#🧬│ai-chat
1 messages · Page 246 of 1
Hello
LMFAOOO
wheres ur nitro then
Hello
guys help
Ayo? @eager moth level 3 !!! 
theres like a 1 second delay
sematary
Just a joke when’s Bondrewd ai cover ?
Idk, because i need to make model of bondrewd first
I have no time for now to do that model
the voice ai doesnt change my voice? any fix
hi
yo
Hey there
can someone help
Ayo? @rapid storm level 1 !!! 
pls ask on #🔍│help-w-okada
mb
Hello
hai
I have a picture that I want to turn into an animated gif for a wallpaper so how I do it?
Ayo? @gaunt bough level 1 !!! 
🌹
How can i do to search a voice model? idk the comands
it says search at the top
Ayo? @unkempt marsh level 1 !!! 
Just search on #1175430844685484042 or use slash commands on #🔍│find-models
Which program are you using
Could anyone recommend me a Pretty decent mic? Nothing too too expensive. My Okada, despite every setting I change, and every voice I use struggles to form some words, And I believe its just cause my mic is shitty. Because sometimes it'll get it perfect, then mess up next go around
I must continue to yell at people with Shaxx's Voice
Go for a Redragon headset or something
I got a Redragon Hero, has a pretty decent mic and it worked well for W-Okada in my experience
But idk, maybe that could be the model or something
I wouldn't really recommend spending money if you're not 100% sure on what the issue is
Hi
Ello
wow
how to use these voice models live?
𝒽𝑒𝓁𝓁𝑜 𝓂𝓎 𝓁𝒾𝓉𝓉𝓁𝑒 𝓈𝓀𝒾𝒷𝒾𝒹𝒾 𝓉𝑜𝒾𝓁𝑒𝓉𝓈 /j
most decent headset mics work, and yeah go for the affordable one
(though Hyperx cloud 2 is pretty solid)
EOL - No further Updates
Github - Blanc-dot
Discord - Blanc_dot
Despite being end of life, most if not all information has not really changed, so should be very accurate until actual new stuff comes out.
Other Links
Antasma's Local Error Fixes
Antasma's Colab guide
Sushi's useful Links - You need...
^
Use W-Okada
Thats actually the one I have lmao
I don't know if its because its worn out, or what
If I say Sword, it'll say Swarsh, but SOMETIMES it'll say Sword
But thats literally any voice model I try too, not just one
where do i download it
A good tip, increase chunks, use higher extra, and keep the mic close to your mouth, decrease gain in if you have to
download what
Ill keep that in mind
New Juice wrld cover by @abstract tundra is 🔥
https://www.huggingbass.com/song/66478f588f45770dc552775f
Ayo? @signal wave level 1 !!! 
-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
- 🆕 UVR5 NO UI for Google Colab, by Eddy 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
Mind posting on #1159290752195633273? Ty
thx m8
you're welcome :)
hi!
ello
hi!
hi thereee
hi
hi
Tf why weebs calling anime incest hot
💀
We are not making it to the familly diner with this one 🗣️🔥
How is everyone
I mean... 😏
What you expect from our degenerate hentai enjoying asses?
bruh
Art by JaheTerban
Awkward
No embeds? I was about to report a scam/spam/bot/whatever message coming from someone on this server
Which isn't the "hey bro I'm too broke to buy doshirak pls commission me" kind
try /report
Zdrastfujce
Hi, This is Michael based on US.
I have 9+ years of experience in software development.
hi
hi michael
Nice to meet you, everyone.
Ayo? @kindred pond level 1 !!! 
hi
Tes
Juri
Hiii i am searching for an ai voice model for a video game.
does anyone know any tts sites that take an audio file and use that voice? much love
There's licensed ai voices with no legal repercussions btw
There's some free ones
One in particular is making BANK
My friend is making thousands with a free voice
True and there's TTS ones made by companies that are technically free after those many years for OS
You can use weights.gg to make TTS and it would convert with RVC
yea i just need the text to speech part and i can get the voice from this discord to make what i want
alternatively you could also use the Ilaria RVC Colab
Last update: Mar 8, 2024
^^
Though I feel discouraged from doing Ai covers at all so I'm just gonna take a break from doing that. Mainstream music copyright is just utter bs
i am a big noob in terms of ai so let me try one at a time
Ayo? @covert hawk level 1 !!! 
