#π§¬βai-chat
1 messages Β· Page 388 of 1
someone help me with settings. my ai voice keeps cutting out in the end
someone has working colab to training model
darwin model on applio no ui is not working
@solar torrent maybe u can help, do we have other pretrain model to training new one for step 3 on applio no ui
https://huggingface.co/Darwin/Darwin/resolve/main/Darwin.zip that link is not working
they aren't stupid, many games use AI Upscaling techniques, like your Nvidia GPU's DLSS (the Nintendo Switch 2 uses an Nvidia Chip and something similar to this), AMD's FSR 4, Intel's Xess and PS5 Pro with PSSR
It helps to achieve better performance as it's running a lower internal resolution and then upscaled to an higher one
What I'm saying is, that in some way or form AI is already used in all games, like AI Upscaling or NPCs
@glad flicker we don't allow promoting
use #1192011222023950368 for help
RVC v2 is currently the best since years
would be better to use help channels, also not sure if @stark scarab is aware of this since it's her project but hosted by Ilaria
409-782-3384 CALL ME
why
I think the number doesnβt belong to him, and he just wants to annoy the original owner of the number.
Greeting, Have a great weekend!
I've participated in many AI and Web3 projects and received numerous compliments from clients, praising my efficiency and reliability. I won't go into specifics here because it's perfect, If you are building or have idea about incredible things like Dapps, AI stuffs or AI Agents and mobile apps, please let me know.
most likely
Guys what website can I use to get a pth and index file from audio file?
and here is the guide
That's a Windows app?
I can use an ap but website is prefered
Or is it the one on hugging face?
?
it's something that can be used on a browser or on your pc as an app
6 + 7 = 13
M is the 13th letter of the alphabet
M stands for Mangos, Mustard, and Massive
you know what else is Massive?
reminds me of one time when a person came into chat and posted a phone number and said to cell their friend and scare them

some friend he is
Can you download older versions of Applio? I need the version from around may 2024
can i use inference during training?
yes
alr thanks
https://x.com/cocktailpeanut/status/1996724204985565195 nice project for who's looking for realtime tts
you can just check older github releases, but why tho? some older releases even had a broken discriminator in the training process
it's always better to just get the latest version
woops wrong channel, sorry
Unoriginal
do you think the more that we interact with them the more show up?
Not sure, I'm not active here
those weird image scam accounts are like insects
flies that show up randomly that never go away
one gets banned and another joins not more than 10 minutes later
guys i need help about finding a voice model
wdym, you look in #1175430844685484042 and #πβfind-models ?
also been here when the 1st server was deleted
anyways how do u make like
an ai song
like the accapella
i used something with google colabs back then
i forgot literally everything
oop chat got nuked
how do u get voyages on geights gg wtf
why do they cost
or does literally everything cost money
omg
if they did an ai cover of elizabeth lail singing a random lullaby omg,that would be a blessing
are the ram prices gonna get any better any time soon within the next 2 years π
wheres e-girl category
pth files are safe and so are index
thanks, coz huggingface marks anything as safe nowadays
Omg i hope someone makes an ai elizabeth lail voice model and makes her sing a lullaby,i couldn't sleep well since yesterday
Do you think I know about pretrained models? Use #β¨βai-help or #1192011222023950368 to discuss about your issue, rather than expecting one helper like me to provide an answer. 
@fading cape what do you use for your images again?
interesting
Actually now that I think about it, what do you like to do with AI anyway?
generate images, have arguments with my AI i made, not a huge amount
You made your own?
ya, i had to use APIs tho, dos i can't run any locally
do you know Python?
I can't code at all
I downloaded Python a while back and never even opened the editor lol
I have off tomorrow
I'll try to get started
I might need to ask you some questions if I get stuck though
You're way advanced so I appreciate your willingness to put up with my idiocy lol
inevitable if you need to vibe code
Vibe code?
Nah, im not that advanced
Well you know enough that it's a pleasure to code
I'm still grinding just to understand basic concepts
true, true, and i see
Yes, but one of commands I know in Python is "print".
NO WAY
That's a hard one
hi
how's this sounding so far
r u using refinegan
sounds pretty good
thanks
how many steps
ima train it up to 200 and send it to YLW
6K
how long is the dataset
6k steps only on a 40 min dataset?
itβs only 150 epochs
what was your batch size
its already at 1700 steps abt
no but im wondering why your step count is so low on a much larger dataset
if its 200 per epoch then it should be 30k steps not 6k
my ears arent that great
but it sounds fine to me
thx
sharing datasets in any way is not allowed
where do you guys use nano banana pro and veo 3.1
???
is there a difference between diff websites and where is the best
Can anyone help me on W-Okada in dm?

