for a model that can do a version of "similarity search" when given a video clip. For example, if I have let's say 2 hours of video footage, I want to find 20 clips of that raw footage that match a given video clip.
Hello
like getting only someone talking from a long dataset?
heyo
hai
not necessarily, I think more like from a list of b-roll video when something logically makes a 20 second video that can be shared to social media
The current time is 10:42 PM
React with š or š to give feedback.
i cant use the "-colab" anymore
cuz there are no channels that i can type it in
-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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- 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
nvm
hello
I am now introducing myself

rejected
f
:,0

This guide is back to receiving regular updates, however do not ping me for anything on it. The guide itself speaks for me. - Blanc
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
Frequent...
Sorry for the late reply btw
This should tell you what you have to do ;)
what the hell is an epoch, is more or less better?
An epoch is basically how many times a dataset has been trained fully.
E.g. 200 epochs
Means the dataset has been trained fully 200 times.
more or less does not mean better
And more means it's closer to the provided dataset
But if the dataset's bad, it'll sound bad
interesting
If you do too many epochs it can also lead to overtraining
Which would make the model worse
whats a good range for simple sentences and/or for singing?
It's more like, overfitting
This really depends on the model
Every model's different and we can't say for certain
200 epochs could be massively "overfitting" for a longer dataset but 200 epochs might be "underfitting" for a smaller one
So yeah
i see (kinda)
Ayo? @formal burrow level 1 !!! 
Bonjour
hai
uwu

In Uvr for covers was a good Vocal Split?
like instrumentals?
for seperating vocals and inst?
or
2 voices
I want to grab a model, grab a song, and make that model sing that song
try ai covergen colab ig
ok
-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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- 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
Nah for back vocals and lead vocals
Want to ask the same question for vocals
Cuz I know uvr have a vocals split options
try 6_HP_Karaoke i think
ya ne ponimay vas
hello
hai
applio has real time voice changer?
hola :D
for realtime use w-okada
bro
w-okada its from github?
thank you
no its closed rn
nvidia
thank you for the link
here is the guidehttps://rentry.co/VoiceChangerGuide
This guide is back to receiving regular updates, however do not ping me for anything on it. The guide itself speaks for me. - Blanc
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
Frequent...
:0.
is coverting the back vocals a good idea?
separately?
ehhh if they aren't layered as harmonies maybe
if you can sing a bit you can recreate them 
Sometimes

is the colud ai colab still works ?
ty to the person who made a bslick ai
do you have server discord Ai Hub France please
it was deleted
Do you have a similar server with French voices?
https://www.weights.gg/ has a french tag
Have anyone here try the UVR Bve?
hi
hi
hai
I used it when Hp_5 didn't do the trick
for harmonies
bro both the windows version and the web version doesn't work for me
Sup
What are you trying to do
It exists for a reason
Just wanna remind some of yall that I joined this server and wanting to learn and figure out where the all animated character ai voices are from originally that people are using and not Uberduck
RVC
Models can be found in #1175430844685484042
Or on the website weights.gg
lol
Ayo? @hidden bolt level 1 !!! 
Link that to me please
Ayo? @elder willow level 1 !!! 
Thanks
On the shelf
You can Legally profit with them
Whatās the fun if itās legal

merhab
Anime
is there a sean connery voice?
someone should make a sean connery haha
hi
Hello there
Hi people, Im actually a bit new to this cuz I was in Kits ai before
But I need another way to make more AI Models for my covers cuz Kits ai became a paywall
Oh, you said make models. Look at the guides for RVC disconnected
HI GUYS, how do you delete a voice from the library?
Alright thanks ^_^
I'm gonna have to wait until next month for audio conversion minutes to reset on my kits-dot-ai account. I paid for the converter plan, which costs, like 10 bucks per month.
Oh my god... š
You could've gotten much better for free
Heck even paying for it too you could get decent stuff for probably less
Bro got scammed
Imagine paying for kits
Good evening guys, how do you download the model in app.kits?
Like how did Kits gone from "Free for everyone" to "LEMME MAKE MORE MONEY"
Ayo? @coral harbor level 1 !!! 
It went from āOk RVC online app thingā to ācorporate ai cover scamā
fr tho like aint no way Im paying for 10 dollars
whats the news
hey i need help how do i use it on discord?
also i have a question
RVC works on amd?
it wouldnāt use your it and instead would use your CPU
So yes for inference, no for training
how do i use it for discord?
oh i just wanna use the existing models
Ayo? @cedar portal level 1 !!! 
use weights.gg, Kits doesn't have the same training as RVC
can some anime girl pfp call me pookie, idc if you're a guy I need it
Ayo? @gloomy halo level 3 !!! 