No restrictions
It's alrighty
Lmk if you have any issues
Again I can't afford to buy vocaloid music software
Good thing she is free and open source and has been in the top 100 multiple times this year
OMG i dont know notrhing when i clicked this 😭 let me try the weights.gg site
But I also don't feel motivated to make music at all. I just need a break.
like do u guys know those drake disses that people make that are clearly fake i want to make one of those
If you have a good CPU you can use RVC GUI but modify it to include RVMPE
no
please no
😭
It's open source tbf

Not everyone can afford really good GPUs
Like my computers
Colabs are just limited especially if it has the gui thing semi banned
That was the way I can do it without doing it online
It can work for CPUs by around 8GB of RAM
It's slow but it works
No GPU needed
Vocoloflex is a realtime voicechanger on cpu
But it needs a update to include the newer features
Colab is limited bro
locally
Ilaria rvc model sounds nice too
mainline RVC 1006 and Applio do the same thing, but newer and better
you do realize, there's ilaria rvc locally
I don't have a good GPU
And has limited legal use too
I'm not sure if this is the best place or the help-rvc sub, but I'm having issues with a model sounding too robotic(?). I think its due to some artifacts I had in my dataset. I know there are paying gigs to create a model from a dataset, but are there paying gigs to help clean up audio and make a good dataset?
My GPU is Nvidia GT 1030
inference can even run on average laptop cpu, but training only does on Nvidia gpu with 6+ gb vram
Sadly I don't think there are any gigs for cleaning datasets... but there should've been
@signal cairn dm spam scammer -> @elder willow can show proof
Locally
For a good model you will need a good mic and a sound treated small room like a foam packed closet with no reverb
When I used one of those RVC software it did nothing only the realtime RVC worked on my computers
A wide variety of singing, 10 to 20 minutes
send on my dms
Ilaria RVC should work though
https://huggingface.co/TheStinger/IlariaRVCBeta/resolve/main/Ilaria RVC Mainline GOLDEN BETA.zip extract this, open it with go-web.bat or whatever, and it should work
VRAM is 2GB has anyone tested that GPU
😭 Regardless of the GPU you have rn, it should work
Even if it uses your CPU to do so
Wow you guys never tested in that old GPU at all
Does it even have CPU supported at all?
All RVC forks have CPU support
But they never worked
We don't have all the GPUs in the world tho
ugh
Ayo? @plush barn level 19 !!! 
Minimum inference according the wiki its 4GB-6GB of VRAM which only my friend can run. I can only use CPU for the inference. But I don't know where to turn that shit on
For training it needs NVIDIA GPUs from the RTX series (8GB of VRAM and up). But I'm not concern about that at all because most of the models I train only needed a few hours to train anyway in colab. I only want to use inference locally
But I need to know were to enable CPU.
I have a RVC thing that has RVC realtime I just need to find the way to enable the CPU for inference
🖐️
Automatically does it, probably
wait
😭
.u
✅ Cached playcounts have been updated for BabyShreddy based on 24 new scrobbles.
.wk
👑 sred - 503644 plays
2. babyshreddy - 235508 plays
3. rxzvxn - 138579 plays
4. - 2129 plays
5. Floare de nu-mă-uita - 1830 plays
6. shenko - 1432 plays
7. kiko - 628 plays
8. Niko - 62 plays
9. kelvin - 48 plays
10. sev - 43 plays
11. 🌷 vivi !! - 29 plays
12. megiu - 12 plays
13. ⁷ - 7 plays
14. TheMihaiノ𝐈𝐂 - 2 plays
Crown stolen by sred with 503644 plays!
Previous owner: babyshreddy with 235508 plays.
😂
4 months of listening to scenecore got me to 3k
.wkg scenecore
1. Kainé - 111541 plays
2. vici - 39360 plays
3. Übel - 10673 plays
4. maxi - 10643 plays
5. ⁷ - 8585 plays
6. exan - 7384 plays
7. chikin - 4538 plays
8. peri - 3953 plays
9. cud - 3802 plays
10. Pout - 3552 plays
11. yucc - 3450 plays
12. Vyzer - 3391 plays
13. mel - 3204 plays
14. mionek - 3181 plays
15. - 2982 plays
Who would be interested in a PyTorch learning group?
.wkg stevie wonder
Do you even understand how all of this worked because it sounds like you don't. Like no offense but it feels like your pushing me to get it rather than saying that it works without thinking of testing that damn thing.
.gwk stevie wonder
1. Private user - 43070 plays
2. missingfigure - 13645 plays
3. uzacc774 - 12800 plays
4. Salrt - 11692 plays
5. imBoltic - 6701 plays
6. Private user - 6465 plays
7. samglogan - 5941 plays
8. altostar482 - 5016 plays
9. Corehero_ - 5008 plays
10. Cready93 - 4576 plays
11. Private user - 4009 plays
12. Private user - 3987 plays
13. Candide1010 - 3970 plays
14. Private user - 3959 plays
25. not MJ - 3125 plays
Mhhh kinda. I'm not all too sure - but you can always try it out. It should work either way, that's my point though
Yeah then you don't need to say anything else other than that
If that's the only thing you know
Sorry btw
.