If you didn't know, you should know. Sharing dataset audio (regardless of if it's a copyrighted one) is not allowed here, as to avoid getting this server banned or questionable by some copyright holders either way, as stated in the server's rules. For conversation about model making and quality, there's #π₯βmodel-maker-chat by the way.
The thing Nano Banana and Veo shared is their creator, Google. The difference between them is how they used to generate media; Nano Banana as an image generator, while Veo is a video generator. When you generate either one on mobile or website, there would be less or no difference, unless a non-Google service also provides access to both tools.
For help about W-Okada voice changer, there are #β¨βai-help and #1192011222023950368. Not everything would have to be in direct message.
wsp yall
π
π
how the fuck do u make a uhh
ai cover nowadays
i remember using rvc or sum back then and i forgot literally everything
if u wanna just throw a song over an ai model without having to seperate vocals and stuff use weights.gg
but to make one where u seperate the vocals, bg vocals, and instrumental use uvr
hello
what gpu do u have
???
nvida rtx 4060
youtuber
get vonovox
what?!
whatever he's giving u isn't new
wdym
send the yt link rq so I can see it
okie
yea that is old
damn
ok
kk
https://github.com/dr87/Vonovox/archive/refs/tags/v1.6.9.zip
Last update: November 21, 2025
here ya go
is there a tut
nah, all u gotta do for vonovox is after u extract it, run setup, then run start after setup is done
oh
and vac lite after extraction run setupx64
ok
Reminder that this isn't an help chat, check the help category and respect the #1402790586028789830
oh ok
I had to save him
before he got an old yucky thing
how do i get the voice to not be glitchy
like it cuts out f u talk to much yk
how do i prevent that
can anyone translate what my ai is saying
π
Borrrrrrring
whenever i say something 1 second later it repeats itself how do i fix
If anyone is an expert in AI, I am trying to use it to make a voice like the type of video link I have sent above.
I have tried Kaggle so many times and the website fails and the training only takes like five seconds and I follow like all of the steps. Please help. And if you can, DM me.
Trying to use Applio.
π
π
π
Hi
worth to pin tho 
ai>
π€―
do we have AI audio editing software yet
what would you want to do with its AI slop?
There's AI-generated music, but for a 99% AI-assisted or AI-based DAW software is a no. 
This is the digital audio workstation (DAW) program I use for remaster tracks.
For help about Applio and Kaggle, there are #β¨βai-help and #1192011222023950368. Not everything should be in one's direct message.
That one doesn't look like a braille or morse code. The LLM/chatbot is either crashing out or glitching. You either try re-send your message or restart it. 
You didn't read the channel's pinned message? For help about W-Okada voice changer, go to #β¨βai-help or #1192011222023950368 and explain about your issue, your PC GPU there.
reply
π
π
@deep wren no advertising
Very much
borrrrrring
what I don't like are these image scam accounts
they're boring and unoriginal
i got a .index file, a .pth file, and a .npy file, what do i do with them?
u can put them in w-okada to use for realtime
or upload them to applio to make the model sing
i wanna upload an audio clip
and change the voice
ngl im new to ts idk what any of this means
.npy idk what that does but it's useless for both things I said
bet
btw for further help go here ^^https://discord.com/channels/1159260121998827560/1159290139609137264
since this is to chat and not for help
preciate it
myb
ur good
booooo
π
π
π
ok im #sus π
Meow
how can i upload my model on here
ok so I recently got back into voice model training and I found this, I don't remember messing with this lol. What's the difference?
hey guys can somebody make an Dorozea AI model ?
#1159289738314919936 No ask here, make a post there
okay