I haven't used w-okada in a couple of days... started it up and it's virtually unworkable now? uses up 100% of the gpu somehow and won't process xD
are there any other free alternatives?
or decent ones
what gpu do you have
3080 rtx
It should not use up 100%, that's abnormal
So it would be fine if i was just using existing models right?
yeah, it was strange... used 100% when I had unity and one other game open which was fair - but now gpu was only using 10%
and it spiked to 100% with w-okada
honestly don't think this is gonna work for me to put into real use which is a shame D:
This may be an odd question but what is your cpu? Maybe heavy bottleneck, I'm not sure because I do not have a gpu so I cannot use w-okada but yeah
I don't have a gpu but a couple days ago my friend got a 4070 ti š
50% bottleneck 
well not bottle necking!¬
Ayo? @rugged prairie level 5 !!! 
at least I don't think so.
alright then I don't know, some minor issue back end I guess
you should probably use an alternative
there are some alternatives but not ones that I've found as good as this one
voice.ai worked ok before I installed w-okada, now it doesn't so i'm not sure of any others
w-okada modified something which made the voice not work properly on voice.ai xD
so I'm kinda stuck now
Not sure I'm not exactly a perfessional as I'm using older experience because no graphics card = no client side = server side 
question, it seems i cant access the RVC help section
they moved to #1192011222023950368
ah isee
how do I get verified
Also is there a way to get the rvc training ui on something like vast.ai
I mean if you have network access to the instances yes
By getting to level 5
You level up by just talking
It's a little bit difficult. U need to configure ur docker image and upload ur .ipynb file and open ports on the instance
Ayo? @exotic flame level 1 !!! 
something like this?
Yesn't
ya, that's an .ipynb file but u need to do some changes
xD
Such as?
Not sure which operating system you'll have available but I'm assuming linux
Tensorboard porting
Just gotta make sure it doesn't use anything specific to google colab
Yep Linux
such as mounting gdrive or colab-exclusive filepaths
filepaths.....
I don't remenber exactly but i think is /root/workspace
is there a way to add a display image to my Ai Model?
I have extracted it btw
This is the docker configuration part btw
Thanks so much, with the ipynb I set the "mount gdrive" to "false"
yesn't
You will need to make more changes
In vast you can use the notebooks that are banned in colab
@exotic flame I have this Applio notebook, it's from before Google Colab restrictions, it could be useful
Hi thanks so much
ur welcome
@hollow tangle
hello nice to meet u all
Pls make voice from Lego Ninjago (rus dub)
My video card is from AMD, how do I download it?
hi
hola a todos
Hola! 
Thanks!
does anyone have the cupcakke model? š„¹
hello everyone
hello
where i can send my first ai cover?
Ayo? @dense basin level 1 !!! 
@north inlet kid models are not acceptable
im sorry i didnt know
some people use them for... other stuff
are they weird?
@fleet ingot 
@lethal swan 
hi
guys
Ayo? @cedar pike level 2 !!! 
is anyone still keep the hitori gotoh model
hı
hello
where can i get all the required pipps?
hi
Doesnāt RVC automatically install them?
No
Well, I sent my first AI model of 2024, I'm waiting my model maker role.
Unless the link I keep getting doesnāt have it?
Ayo? @lusty thistle level 1 !!! 
Hi there, Nice to meet you
You too
- Applio-RVC-Fork, by IA Hispano GitHub
- Mangio-RVC-Fork, by Mangio621 Huggingface
- RVC Studio, by SayanoAI Huggingface
- AICoverGen, by SociallyIneptWeeb GitHub
- Replay, by Replay Team Website
- Original RVC, by the RVC-Project team GitHub
You can find more info on the #1159513888199540817 channel. If you can't find your answer, feel free to ask for help in #āØāai-help. Credits to Faze Masta and Antasma for compiling these links.
I recommend Original RVC
Check above
hi
i'm new in all this, but i've tried and for some reason i don't hear anything when i use this thing
(i'm a mac user)
Rvc v2
Ayo? @slim mist level 3 !!! 
okay
https://discord.com/channels/1159260121998827560/1175430844685484042
/search https://discord.com/channels/1159260121998827560/1163592055830880266
or take a look at weights.gg
https://discord.com/channels/1159260121998827560/1191429836321849435
cant find the model you want?
you can commission one from a Model Master š https://discord.com/channels/1159260121998827560/1159289738314919936
@cedar pike āļø
is there a site where we can download voice models that were deleten from the other server? im trying to find the Markiplier voice model
yep thanks
can someone help me whats an index?