.
where the polish people at
where and how can i use models i downloded some but idk what i have to do with it
https://docs.aihub.wtf/ has everything you need to know about RVC, if you have a decent GPU you can use it locally to convert vocals n stuff
Last update: Mar 10, 2024
thankss
beans
coconut
есть русские?
can anyone help, all my voices come with an accent (personally, I’m Russian), how can I fix this?
Ayo? @errant edge level 1 !!! 
Try increasing Index, but it'll inevitably have some accent 
Depending how you speak ofc
there are voices of Russian characters, but they still have an accent ._.
You can increase index if you want more of that model's accent, that's what I'm saying
Can anyone make a voice model for the Sphinx from Dragon's Dogma 2?
You can request in #1159289738314919936
k ty for help
Ty
Ayo? @elder willow level 1 !!! 
Hope it helps ya :)
can help pls
Ayo? @elder willow level 1 !!! 
yooooo
Donc you know a free colab to make ia song instru and vocals (whit my own lyrics) like Suno?
i finally figured it out
i cant use ai
Ayo? @hot steeple level 1 !!! 
Hi, I'm new here. Happy to be part of this!
am i the only one who joined for the emoijs?
hello
if u got issues, ask them in help channels and be specific
ai hub
what? what google spreadsheet cus we had one but its like old asf from last summer
i mean they are fire

if you mean to make an rvc voice model of that youtuber, check our docs https://docs.aihub.wtf
Last update: Mar 10, 2024
yw, if u need help ask in help channels
nahh dwwww
Ayo abuelos.
Guys.
Do you also get lost in discord's indicated messages & selected channels?
I give up on making ai covers. Every single time i try to make one the inferencing is terrible no matter what i do. I did the correct process for MVSEP and used the 2024 isolation model and still it makes garbage results when inferencing the voice
Never had an issue with this one

you can sing the song yourself if you know how to sing
Vocaloid
hello, i wrote a piece on sheet music with pencil. is there any software i can use that will be able to scan it and then play it for me?
i cant sing well
At least u have a great sense of profile aesthetic and choice of color
you can also ask or commission someone to sing it for you
or do i have to just manually put it into a software myself

You can import vocals and edit
are you making ai sing from a voice model or are you inferencing a voice model over a voice isolated song?
Hello boys and girls
Ayo? @stuck frigate level 11 !!! 
Hello
no no, it’s like sheet music with SATB, there’s no vocal recording or anything just the sheet. can i take a picture of it or something and it can be turned digital
I came from work
Convert it to midi, load it in and input lyrics
Goodnight
You can make it sing anything. With a variety of inputs
i dont understand. How are you doing this?
I just went to have a meal after work with a colleague
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
So I came late, but it is totally fine
Welp, everyone
Goodnighty night 🌃😁
It was a beautiful day 😌
ok but how are you getting those raw lyrics? i cant record any myself because my mic isnt good for singing lyrics. are you using ace studio as a base?
wait that is ace studio isnt it?
wdym with raided? its just outdated anymore lol
oh ok
i dont understand how do people get away with having ai covers on you tube and they stay on you tube for quite a long time yet when i make one my first version is copyright striked and removed but my newer version with only slightly clearer vocals doesnt get touched? (i privated it just in case)
No, we don't use ace studio because the developers are controversial.
You can write your own lyrics! These are legal to use
what i mean is are you singing the lyrics yourself or are you using a prgraom and if youre using a program to make it sing what program are you using?
You can do both
This is the best selling ai singing voice available
yes i know you can do both but i cant sing so i cant do that, i have to use a program
what program are you using
There's english available on the site
. . .
what are you talking about? i said what program are you using? i have my own voice models already
SynthesizerV
There's also an open source version called openutau diffsinger Ai that you can use your own voice models in
i cant do it then. i know nothing about modifying open source stuff
The commercial version is already pre made for you. Legal and can sell your music
any beginner tutorials?
Yes, there's a whole series. If you can type and speak you can use it
Just bought another album from someone else made with it yesterday
Hello
I use rvc for making new sounds, it can be used for more than just vocals
where is this supposed english open source?
Rvc?
Rvc and openutau diffsinger you can make your own voices free
is the program free or not
Rvc and openutau diffsinger you can make your own voices FREE
omg... not making the voices but the program itself
The program is already made. You use the same singing dataset you already have.