π
π
can somebody please guide me fully on how can i create an AI Receptionist? I tried million times but nothing works
π
it is an LLM, it does not think, it only predicts a sequence of text, it predicted that it needs to return 'Yes'
π
π
hello
π
Hey everyone. Im new in ai ml. If there is any teenager who is interested in learning ai ml with me, please DM me.
π
Hey
π
π
π
Hi
π
Don't fuck. 
To make a voice model request, there's #1159289738314919936.
Look up in #π₯βmodel-maker-chat, and search contentvec and spin v2.
Change your password
emm
I think that's the first time I've seen one of those image accounts actually online
hello any Ai's here wanna chat ?
hey yall, i have a question, i'm very new to the ai-voice changer scene, and i'm interested in potentially using w okada's voice changer to start streaming using my OC robot's voice, since she uses an old text-to-speech engine as her voice. What would be the best way to learn how to train my own voice model on clips I can generate using the old TTS engine?
I've no idea at all where to even get started with learning how to train my own model so any help would be very much appreciated! thank you :)
π
π
π
π
what gpu do u have? and what operating system do u have as well (windows, linux ect)
windows 10 and amd radeon rx 5700
oki so I would use wokada tg fork as it's the better one for amd
btw for more help go to https://discord.com/channels/1159260121998827560/1159290139609137264
yeah i did but no response thus far
π
π
π
π
π
π
Is 500 epoch too much?
π
π
π
π
π

it alright. i train to 1k epoch on my models while at school
Does it really make a difference
Chatgpt told me there's no point going to epoch 500 because the best of my audio was maxed at 80 epoch or something like that

the W odaka isnt working for me
ask in #β¨βai-help and read the help guidelines before asking
in my case yes and no it depends on your data set
For help about W-Okada realtime voice changer, better ask in #β¨βai-help or make a thread in #1192011222023950368.
But if you're currently gone from Discord for two months, you can come back in the next year. 
π
"ChatGPT" 
How to make a voice sound good like my specs should be enough and idk what settings to put
Hey everyone! ππ
Happy to be here β Iβm exploring the server a bit and really enjoying the projects and discussions.
I love learning from other creators, so Iβm looking forward to hanging out with you all and chatting about AI! πβ¨
π
?
This was long ago lol
i wouldn't wait very long talking to Echte, its a once in a lifetime chance
π
π
π
π
π
π
π
π
hallo everyone my name is amwar. I have never had a real girlfriend before at age 38
I was wondering if anyone here have experiences with AI girl friend
thank you
-Amwar
how can remove autotune of a audio for a model
i don't know any other way to tell you this but, AI is not the way to go on this one buddy
get off of Discord and go out in public, meet some peopel
*people
π
π
i'm going to make reze model fn
I wish this was real
If I was that guy I'd rather be lonely than turn to AI for that
i know right

It was a joke
What's sad
a joke that could ruin your life is not that good
trying to turn to AI in place of a real relationship, that could ruin your life as seiso already said, its not that good
So... a AI system that lives in your home with the leash off.. bad..good...wyt?
does it have to be something touchable physically?
Guys what ai cover makers that can be used with google colab
Ai girlfriends are the fakest fakes i've ever seen faked
Yo guys do you know the easiest way to get AI celebritiy voices with PlayHT is the best way to just find some inteviews on youtube or smth?
I actually posted a video in AI videos that is about that very concept, of why people might want to turn to LLM's for romance, because of the patttern recognition areas of the brain that AI can trigger.
Hmm interesting
What's the best way to generate celebrity AI voices with Play HT? Are there any good GitHub repo with audio files or something to train the models or should I only find YouTube clips to upload? Thanks in advance π
What's the best ai song cover without using local?
Haven't figured it out yet, still working on it myself.
π
π
hi
im new to this server
and i got the ai thingy installed right
and then i click start.bat or whatever
and a screen pops up for a milisecond then ckises
For help about W-Okada (MMVCServerSIO) realtime voice changer, ask in #β¨βai-help or make a thread in #1192011222023950368 and explain about your issues there.
BBL Drizzy. 
no i mean like places to make ai covers or smth like that
like with using google colab
When you say "the best AI song cover", I just thought you meant by a song track name and not some software/program that was supposed to be Applio RVC.
Applio, use it on kaggle
Or if you don't wanna remove any of the background music from vocals use weights.gg
How to get rid of draft on kaggle?
dont they have an option for thay
Applio used to have a feature that could separate vocals and instruments
π
A separate UVR program is always better for separating audio into stems. An UVR option/plugin within an RVC fork would likely have limited UVR models to begin with.
Can I ask for some help?
Guys, any local alternative to Adobe Podcast AI? https://podcast.adobe.com/en/enhance
sure, in #1192011222023950368
π
π
Itβs for captions
π
π