A part of a model that mainly controls accent
Goes in the logs folder on local forks
i try to put it in the download
Ayo? @steep apex level 1 !!! 
an it woulnd't let me
For what?
i try putting in the goku voice but it's not letting me use it
On what????
okada
Hi, I just joined, please guide me
okay i have got you
there are 3 bombs in 1
there is the fusion bomb, the fission bomb and the regular tnt bomb
lithium dueteride is the fusion bomb
plutonium-239 is the fission bomb
the regular tnt bomb is self explanatory - but it encapsulates the fission bomb to make sure it goes critical
there is a little capsule in the centre of the fission bomb that becomes larger when the tnt bomb compresses the material - we are trying to get neutrons to definitely hit the centre of the atom
Ayo? @hoary barn level 1 !!! 
Is this like a test for me?
For covers or realtime?
i can give guide to gather materials
how hard is it to make a model and do i need a good mic
Good I guess
Doesnāt have to be the greatest
i got a webcam mic
Does it sound good enough?
Realtime
Ayo? @gleaming ermine level 1 !!! 
plutonium-239 is made by adding a neutron to uranium-238 - and then you force the beta-decay of 239 into neptunium-239 and repeat once more to get p-239
No thanks
You can try training with it
It will probably work
btw idk which rvc i have
"Abelson placed pressurized liquid uranium hexafluoride into the space between two concentric vertical pipes. With the outer wall cooled by a circulating water jacket and the inner wall heated by high-pressure steam, the lighter U-235 isotope tended to concentrate near the hot wall and the heavier U-238 near the cold" - this is the steps you need to take to get u-235 if you want a weaker bomb
all that matters is the fusion bomb though - if you can make some go critical then the process can create enough energy to make more fusion fuel go critical
I'm an AMD user, there are some files that are damaged, I'm trying that
hey
so it is like the match-stick which makes gunpowder go off - you can have a much larger amount of gunpowder
you need a material which generally holds onto neutrons - which is boron
it was used in the tsar bomba around the fission bomb
someone know why the go-applio.bat says "no supported nvidia gpu found"?
Ayo? @eternal bronze level 2 !!! 
the tnt around said fission bomb is in a spherical shape - encapsulating it and there must be some trigger which sets it off
nagasaki had an off-timed clock and so the bomb went off above the ground
better learn from history and make the bomb go off on contact with the floor
what epochs and batch size shall i put for 1min dataset?
@storm meadow guess?
wasnāt talking to you but ok
also sorry for the ping š
Also you canāt train locally on an AMD GPU
lithium deutride is made from extracting large amounts of d2o and converting it into the lithium dueteride
Ayo? @hoary barn level 2 !!! 
Just do 500 and see what plops out
batch size?
the bombs are held together in a semi-solid structure by a kind of foam
i have not even gone in depth on the important things
but everything is online for some reason
there is a fissionable spark plug inside the fusion bomb
it looks like a large toilet paper or paper towel on a stand
you are very welcome
: )
hi
Hi guys, just saw this on YouTube! If anyone can point me to a Taylor Swift song generator that'd be great !
where can I ask questions about A.I stuff?
depending on the topic we have multiple help channels
Where can I request somebody to make a model? I was looking for a specific voice model but could not find it
https://discord.com/channels/1159260121998827560/1175430844685484042
/search https://discord.com/channels/1159260121998827560/1163592055830880266
or take a look at weights.gg
https://discord.com/channels/1159260121998827560/1191429836321849435
cant find the model you want?
you can commission one from a Model Master š https://discord.com/channels/1159260121998827560/1159289738314919936
@remote basalt āļø
Thank you
Ayo? @remote basalt level 1 !!! 
how to fix the goodle collab constantly disconnecting due to inactivity
what language did you use for code the model?
hi
Hi
im using rvc pre existing model to clone voice but the collab disconnect every few minutes due to inactivity
Ayo? @patent dagger level 1 !!! 
hi
Perfect
hello
;-;
Owo
when i was a child i used to try and search for people who i would enjoy meeting and knowing
and i succeeded
soul_spur it was your mother
i
am your father
@tardy vector how do i get the 'verified' chat
You have to get level 5 on arcane
Vote Booster: Vote now for a 10% boost. https://arcane.bot/vote
⨠Tip: Use /card to modify your rank card
oh my god
72
i think that would be about 260k of the xp thing to get that
that is if it increases linearly like i think in the same way or whatever
okay so 14xp per message
okay nvm
22,802 messages
i underestimated
m
319228
go to sleep
no
bed
nuh uh
Ayo? @hoary barn level 3 !!! 