Hiii…I am new
is the PROGRAM free not making the voices
link to the open source file? i cant read japanese
Both traditional and ai methods
Depends on the artist and song sometimes
Because record labels
Vocaloid songs they ENCOURAGE you to cover their songs and even post the stems so you can remix it for epic dj raves and parties. They are going viral
sounds like an excuse to me. There are tons of juice wrld ai covers with thousands of views yet my ai cover with less than 100 views gets copyrighted despite you tube saying the creator allows it to be used on you tube
skill issue I guess
you know maybe it is. Maybe mine are so much better than anyone elses that they felt like they needed to take it down

Or the complete opposite 👀
“Yo these ai covers are so trash, we can’t be ruining our artists reputation and fanbase like that, take it down”
thats a load of bs. i have the same exact cover on my channel with the same exact song but slightly better vocal inferencing and that one isnt copyright striked
Hey guys
This is literally going to sound like a Karen but can I speak to a manager? @signal cairn
I want to extend an opportunity
with what 😭
how can use this server
It's a server wide thing, it's like a mutal beneficiary but an ai company is looking for some help
Has to he server owner/admin
So like a partnership?
Does this ai thingy have piccolo voice
So to be fair more like product testing and early development assistance for a new startup.
just search on #1175430844685484042 or weights.gg, there should be
It's free for the server and it has a ton of benefits
Otherwise I wouldn't be here offering it xD
Thanks
But I only back things I belive in.
Yw :)
Write an original song and sell it then
We've reached 400000 members,just 100000 & we will reach 500000! Keep up the great work everyone! 👍
Wait could you help me? I found piccolo ai voice and theres some link and I go click it and it asks me if I want to download it. I dont know what to do
Ayo? @ionic anchor level 1 !!! 
You can either use it inside weights.gg, or if it's a link in here you can use one of the Colabs in https://docs.aihub.wtf/ and use the model
Last update: Mar 10, 2024
RVC > Cloud section
SO it is definitely my mic, i just tried RVC ASWELL, same results, no words with the letter S will register
ah, damn
so I bought an actual mic mic, with an arm
niiice
it was time already
balls
im sure a lot of those are fake/spam accounts
Oh,I see
For the 2nd time
were can i ask for help setting up voice metter i have tried everything watched a vedio but no matter what when ever i turn it on according the settings this guy says to set up my audio on anything doesn't work and to add more issue no one can hear me on discord when i try to speak i'm confused fustrated and annoyedhttps://www.youtube.com/watch?v=pEd7jw0uNE4
Ayo? @dense dawn level 1 !!! 
Instead of using VoiceMeeter, use VAC (lite) by Muzychenko instead:
https://software.muzychenko.net/freeware/vac470lite.zip
Also, make sure to read this W-Okada guide: https://rentry.co/VoiceChangerGuide
EOL - No further Updates
Github - Blanc-dot
Discord - Blanc_dot
Despite being end of life, most if not all information has not really changed, so should be very accurate until actual new stuff comes out.
Other Links
Antasma's Local Error Fixes
Antasma's Colab guide
Sushi's useful Links - You need...
alright is it easy to use?
Just set Line 1 (the cable) as output in W-Okada and the input anywhere else u want to use it on
Well, no, but the written guide explains it pretty well
And also how to use W-Okada too
okay.. so i downloaded the file what now.
now just install the thing
the guide explains how, too
whats he tryna do?
if you have voice meeter i can help you set it up
thats what i use for the voice changer
i dumped it for what mj said to get instead.
i couldn't get the dam thing to work. I set it up how i was told but no one could hear me.
mind hopping on a call so i can howe you have it set uip?
what the new one?
i don't have the meeter thing .
but sure i'll give it 1 more chance
idk if it is or not
hi
Hi i don't understand how to skip the queu in Hugging face facefusion
how do i uninstall the voice changer cuz i think i downloaded the wrong file
mostly, but we do have sections for ai art and llm as well
till UMG decides to come after you
thats my render
Me too.
It's some kind of Kasane Teto cover
@quartz roost oh its synthv
dont scare me like that
Piola 👍
Note taken: label all copyrighted content as synthV
i mean i can hear the synthetic voice
thats the point
i didnt even hear it on the first time
Have the first part be synthV 
shes processed on purpose
Bait and switch
now you get it
now check out this voice
is there any discord to find vtuber model :O
Don’t use voice ai it’s hot trash
Use RVC or applio for covers and w-Okada for voice changer
what about voicemode
idk but does not work well for me .-.
i want to use voice changer
Ayo? @buoyant hinge level 1 !!! 
live
bros just stating even worse ones
Prob your settings
Send a screenshot
ahh
i deleted already cuz didnt work well for me
the voice sound so weird
i tried all tune but ...