hi
hello thief
i found it off a youtube video
i didnt know it was yours
sorry
https://www.youtube.com/shorts/3YIhRLf0_Fs i think this video
idk
it literally couldn't have been in a youtube video the model was made 4 days after u did that, plus the link had my name in it
Sorry I found it off YouTube I donβt know where but someone else took it on YouTube but Iβm gonna leave the server for you
you don't have to
LOL
they did leave
they were trying to defend themselves after shamelessly reposting your model os their own
π
π
π
π
π
π
emm
whoops im just an AI enthusiast and wanted to inform people about a Google event
whoops, maybe read the server rules next time

Good moorning π
yo what's the best pretrain for isolated vocals
This is the modmail account, you may DM the bot to make a ticket about an incident in the server.
To make and own a voice model, you simply use your brain, learn things and train one yourself. Don't try to take something you didn't own or make; otherwise, no one here would welcome you either. 
I don't think anyone here would like to join some Google event you claimed. If other people know where to go, they'll go by themselves, just not through some sketchy and questionable links. 
More like UVR5. If you mean by RVC, it's a different one.
hiya hiya i've been out of the loop for a while. what's the go to ai voice changer or is it still the good ol mmvc server?
Both W-Okada and MMVCserverSIO refer to the same voice changer program. The best working W-Okada version is b2397 (not v.1.5.3.18a). There's also Vonovox, which is a complete alternative to other W-Okada versions, but is currently for NVIDIA GPU only. For help about W-Okada, see #β¨βai-help.
hey guys this is probably the third time im trying to download thijs voice changer cause i lwk am too slow to install it lmao π
can someone pls like throughly guide me through ts
Look up my message above yours.
uh im using macbook pro
π
i have a pc as well though
using a 4070tis
so
but id like to use macbook
is that a problem?
A "PC" with a powerful GPU is better. Using W-Okada fork on Apple Macbook Pro would simply run on your Mac chip (CPU), which would likely be slower. Sharing W-Okada from one PC to another over the same network (LAN) is possible, but both your devices have to be on the same network. Again, for help about W-Okada, there's #β¨βai-help to discuss.
ohh im sorry
ah nice thanks
hi
is this the chat where people want all things AI related?
is there any biological intelligence here?
song I wrote about the how AI can be used for cognitive scaffolding in psychology
alt version:
https://suno.com/s/4UfSHVVp8q5ufkLe
this one is about
simulation theory:
https://suno.com/s/kPlxwjwEh1A1LZDK
this one is a trick song:
https://suno.com/s/UjfHoYdUOaFnqGOR
@timid musk sorry to bother you but there is an issue with a member here
seems everyone is a sleep at the keyboard
welcome to the void
where you talk but get no replies

π
π
π
umm where is the e-girls tab for voice models?
thats what im trying to find
Ew...
do u know?
Nowhere to be found, it was deleted long time ago.
yeah i understood
boo
whats the best vocal model for UVR
not strictly for datasets
just the best in general
and it doesn't have to be just uvr
it could be RX 11 or something
i just want the best of the best
this is not a help channel, ask in #β¨βai-help or make a post in #1192011222023950368
I would personally recommend Gabox FV4
But you would have to add it manually to your UVR. Since it's (probably) unavailable on the Download Center.
it is there in Eddy's version
Yup
what about BS Roformer 2025.07 or Gabox Vocals FV6
i use FV6 normally
My favorite one is still FV4.
Works great for me.
π
π
π
what's this?
ad
please don't get scammed by that
Finding deez nutz. 
What's the best way to find audio to clone celebrity voices using play ht?
Good morning Guys
TypeError: Cannot read properties of null (reading 'modelSlots')
at i (http://127.0.0.1:18888/index.js:2:1305771)
at Object.updateServerSettings (http://127.0.0.1:18888/index.js:2:1306003)
after switching to a model, default ones work
Slot system is meh in wokada
π
π
π