i see - it seems that the longer the message the more xp or some crap
I hope you realize there is a cooldown
i was wrong
: 0 this is news to me
i totally did not realize
oh you were talking about updates for /level lol
unless there is proportion between being an asshole and level
limiiiiiiiiiiiiiiited lifespan

well as you can see i am heavily intoxicated
@hoary barn Stop spamming
Buy google collab pro
I linked my ai model zip on google drive that had a pth file and it was public,but it keeps telling me that it had/was neither of those? What was the issue in here?
i have the same issue, pls help
question, why does this use so so much cpu, can i make it use gpu instead?
it's using 30-40%
aint the gpu a video card?
we're all built different
wait tho, what abt stuff like RTX voice?
its fine
funky words too complicated for my brain
(idfk lmao)
it uses gpu to do noise cancellation iirc
Gdown has some problems now use hugginface
so, no way to lessen cpu load?
dm if soemone got the weekend
no
Ayo? @eternal egret level 1 !!! 
sobbing, altho question, would a ryzen cpu work better than intel? currently on an i5 8600k
i mean not rlly it depends on the performance of it
Decrease "EXTRA"
Put the index slider to 0
It's voice-changer
then is there a way to turn client off?
Ayo? @brittle nova level 2 !!! 
What do you mean by that ?
also i'm assuming the mmvc thing is showing up instead of command prompt that showed there last time
close the client but still keep the thing going? client is using like 13% cpu
MMVCServerSIO just runs in command prompt yeah.
Also opens up voice-changer-native-client which is what you don't want to see ?
Try setting the audio mode to "server"
Could help regarding that
lol
Unfortunately you can't make it not launch the native UI thing
Unless you tweak the code manually
aa, yeah not built for that
or deleting the voice-changer-native-client file could work if I remember right
But then you'd have to use a web browser to make the UI appear and start it
too big brain
but aaa this is already taking so much i can barely play games with obs and discord on, might die if i stream lmao
could OC tho
also server not working, do i need to do smthn besides pick my input output stuff?
And not get the performance hit
It's unstable lol ĀÆ_(ć)_/ĀÆ
wait really? would it work if i use a linux-ed switch?
HI
nintendo switch with linux yes
Too slow for it unfortunately
sadge, so phones won't do either?
Nope !
hmmmmm
i may get a uni laptop soon, could do this on it while streaming, how would i do it so i get the audio onto this pc and not the lappy
you guys know where I can find the archives for the voice models?
It might be able to, latency will vary depending on the specs, but that would work
how much extra latency we looking at? š¬
offloading is pretty much as simple as having both PCs on the same network
launching voice-changer via start_https on the laptop
and connecting to it via https://laptop-ip:18888/ on the main PC's web browser
leaking answers for model master application!!!
Whatever the laptop can do without choking + whatever small delay for the network transfer which might not even be significant
so abt 100-200ms extra? or like, same type of latency as wireless PCVR
Can't tell, depends if the laptop is fast enough to deliver those speeds
also what's S.threshold?
The laptop I'm using right now can run voice-changer with 170ms of delay on it's own
When connecting from the desktop, I can run the same "CHUNK" setting of 170ms without issues, even with the small extra latency from the network
Minimum volume that it processes
64
isn't that super cut out-y?
yeah i'm trying it out rn it sounds like i'm talking through 4 fans
.
It only works if your computer manages to process it fast enough
Ayo? @waxen grove level 62 !!! 
Anything lover than 64 just cuts out here as well
so, the faster ur cpu, the less chunks u need?
Ayo? @brittle nova level 3 !!! 
GPU
2070 goes brr
Wait there's a "GPU" setting, what is that set on ?
gpu 0
that's integrated gpu isn't it?
how many epochs should i put for 21 minute training
OH FOR FUCKS SAKES
There's a diff one for nvidia which is significantly faster
help channel moment
it not letting me switch now
nvm just had to spam like a good user
https://huggingface.co/wok000/vcclient000/blob/main/MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.17b.zip
That's the nvidia one, the GPU selection looks a bit different
there
so do i delete this one i currently have?
too late i already did
Hey guys, is quality of RVC Voice Changer dependent on Voice Models or Pre-trained Base Model?
There were some default Japanese voice models that came with RVC WebUI, they perform much better than any of the voice models I find online i.e. trained by community.
still downloading
raaaaa
the zip downloaded but script download taking forever
WAIT IT WORKS IN BROWSER
Oh that's funny
the files couldn't be extracted
huh?