It’s not just tune
is there a way to get the ai voice changer to work faster? i remember a while ago i used some real time voice changer from here but i cant find it and i cnat find out how to use it again
hi i am bored d'm me
i watched some yt video about and used their setting :O
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
hi modzzzz
Live2D or voice model? (cause it could be confused with either of the two things
)
hai
https://twitter.com/DieselbrainArt/status/1386811841184206848
songs used: Cowboy Bebop - N.Y. Rush, Green Bird
public voice file library: https://patreon.com/GetGianni website: https://GetGianni.com
Hm
hm
live2d :D
Ayo? @midnight saddle level 2 !!! 
check steam workshop or paid an artist to make a model
damn i hope some really fucking hot chick paints my brains in the hallway
there's also booth, there are some free ones as trial there
sorry general chat
Hmm.
HmM…
whats the best application to use if your trying to use a ai voice changer like its normal, like with as little delay as humanly possible
HmM... m
well
it’s more of a hardware issue than software
W-Okada is the best rn
how do i set it up and use it?
-realtime
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
Yesss tea ran the command
First link!
SHOES BB
local or online @timid musk
I dont really know, but im sure there are some usable models you can search for (not the project files tho). for example, try search liveroid model.
Every time I've heard someone use online/colab it sounds like dookie
Try local first. If it's horrible, try colab
Is there a way someone could make an Earthworm Jim voice model?
RTX 3080 or smt
Ayo? @terse nest level 1 !!! 
its either 3080 or 3090
i used to use this like a whileee ago like back in febuary and it was wayy more simple lol
thank :3
but ty for helpin me
i mean you can still use realtime
-realtime
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
Yapperton City
Second link
is as simple as it gets
but UI sucks
and i actually haven’t tested the new version
i tried to use one like that but everytime i clicked "start" it would just crash
so actually just one the first one lmao
Skill issue perhaps? 🤷♀️🤷♂️
fuck you gavin
No u
np 
i can’t read
Same
oh
Im not a bot, don't want to join the npc squad
Certainly, I will proceed to post in the promotional chat promptly.
not you!!
Ok done
thats fire tysm
How do use, I wanna make yukari akiyama sing Erika so I can put it on Reddit for gold 🐱
does anybodyu know the best Wokada settings and how to not sound like a robot? or reduce i9t is it the voice im using or ? my settings
GM
하이
@forest juniper한쿠닌?
@stiff mirageUse a quality model and you won't sound like a robot!
Hello guys
this is an ai voice server my friend
Ohh sorry
We are a team of creative individuals striving to provide the best of AI digital solutions.
Ayo? @junior bone level 1 !!! 
nice, just dont advertise here please
Ok
Their settings would be different from yours
Since your model and your own voice is different
And plus your GPU

hello
Hi
im supposed to be studying but here I am making ai covers
Nice profile aesthetic
Ayo? @inland hull level 1 !!! 
is it gay if i want to smash a femboy?
anybody got "luongtienthanh" in contacts
If you want to smash because of the D then yes. But if you want to smash because of the looks being like a girl then no.
Thanks
I like your background
nah
just say no homo before you do it and you should be fine
Hi,
is there a passable to make AI sounds crispy with high pitch voices? or no?
yes
you need to mix them
in fl
give me the vocals if u want i can help
@sharp cedar
@random geode hi thanks,
I see so it needs manually to clean up? like photoshop with stable diffusion?
Ayo? @sharp cedar level 1 !!! 
yes exactly
but it s harder you need to know mix master
it s a complicated process
i can give u an example i made
i l send a link in private if u want
I see ,do you think Audacity will be good ofr this job or i need better software
ok np.
A poll shows that out of 25K people, 76% of people are still upset that the headphone jack is gone on phones
Ayo? @mossy pine level 1 !!! 
how do you decrease the response time in w-Okada
hello disculpen ahora que ia se usa para hacer los cover?
Hello, sorry, AI is used to cover
help
help
Ayo? @lost nova level 1 !!! 
:c

.
RVC, you can use that on weights.gg or on Colab
erm, comparing TITAN vs OV2SUPER, which one is better ?
Depends on your use case tbf
OV2's much better at handling shorter datasets, while Titan handles everything in general
Though I have seen that both don't really handle noise that well - they'll collapse n stuff, despite giving a decent result overall
I just prefer original because it has no issues across the board, when it screws up I might think of using the others
how many minutes will a dataset be called "short" 😮
Like less than 10 minutes
dammm then every model i train is short lol
i usually use original
but i think today ima try OV2
hope you've cleaned your dataset well

my dataset are not really well, it is 50 audio file, each have length about 3-5s
and i am too lazy to denoise all of them
sadly i already started training
i will try both ><
sir, i keep getting the error when run training
FileNotFoundError: [Errno 2] No such file or directory: '/content/drive/MyDrive/rvcDisconnected/ModelName/%s/ModelName/filelist.txt'
H
helooo
I need the voice of JC Reyes, can someone put it on?
I also need the one from luar la l
can you put it on?
yo
hi
hello!
who can help me generate pictures of cats in ai? please?
you can help me?
i write you
pokkan pokkan umarechatte
Autosave might be on
HELLO!