wat
For help about W-Okada, there's #β¨βai-help. Asking in chats won't always give you a proper solution.
Hello everyone, can you tell me where to configure Epochs?
RVC (retrieval-based voice conversion) or W-Okada realtime voice changer? For any of these, see #β¨βai-help.
You can't configure epochs
Unless you're training a model
what is the best voicechanger real time
with custom file (rvc)
hi
Booooooo
dumb fuks falling for this shit
I have a friend that has fallen for it twice
It's like some ksi crypto bs
Dumbest scam I've ever seen
seen it for 4 long months, and it has not ceased
It's the most useless scam ever it does literally nothing but annoy people
yo its working
dyk any good settings for it?
Just use the base settings for Vonovox, btw make sure to type in the AI help section for help lol
ops π π
One message removed from a suspended account.
One message removed from a suspended account.
???
One message removed from a suspended account.
Testers for what tho
One message removed from a suspended account.
the problem is automod can't catch them while avoiding false positives for benign users
probably shouldn't
true
Yo
Could someone help me out, I want to make ai stories but I donβt know how they animate the characters ( itβs not sora it looks worse)
And thereβs some sort of lip sync
I will genuinely tip you 10$ on venmo for assistance on where I can get started. I have no idea on what to do
Channels like this
@aidramalabs.tv on tiktok
Seems like they use chatgpt and some tool to animate
25$
Bro come back
π
π
Ρ Π°ΠΉ
xanex
π
π
π
π
π
π
π
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
wtf, thats the husbands job
vonovox is stuck on the launch screen i see there was some sort of patch released 3 days ago is there anyway of applying it
can some 1 do my request
To request someone to train a voice model, there's #1159289738314919936. Anything else that not specified is better not.
i did
Or should I just uninstall it and reinstall
Itβs stuck at the black startup screen
Just don't try to demand your request in here.
I like to draw art, and I like do art better than your fatass yapping. 
Ok, Ruby Hoshino profile. 
Tu-tu-tu-tu, Max Verstappen. π£οΈπ₯π₯
π
I changed it but it sheβs peak ( sad she ended up just like ai in the ending..)
Don't spam various bot commands in here.
i aint even doing it right
That doesn't mean it.
i wasnt even spamming it says on https://rentry.org/fdg_guide_new to submit model to #β get-model-maker-role using /submit command but neither of those are on here that i can see at Ieast
That one doc is long outdated, not sure why no staff member updates that one. No one tells you to use commands in chat channels. The # get-model-maker-role or "submit model" channel has long been removed from this server, neither an explanation, so now it's much harder to get going.
is there a way to get the model maker role still
guys,do you know what is the best tool to change voice from one person to another?
at this moment
Is RVC still the best?
Applio RVC fork.
quick question yall
i just found out about these ai voice changer things
what does epochs mean
on the models
thanks
In artificial intelligence machine learning, "epoch" refers to how many cycle the entire dataset being passed through the algorithm overtime. In "RVC (retrieval-based voice conversion) context, the term has similar meaning but applies to audio dataset that used to train a voice model, which is the exact voice model type also used in voice changer (W-Okada).
Generally, the epoch might not sound meaningful for inference in general, though it's a fundamental feature in training, where one must set epoch to a specific number to achieve best working model as desired.
that's outdated, there was a newer method than that but it's currently being reworked too by @dawn temple
Hi, is there a tutor?
@vagrant veldt you a turkish?
ye
π
-colab
Google Colab is a Cloud (Remote Good PC) Service. While the Free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.
by IA Hispano
Google Colab
by Hina
Google Colab
by Eddy
Google Colab
by Eddy
Google Colab
by Deiteris & Hina
Google Colab
by Shiro & Eddy
Google Colab
by Nick088
Google Colab
by Nick088
Google Colab
by Jarredou & Makidanye
Google Colab
π
<@&1159293140440723499>
π
any new releases guys, like suno software or we still using rvc applio etc
Hey what's up guys
π
Free blocks for me on those image accounts
π
what is a good voice cleaner im on mobile
mdkmd
π
π
π
π
π
I'm not sure if this is the right place to post this request, but here goes. Anybody know of a screenreader accessible desktop app where I could import .zip files of voice models and make text-to-speech and audio? I had Replay installed on my machine without warning one day, and when I tried to use it, JAWS (Job Access With Speech) won't read it, as I believe the program is all image-based and not text-based control parameters. Specifically, I have a few voice models I like from weights.gg, and I wonder if there are desktop apps that these are supported on.