-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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- 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
any got any good girl ones
Searching for the perfect RVC e-girl voice model? 
not that
can you guys recommend me some girly male voices? doesn't matter where from
any1 got a non rickroll one
i downloaded the original rvi but i couldn't get the pipps
...?
You'll never get me

https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/releases how does this work?
the pipps ?
On windows it's just a matter of downloading the correct release for your PC (depending on the GPU)
it makes a voice model using audio files that you give it
Ayo? @molten stone level 55 !!! 
i have lenovo
Ayo? @lusty thistle level 2 !!! 
Can anyone send me a specific link for a faster voice generator of spy from team fortress please
what u mean?
There are two files depending on what your computer contains
Wait do you have the software running right now ?
hi
no
-enable
I have a problem..
I can't get back to working where I was and I don't understand why, yet everything was imported, and I even looked there is epoch number 370 in the logs folders but when I press on train it starts again at 0...
Help me please..
no
Can you open task manager ?
prolly screwed up on saving the d/g files or you're using a different batch size
I checked everything and I still don't understand why, I saved everything correctly, I checked on my drive with each backup if I had my 370 epoch, and I had them, so if you can help me the more I will be delighted š
i can't even share the screenshot
just make sure you're using the same batch size as the last time
heyo
if it is the same, then you probably stopped while it was saving the epochs
It's done I've already looked don't worry
so yeah it most likely screwed up the d/g files
Check out #1192011222023950368 and create a post there
now u gotta start from the beginning
You should have permissions there
I started the save and waited for it to finish
It's been at least 4 times that I try again at 0
just remember to never stop at a saving epoch, always stop 1 epoch after
so it does it very often for me..
that's besides the point
Processing | 90.9 s i just wondering how long does it take to train in RVC it's 90.9s anyone experienced before ?
then ur screwing up most likely
rvc training loading for me never screwed up unless i did
yes ik
xD ok
This takes a while depending on what you're using, just wait for it
Wait, are we talking about the colab version?
How long do you expect?
depends on the audio length ig
I guess we are?
Yes yes, I prefer to be sure
OK, I'll try again, thanks for your help.
Thank you
you're welcome! wish i could help more tho but it happens yk
you're welcome :)
At the same time it's my fault xD, I just wanted to test if the save really worked or not, when I could have finished my model peacefully
Not problem
And just on the tutorial on google docs, it talks about step for the dashboard, but I have epoch on the abscissa bar @forest quarry
Ayo? @hardy rapids level 3 !!! 
I don't know if that changes anything or not?
Xd ads
the what? I don't quite understand
yaaaaaaaa
On TensorBoard srry
@vale swan just go on #1159290752195633273
...I don't think I get it though
prolly a language barrier
Don't you see? the TensorBoard on the colab, to see overtraining
This still doesn't explain the "step on the dashboard" and "abcissa" thing you just cited
So yeaj
It doesn't matter
Just in the tutorial we see that the person does a calculation to find the epochs with the steps on the TensorBoard, but I therefore have the epochs directly and not the "steps"
I would like to show you but I don't have the right to send images here
MJ already told him
yeye
hello, I've just started to use AI for fun š
ello there, welcome to the server, nice to know ;)
nice idea
Senior RVC what is batch size per GPU my gpu is 3050 8gb what size should i chose
nice
anyone know the jschlatt model?
u can try 4 and 6
Thank you again you're amazing senior
you're welcome ;)
can someone help me how to use voice ai and how to put models in the program
can someone tell how to you rvc ai to change voice?
it says the file is unsupported
hello
hi
š¤
Anime women arent real
We don't give support to voice.ai specifically. You can use RVC tho
hi
š
@waxen grove can u help a bit, i'd like to run using client app, not a web page, but it not letting ,e
just use start_http bat and not start_https
i cant hear myself in voicechanger, why ?
start_http.bat
start_https.bat
hi
add ur headphones to monitor
tysm
ty mj and squishy!
you're welcome :)
hii
hi guys
ello
i need help
post on #1192011222023950368! make sure to add details and screenshots so it can make our life easier ;)
ty ^^
you're welcome!!
How can I boost instagram followers?
hi guys
This is a server about AI btw
We don't know about dat
I recommend uhh sharing your @ to friends and family
:3
People will know
Sigh
It's not for me
We ain't helping with botted followers. That doesn't even help your Instagram grow...
Sigh... I am not using it on my content
Regardless - it doesn't help.
Can just tell me the software pls
what's pasthrough do for this thing
Nah just tell them to go to #1159290752195633273
just passes your real mic output through the cable
and not the AI voice
Seemed to be a copy paste or a bot to me
Oh so it uses your real voice?