RVC
Does anyone know how I can run a real-time voice changer from Amazon SageMaker?
I hope everyone has a good day/night
no it not 😦
oh wait it is, my bad
lemme try fix the auto save 😦
wait wtf, i don;t get that error now even when i on or off auto save
Preguntale a la IA como hacer covers.
hi
nice beatsaber to find there 👋
Do you need the model or the voice changer
@edgy bay okay
Ayo? @nimble jackal level 1 !!! 
what
HI all!
Stumbled upon this place and seems interesting!
I understand fully the copyright issues and so wont straight up ask for the obvious. However, im taking the wife and kids to D/Land in France in a few weeks and i'd love to make my own voiceover for a certain .. Mouse creature.... Can i be pointed in the right direction to either Find a good one or learn how to train my own and what kind of raw data works well to train one - and also whaat software i'd use to train one 😄
well that AI bot on here already kinda helped! haha
hi
hi
how does one become verified
chat until level 5
that is actually so much effort
a
skill issue
N
Welcome! You can find one in the #1175430844685484042 channel. :)
Ayo? @sharp mulch level 8 !!! 
awesome thank you - i thought i was clever until i saw some installation instructions installing timm then loads of dos style typing haha
Ayo? @prisma ermine level 1 !!! 
That is only if you have the specs to train or convert audio locally. Otherwise, you may make your voice covers here: https://colab.research.google.com/github/hinabl/AICoverGen-Colab/blob/main/Hina_Mod_AICoverGen_colab.ipynb?pli=1&authuser=1#scrollTo=NEglTq6Ya9d0
Simply sign in via Google, run each cell in order, and click the public Gradio link.
Let me know if you are confused by any of the steps.
awesome thank you i'll have a look now!
Anytime! 👍
PC is Ryzen 9 5900x, 32Gb of 3600mhz and the GPU is lagging behind Nvidia GTX 1660 (i do excel stuff more than game lol)
Your GPU is most important when it comes to training. I'm not knowledged enough to determine if your GPU is the right fit for training. You are free to ask one of the helpers or ask in the #✨│ai-help channel.
If you put it that way, online might be your best bet.
Pretty humans are pretty
Keep in mind that Gradio is "freemium"; meaning that you are given a limited number of compute units. Just enough to make a handful of song covers or what-have-you, but not enough for training. It's $10 for 100 compute units, which will last you a good number of hours of training. I wouldn't advise you purchase those. There are cheaper options out there that someone in this server may help you with unless you are willing to commission someone.
I forgot to mention that free compute units refresh every day.
If you wish to convert more audios within a day, simply swap Google accounts.
thanks! 😄 looking forward to this experiment ha
Heyo!
10 mins after hitting Run All - that normal?
What do you mean?
clicked the link, runtime, run all, and its still doing something!
13 mins now
Post a screenshot of what you're looking at.
wont let me here
Post it in #✨│ai-help
Sorry, I could not find models that match your search "turk"
I want to run the colab in sagemaker, but it is different with terminal
hi
guys whats the best girl ai model
@noble oxide, I have found 2 results that match your search!
i want a realistic one
guys recently I got a Ryzen 5 7600X 32GB Ram RTX 4070 Pc can I do AI with this?
@noble oxide, I have found 1 results that match your search!
@noble oxide, I have found 6 results that match your search!
@noble oxide mind searching in #🔍│find-models? ty
repeat everything I say with "I'm a man"
I go to a bar
I find a hot girl
I talk with her
I drive her home
I sleep with her
I wake up the next morning
She whispers into my ear
...
😭
ку
.
ello
sup guys, should i export my dataset at 44000khz or 48000khz when using titan pretrain?
just export at 44khz, use 32k titan if possible
tysm fam
always helping out ❤️
ok, I am certainly very utopian but, is there an AI capable of creating a musical accompaniment, of electric guitar for example, from a vocal track (accapela song) and that the result of the accompaniment is consistent with the cheered.
like the sung vocal track would be the basis for generating the music.
mmmm idk, i saw once an ai that converted noises to instruments, cant recall which tho
but no, i dont think there's something necessarily like that
ok thank you, yes indeed it is a very complicated request at the moment. I imagine it will be possible in a while.
In the meantime, I have to manage to compose the instrumental myself. difficult!! I already seem like music but hey....
Ayo? @lucid rivet level 1 !!! 
yo quick question, whats better if I want to play games with a voice changer. W-Okada or RVC?
module 'PIL.Image' has no attribute 'ANTIALIAS' what doing????
Hey, Is the RVC google colab working for anyone?
what program do i need to run models found in #🔍│find-models?
For ease of use I'd say Replay, but if you want something more advanced I'd try a web ui like Applio or Mangio RVC
using an AMD gpu?