I'm not sure how you would report a user, but you can always block them. I did that on another server when someone was being a bully to me regarding my visual challenge. Out of curiosity, is this user teasing you?
What's on that link? Is that an app for RVC?
just murder them
Itβs a skribble Io game
i trained a model and it only shows the .index file, what do i do?
Good morning
That sounds like fun. Scribble Io? Sounds like an online version of coloring with crayons or pencils.
If anything help about AI program has to be in one's direct message, what is #β¨βai-help even for? 
Maybe you can use applio, applio uses gradio and possibly it is compatible with "jaws.
Even if it sounds bad, most applications that use RVC are not prepared for these cases, because no issues related to this were ever reported.
π
my voicechanger is a bit laggy and ive changed almost every setting. can anybody help me ?
How much does that cost, and where can I find it?
I'm not sure I understand what gradio is.
www.gradio.app
#β¨βai-help -rvc
Free
Open source
If I download Applio, hopefully it won't give my computer any viruses.
It won't
As long as you get it from here
Is applio rlly that good?
I've heard ppl say a lor about it
Ion know because i use weights now but its getting corporate af
so i need to change
my voicechanger is a bit laggy and ive changed almost every setting. can anybody help me ?
anyone have the Amanda Silvera rvc voice model file?
die
ask in #β¨βai-help and read the help guidelines before asking
Does anyone know which AI is used for the ai cat dancing videos?
God, Chatgpt has gone off the deep end since 5.0, it can't retain any memories, follow any hard rules or follow any maste rprotocall or rulesets for more than a few prompts.
Lmao. You're really on a crusade to find out how its made.
To go google, type in AI video generator, click literally any one. type: Static camera, cute kitten in a [stupid] costume dancing, still camera, center screen, background of [whatever] focus on cat in the center, and dance moves. No other movement.
yup!
No one would answer you because its so obvious how it was made anyone should be able to figure it out
yes! After like 3 prompts it forgets the previous instructions, and then you go into a loop of failed generations.
i cant
I want to import a specific emote, not sure where to find that
You want the cat to be doing a specific dance?
Then you need blender not AI.
Unless its a popular dance the AI knows.
exactly
its uncanny, i need it!
You want more anthropomorphic hamster people?
makes sense
what about uncanny hamster?
That should be burned.
oop, rank up
yes, thats the great part!
i dont know which tool
i suppose? "Image to video?"
Well you can do it a couple of ways.
Viggle?
You can write it down, describe it in great detail the way you want it to make the video and then do that. Text to Video.
Or you can get another AI to generate the image that will be like a screen shot of the video you want to make, then do imge to video.
but that gif i sent of the hamster, that was a specific video that was inputted.
I use chatgpt to write my image prompt.
which ai, may i ask?
Sora 3? is that out?
Lmao, I mean Veo 3.
ah, now i understand. Thanks!
But, Veo 3 can't make anything look bad. Everything it makes will look SUPER polished and shiney and too good.
It wont make your creepy hamster videos.
Most AI have a pretty bias, so they can't make ugly things.
on purpose.
But that video is an image attached to a video
Gross.
these 2 are made using the same Ai, i think
its the same video. one is just slowed down
The hamster one, was heavily edited tho.
Everything AI makes is pretty.
Shiney.
Someone prob edited it to make it look old and cheap and badly lit.
Which is hard.
i believe its an ai
You can heavily edit AI.
exactly
They took a video and overlayed AI on it.
None of the legal models are gunna do that for you, maybe open art AI.
You'll have to find a personal model someone made.
Learn how to use it.
And do that
ok, thanks. You got any in mind?
or where to find them
I only work with the commercial ones right.
You find them in that channel
Find models
XD
i can only seem to find voice models
But thanks for your help, have a great day!
Np gl
yo what's good
π
How to use applio in keggle?
Try viggle
can somebody help me ? My ref time is at 12k and the voices are very laggy
i need the best female voice model
like it works for me
but sometimes its just not smooth
and i lowk need suppression
can anyone help me with that
please
@viscid lion no advertising
DYWMTCOASYDTYNIMM
π
π
hi
This is true