If they are I will kick
if you're on passthru, it passes your real voice through, so yeah
if you disable it it will only play the AI output
The AI voice on my end is always really choppy
bro i have a problem. after using it for first time and i got error, the site gave me error to connect
hey guys how do i make sure that my mic picks up my AI voice changer instead of my regular voice? w-okada btw
like it said something to go on waybackmachine
What GPU do you have?
What program were you using
can anyone help me tune the sound
gradio
You need a virtual audio cable. Check https://rentry.co/VoiceChangerGuide
This guide is back to receiving regular updates, however do not ping me for anything on it. The guide itself speaks for me. - Blanc
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
Frequent...
change your discord input to vb audio output
rvc
it's sounding pretty meh compared to yesterday
already done
...This isn't necessarily helpful
What fork of RVC
it says "No interface is running right now"
then change your okada vc output to vb audio input, click start, not passthru
Are you on Colab
done and still regular voice
yes
Ayo? @elder willow level 1 !!! 
I just got it nevermind
@nimble finch vc?
Probably an old Colab then
sure
- 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- 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
join my vc channel
Couldn't find your model files. Check the found file results above. (Did you run Index Training?)
I have that, but i have run index training...
oh man check your Drive folder
yeah and ?
I'm still waiting the approbation of my RVC model.
Ayo? @vital fog level 2 !!! 
do ya see the D, G files
and the index
inside your Model's folder
Autodetecting epoch count... Model with path /content/Mangio-RVC-Fork/weights/LivaĆÆ found automatically. Searching for weights files in: /content/Mangio-RVC-Fork/weights/LivaĆÆ.pth Searching for logs file in: /content/Mangio-RVC-Fork/logs/LivaĆÆ/added_*_1_LivaĆÆ_v2.index Found weights files: ['/content/Mangio-RVC-Fork/weights/LivaĆÆ.pth'] Found logs files: []
oh wait srry not that..
you may have just saved the weights and not the logs
No features exist for this model yet. Did you run Feature Extraction? that is my problem
probably no-feature-todo thing again, this is probably your dataset being funky
Hey
I have a question: Now the AI Hub server reject AI models of real-life singers for copyright reasons?. Most of the AI models on this server are of fictional characters.
I hope not I just trained a French singer xD
yeah ur dataset format might be wrong
no
It's rejected if it has bad quality.
The dataset must be formatted like this;
- folder with your model's name (must be the same as experiment_name you will set on the Colab, no spaces too)
- Audios here (no spaces recommended here too)
YES
That's the problem
You should do .zip
bahaha xD
.rar compresses the files and RVC can't decode it properly somehow
ok i'm srry,and does this have anything to do with the problem earlier?
And why then does it not show me any errors? he did the training for 30min in a vacuum xD
Ayo? @hardy rapids level 4 !!! 
When I didn't want to recover the backup to resume at 370epoch
AI Hub being on discord is pretty smart ngl
Since they cant really be striked since no one will suspect a thing since it isnt a website.
Well they will find us all eventually
come back to vc @trail coral
The shameless Homer Simpson rip-off now has his own AI voice model: https://app.kits.ai//convert/shared/gela-samsonadze
at one point this worked with 48 chunk but now it's not at all
Anime women arent real
lies
Ayo? @brittle nova level 4 !!! 
am one
sup people
can someone help me
sa
selam türk müsün
he
yardım edebilir misin
buyur
what transcate do?
ƶzele gelsen
Can someone give me the
best colab link?
And The models are not working for me in huggingface
u are a skull emoji @crisp verge
?
idk
Could u recomend good egirl voice? i am tired of searching, there is nothing really good?
Searching for the perfect RVC e-girl voice model? 
Ayo? @fair frigate level 1 !!! 
Astolfo
bro))
where i can see number of steps for my models ?
Wait what are you running on again? Is it two different machines or one only?
dw got it, but now struggling with tuning
one machine btw
wassup guys
ć ¤
Inst https://drive.google.com/drive/folders/1ld-orYQ_uXjVr9s3X7e4nRkZZfMJwS5f
ćÆćꬔćå§ćć¾ććć
čŖćććć£ććéć³ę¹
飽ććć«ćå”ćć¾ććć
éćčć貓ę¹ć®å³ę¹
ēććēććå
é«ććé«ćć空ć®äø
赤ćć赤ććēćæ
ęę„ćÆćć¾ć ććä¼ćæ
čćććļ¼
Witch Hunt
åÆć話ćć¦ć»ććć®
Uh-uh-oh-oh, Uh-uh-oh
é ćć
ćć®ć
Witch Hunt
ēćčæćć¦ć»ććć®
Uh-uh-oh-oh, Uh-uh-oh
ä»ćć
éäøć®ēę§ćęęćć©ćć
ē°ćęććć£ćććę¤ę¹ć¾ć§
éäøć®ēę§ćęęćć©ćć
ęć欲ććć£ćć ććęćć®
ćć®ę„ć®ē“ęćÆć©ćć«...