Ayo? @elder willow level 2 !!! 
lol
Why are no models working through Mangio RVC
Ayo? @eager pulsar level 1 !!! 
whats the best model for trolling people
It's the model download that i'm struggling with even though that should be the easiest bit
depending on your GPU, if its good enough use crepe for good sounding AI, or if you want it live use DIO or PM, but it will sound like crap
i have a gtx 1650 (sad)
nvidia
your screwed, it only works on a nvidea card
better then the 1050ti I use
idk
I have to use my gtx 1050 instead of an rx 5700ti
my rx is 8x faster
if you got 40/50$ tho you can pick yourself up an nvidea k80
those go for cheep on ebay
but it needs another MAIN gpu to work
(and another pcie slot)
hey
There's google collabs
i dont think google collabs work anymore
I just used it last night
You just have to use it very rarely because they are picky with GPU usage
i tried to use one a few days ago and it did not work
Are you training or singing
and i know the collab rvc training notebooks dont work anymore
training
whats the most epochs you have gotten out of it
@elder willow, I have found 1 results that match your search!
500
and extraction is it rmvpe??
dang how long did that take
yup
it was 5 minutes of audio so like 2 hours
not bad
that would be hilarious to see one of my friends sing somenthing like ballin
lol
if you screen record on your android while vcing
it picks up his audio, and UVR supports mp4
i did not know uvr supports mp4
it does lol
that worked
alr
alr
bruh what
Welp... who ever's running Weights GG will be hearing from lawyers!
isn't that free
Yea
Anyone know of a faster and easier solution?
then be glad its even free
google colabs ._.
the best time to do it is 6:00 am and pm
trust me it is the best
most people are offline at that time
i have a rtx 3070 so i usually train on that
Ayo? @fallen plover level 5 !!! 
wait
Ayo? @elder willow level 3 !!! 
i find it harder to get right
o
i usually mess up my models by being stupid as well
same smh
i dont know what rvc file to use half the time either
finally done training, you test it and it sounds like garbage is the worst feeling
i know right
i have been trying to find a good rvc fork to train on any recommendations?
i'm still pretty new
🚨⚠️WARNING 🍎🥐 GORE⚠️🚨
Hello, peoples
🤓
2
3
@clear tangle
🤷
.
Hi anyone knows if exist a Gary Oldman voice model?
hi, anyone knows how to generate people in stable diffusion? when i type someone name from internet sd generates buildings
my sd is clean
so if i need to download something tell me
Literally dad
walter wit
AI is so sigma
it has me mewing and allat
hey hey hey
my alpha wolf side is gonna come out🐺🩸 why am I an AI noob
I fucking hate when i search up a character and i get those ai chats
Weights gg moments
wha
Ayo? @main peak level 1 !!! 
Weights.gg being Weights.gg
yo yo
wasup
It was in a server for a artist or something and they were sharing ai covers, one person said “what if we made a server called ai hub” then bam
Or something
cool
I wasn’t here at the very start so I don’t know the full thing
Tho the server has a lot more history then just that 
ye its cool af - would it be cool to share something ive been working on? wanted to share it with people and get ideas for what you think! its pretty early stages so your feedback will help us build a dope product and see what proboems we can solve and how we can innovate!
Ayo? @umbral rapids level 1 !!! 
What do you mean by product?
Like a marketable ai tool?
platform - basically a marketplace for indie devs can upload plugins and music software!
So a storefront for models and stuff
yesisr
what colab i use to do deepfake?
i think models is going to be sometime down the line, the first is just improvong on the general experince of buying pliugins itself. license management, social experiences and plugins
Plugins like as in?
waves audio music plugins etc any software to manipulate sound!
Oh ok
Why are you asking here besides ai voices?
anyone know any realistic girl voice models 😭
thats the waitlist - we're working on a figma prototypr rn to test out and get feedback on features.
K i hope you will maybe get users
You can advertise in #1159290752195633273 if you want to
nah i think tahts about it
Depends on your settings
i mean ive shared it and if poeple are excited, do check it out!
what settings do i use for a less robotic sound, all my voice models end up sounding robotic
and im always here for a chat, if you want to learn more or get involved. I think ti can help alot of poeple make really cool ,music and also allow innovation in the space 🙂
music is evolving!
@orchid slate what girl voice model do u use 😭
Hi m new here🤭
are u fr
the truth is, kalomaze used a spell and used shadow wizard money gang magic to /spawn it

#🧬│ai-chat Can @keen loom Forgive me?
How is everybody
Hi where can I ask if you have any ai models?