is there an EN teto voice?
u got me lol
ćéé³ććććć©ćÆć«ćć„ć¼ćøć«ć« ā Labyrinth by Miracle Musical Cover Ft. Kasane Teto
SVP: https://youtube.com/@mr.fiffles
https://www.youtube.com/watch?v=MdAzl3sOwmY
hi
hi
ā ļø
hi
Hi?
i think i know what was the problem with my old rvc thing
EN teto sounds so good š®
ā ļøI was invited by a friend
whats a good free alternative to create free ai models since kits ai is now payed
Colab was always the best alternative
-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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- 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
RVC disconnected :)
escuchĆ© un cover del utau Pruno Faustino de esta misma canción y recordĆ© que yo habĆa hecho uno con Teto hace algĆŗn tiempo y decidĆ hacerle remake
MIDI/SVP
https://drive.google.com/drive/folders/1-vAxhw5Yvi4MjmEdF4wjW-GvtBGOx9Ae?usp=share_link
ā¤ļø Discord: Zusmo-cat-0 Utau#3424
-rvc
š All-In-One English documentation
ā English FAQ
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
Model training with Mainline RVC
Link: Rentry
credits: Raven (ravencutie21)
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How To Make an AI Cover With Ilaria RVC
Link: Rentry
Credits: š½ Julia (ailen2091)
Here's a guide for it
so like i used that rvc-gui thing last time on my old PC i remember and i think my old PC had like an nvidia gpu thing and now my new computer that i'm using has an amd thing and maybe that's why???
Ayo? @hollow bison level 2 !!! 
oooooh yeah maybe that
hi
oh my god finally it works
not the old rvc gui thing that i used but mangio
finally
YAYYYY
YAY ^_^
@quartz roost https://www.youtube.com/watch?v=ZEYCjcxgZnY š¤Æ
hmmm where's the embedding
bruh
something is wrong with discord?
it wont embed even if your verifed in general chat 
aw man
only trusted member role and staff
is it okay to send links if the embeddings won't appear?
if its promo then #1159290752195633273
okay, i just wanted to share a cover with tetski
hi
hai
Hello Deleted User 61265a4e
yo
hellow noobmaster692722
Hi
hey everyone heads up be careful on youtube
I had an ai music video with 80k+ views and an antipiracy company copyright striked me MANUALLY
so I have to get rid of all my ai songs š¦
Hello
Yeah i have a few taylor swift songs ive had teto sing, she slays it
Make ai original songs and profit
yes ill use ai to make songs somehow
I already have over 750 subscribers and stupid "Riaa Antipiracy" was like NOT TODAY
Vocaloids have ai and are top quality and are a multimillion dollar industry
Rvc is amazing im working on using it with an automatic translator
someone help #1192011222023950368
when update when rvc3
Why? It does what it needs to quite well
ok
but what it do
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
Hey Anyone from India ?
silence is death
how can i use a model in python
What is the issue you are having
like i want to use no gui
@quartz roost Can you Tell me How Can i Make A Pth And A index file of any model ?
Did you read #1159513888199540817 thoroughly?
hi guys, I have a question, does the "Easy GUI (for RVC v2, with crepe) (with improved downloader)" is working? Because on the second step I get this error with every model:
Access denied with the following error:
Cannot retrieve the public link of the file. You may need to change
the permission to 'Anyone with the link', or have had many accesses.
You may still be able to access the file from the browser:
https://drive.google.com/uc?id=1Ny2aZ5xea1mIS92EPyfuDpcl4a5niaZN
Error: Could not find a valid .pth file within the extracted zip.
If there's an error above this talking about 'Access denied', try one of the Alt URLs in the Google Sheets for this model.
Thanks in advance for your help.
No, it is not working. And Google Drive links are broken for all Colabs, I recommend downloading those links and uploading them to Huggingface
So sorry
Ayo? @unborn lark level 1 !!! 
-huggingface
English guide
Link: https://rentry.org/fdg_guide_newer
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
- 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- 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
Thanks fellow Luffy fan!
hey guys, am i able to train a model on plain speech and use it for singing? or do I need to train a model on singing vocals if I want to do that?