-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
- 🆕 UVR5 NO UI for Google Colab, by Eddy 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
how tf do yall get ai voice working im so confused if thats a real thing 😭
Yes its a real thing, please check out our docs for more info https://docs.aihub.wtf/ and for help ask in help channels
Last update: Mar 10, 2024
LOL WHAT
?
someone said i was using voice ai so i came to recheck here
i was trying to get it working 3 months ago for cyper punk but damn ok
anyway ty
ya that thing
-rt
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
i wanted that ai thing for like making models in games or like dialogs for yt
Most users use the first one, wokada
LMAO WHAT
i wont be using it but thx again
Oke got it yw anyways
ㅎㅇ
Hello, does anyone know why whenever I try to download the model via the spreadsheet, it just keeps giving me an error message despite everything being correct?
Hi, is there anyone here have ever posted some ai covers of bruno mars on YT and then they suddenly got removed but not a copyright strike ?
try finding the model you need on weights.gg, all of our models are there. The spreadsheet is pretty much outdated
It was that easy? I spent like 4 hours trying to make a freakin' cover! But thank you though, I appreciate the help!
we need JSONDecodeError meme for anyone who use outdated colabs.
btw there's no excel spreadsheet for the purpose, or u mean colab?
Yes, I meant colab.
Ayo? @sacred marsh level 1 !!! 
there was a spreadsheet w a list of models in the old server so i assumed he was talking abt that one. Glad he was finally able to make a cover
How is everyone?
hi
Hi everyone, whats the best female AI voice you guys found so far?
if u mean the list of models, yeah it's old, even including v1 models and even background noise & low quality audio, also missing index file that could have been removed from #1175430844685484042 by staffs.
how do i use amd gpu in the voice changer
Good Morning Everyone
hi
Ok guys how do I get started to AI
woohoo, just started my first training on Colab Pro! no more random disconnects and access denial!
oi
there is not Ty Dolla $ign voice miodel?
Guys how do I make AI
argh! browser crash after not even 1% was done! -.-
Ayo? @spark raptor level 2 !!! 
oh wow! with Colab Pro, it's even continuing at the exact epoch where it was after re-launching Firefox
hello!
guys what app do i use the voice thingies?
Howdy.
How's everyone doing?
Depends if you want to do realtime voices (for games/discord), or if you just wanna convert .wav or .mp3 files to a voice
where is the rvc colab?
check in our docs https://docs.aihub.wtf
Last update: Mar 10, 2024
thanks <3
what nvidia gpu should i buy to use cuda ai cuz my shitty amd RX580 wont do, sm around 300$
i wan use it as realtime
can you help me set up the ai voice with AMD please
hey all, is this server to make ai voice lines ?
Discover the perfect summer essentials at Sandify Shop! 🌞✨ From stylish swimwear to beach-ready accessories, we have everything you need to make your seaside adventures unforgettable. Dive into our latest collection now and enjoy exclusive discounts! 🏖️🌊 #SummerVibes #BeachEssentials #SandifyShop
Https://sandifyshop.myshopify.com
what licence am I gonna use on huggingface? 😭
I don't think the people creating AI will be spending their time enjoying the beach on such a hot day.. u r in wrong place bud.
R u ok?😭
Ayo? @lofty bough level 1 !!! 
Can you seperate two singers in a song using ultimate vocal remover?
If you guys know what that is
I wish I could
maybe? it really depends on the song, but try the BVE model. If you're talking about diarization (separating speakers who aren't singing at the same time) then no, can't do that with UVR.
@slender elm do you know the best way to change the bpm of a song without making the vocals sound a different pitch
i need a song to be a bit slower but i want the vocals and unstrumental sound the same, just a slower tempo
I feel like most DAWs can do that
should be possible in audacity right ? 🙂
Yeah, I think Audacity has an effect called change tempo or something like that
Preciate you 🙂
What do you mean?
ture !!
Oh, I have no idea.. I train all my own models, but check out #1191429836321849435 for commissions
hi!
Ayo? @dry summit level 2 !!! 
you cant magically get better model quality from the same dataset, if you want a better model then make the dataset cleaner
hello!
hi
whats your name?
Ayo? @wraith herald level 1 !!! 
mia
hi
hi
i cant listen rn

Does the google colab even work, I keep getting an error whenever I try to download the model.
how to download the model guys?
search in voice-models tab
why is mine doesnt have the “index” file?
Sa
That Colab notebook is outdated, and it's broken
You can find the recent ones + guides on https://docs.aihub.wtf/
Last update: Mar 10, 2024
I think the train model section should go over how to make use of tensorboard to see the performance of trained model over time and the importance of creating small weights with each save. That way you can try out a weight according to tensorboard stats. That is what i am doing for the first time let's see how good or bad it turns out.
🙏 good luck
That's exactly what it's for and it's explained?
The whole point of using tensorboard in this scenario is to check for overtraining and pick the right state