yes, but you might need to change the pitch of the vocals while converting em
or maybe not
i got a steve jobs model that can sing lol š
Can anyone send me a specific link for faster voice converting for team fortress spy ai voice please
lmao
so is it just a matter of trial and error with the training settings?
nah, its more of like
the options while converting vocals
and ur dataset maybe
hi
I don't believe there is a specific link for that, the Huggingface spaces shouldn't take that long depending on the audio you use, if it's a song it'll take a while yeah
change the pitch of the input vocals?
Itās more of a text to speech is wut am looking for
hey yall use crepe or rmvpe
yeah, u can do it in the rvc gui itself

rvmpe for life
wanna listen a truth?
i think if u get ur tts from ElevenLabs and convert in Huggingface it will be decent
what's the rule of thumb for training data? i saw somewhere my dataset should be 80-100 ten second audio clips
noooo, you don't need to trim the dataset like that
I prefer mangio-crepe but since it's more prone to artifacting I'm obligated to use rmvpe
You can just use one big audio file
With modern versions of RVC you don't have to trim the dataset into various files
how to upload photos in this ch. sir?
got it, thanks for all your help guys
Ayo? @full night level 1 !!! 
You need to be a trusted user
(trusted role)
^^
you're welcome :) glad we could help
is there some kind of level required?
But you can upload in pics in #1192011222023950368 if that's what you need
Not really. You just need to be trusted by the higher-ups
By being active, helping, improving the server
Stuff like that
rmvpe models builds up white noise on inference at most of the times
And isnt stable on pitch on inferences
while crepe doesnt have that problem tested it with my friend
I don't make models so idk
crepe is hyper sensible to any type of noise
But in my experience rmvpe is not the best at hitting every note
so you have to have the rx10 knowledge, else it'll sound bad
Not hyper sensible, wdym š
- i think Charlotte told this, but it is really good at deep tones
my dataset had to worst parts still hands down the best till i made
kinda like harvest v2

I have only done inference with it so I can't say much
exactly.

training with Mangio-crepe is a hit or miss if you don't clean your dataset
dio>>>>
š„±
Dio
His eyes look like olives
Why tf are my cook instincts kicking in
what's the difference between the f0 det things
Pitch extraction, ones are faster and better (rvmpe) and some are slower and okay i guess (crepe)
what's the difference between rvmpe and rvmpe onnx?
anyone can recommend a girl/woman voice model
Searching for the perfect RVC e-girl voice model? 
ty ig
š
yo how do i convert .index to .json?
ćéé³ććććć©ćÆć«ćć„ć¼ćøć«ć« ā Labyrinth by Miracle Musical Cover Ft. Kasane Teto
SVP: https://youtube.com/@mr.fiffles
https://www.youtube.com/watch?v=MdAzl3sOwmY
there's no teto model in #1175430844685484042 š
And the answer is always teto
She has her own website for the model
that's not an rvc model 
actually using smthn else personally
off vocal https://commons.nicovideo.jp/works/nc323822
music / lyrics: OSTER projectćhttps://twitter.com/fuwacina
guitar: ć·ć³ćć¤ć³ć¦ćøćhttps://twitter.com/koji_onsensuki
illustration: ćåćhttps://twitter.com/fucodoku
movie programming: ć³ć¼ćććhttps://twitter.com/_Be_suke_
Web: http://fuwafuwacinnamon.sakura.ne.jp/
FANBOXćhttps://www.pixiv.net/fanbox/cr...
onnx runs on DirectML, this is recommended for AMD GPUs n sruff
ah, so for my 2070, just normal rmvpe? also it lags on like 96 chunks with least extra is that normal?
how much dataset is needed if you are aiming to train a highest quality model?
dataset quality matter more but id recommend like 10-20 min
this guide recommends trying these two settings for nvidia 20xx series: 112 chunk +16384 extra or 128 chunk +16384 extra if the former is still too laggy
so going to 1 hour won't have any significant improvement?
not really tbh
yeah just normal rvmpe
and yeah, i think only the most recent cards can do 96 chunk
yeah no
š
Ayo? @finite aspen level 5 !!! 
verified chat 

hmmm
how i can have RVC for amd card pls ?
tysm this guide is neat af!
I'm training my new ai voice model 25 minutes sample with 500 epochs it took 6 hour just finished 360 epochs and my pc almost burn š„ššŖ¦
Ayo? @vague skiff level 2 !!! 
how do i uninstall VB btw?
is w-okada the best voice changer out there?
Searching for the perfect RVC e-girl voice model? 
lol the bot
applio didn't work for amd graphique card ?
oh my god....
thats terrifying
š
/search