how to use the ai voices
For what
Hey chat 
Hi.
Hi :3
The best working female model is, look up "Hitori Gotoh from Bocchi the Rock" in #1175430844685484042. 


I am new here, maybe there are other channels for questions 
This server seemed interesting, since I saw a thing "AI voice". I am not too sure, what is meant by this.
ChatGPT speaking, or some revolutionary real time voice changer with ai π
Option 2
Retrieval-based Voice Conversion (RVC).
O_o
How does this thing work? It like picks up your voice, you slap some model on top - and it ... adjusts it? π
To voice of some celebrity, or even another gender(that would be cool)
For help about W-Okada realtime voice changer, there's #β¨βai-help for such query.
RVC is speech-to-speech; applies a given audio to sound like a specific voice model, that voice model can be of a famous person or fictional/cartoon/anime one. 
Careful, don't fall into the pit of weird ai girl stuff
Seems like a thing i should try out. Seems like those are local ai models. How heavy are those things to run?
Like rtx 5070 and 32gb of ram will be enough? π
π
I guess it's dependent on the model ... how can you tell, if it will run ok?
Too late πβοΈ
I somehow missed that ai exist for those 
The thing is you don't need to understand those Tung Tung Tung Sahr-type memes and you'll be fine. 
But for those models to run, what app you should install?
Some are just zip files, and others this weights platform?
For the real time changer, I saw only record voice sample, and it will transform on that weights π
W-Okada is realtime voice changer that mainly uses RVC voice model. Applio RVC is an RVC GUI fork that is best known to do AI cover and even train an RVC voice model. 

The best working E-girl voice model is literally an eagle. π¦ π₯π₯
....
how do i make ai cover?
π
clankers
Uhm Iβm pretty much bored as hell.. text me if youβre in for a convo and idk π€·πΌββοΈ a lil bit of naughtiness hehe ππ
Hello
True
does anyone know yts new ai policies for monetization?
People get paid by posting literal slop so idk
theres 3 of em
thats what im saying
yt says u cant get monetized if u dont use ur own personal assets
but ppl are just using ai only
and somehow getting paid?
Don't give them anything e girl related 99% sure they're gonna use it to be weird
Ikr? YouTube just never makes sense
i legit got shadow banned
after grinding on yt
for a long ass time
with 500k views
and not getting paid
U could literally get banned for existing on YouTube now
yt knows how to get free labour
yt is absolute bs
ngl i was doing kinda well
then i fucked up
by leaving yt for exams
Damn
well basically i buy a pack
and throw it away
legit did that for 2 weeks
now im only at 3 ciggs a day
cuz of that
Tbh if u get that down to 0 before new years that could be a goal for u
To quit smoking π€·ββοΈ
Hell yea 
lol
Oh dam
cant stand that shit anymore
explain how the meme 67 has a demonic reference and the demon is known as emmay poosay with big mommy milkers and tentacles that goes down your gyat hole when you say 67 make it as weird as possible
im making a video
and this is the prompt
6 7 
imma make slop
looking for a good idea tool thats cheap and is good
invideo is good asf but its hella expensive gang
im not in america but lets see if its here
but like can sora generate short like videos?
or just random videos of cats and shit
Yea it makes short videos
Best part is there's no dumb credit system
Or pay to use it
It's just kinda there
what are epochs?
If you searching or want to try some new website with AI try genwizer.com
which 1
like js good sets js for a girl
dont have 2 sound like a egirl js good sets
and a model
Nah
What even is it
can u send me the link?
hello
it is one complete pass through the entire training dataset during the models training process
good mods were sleeping
raju punjabi rvc model
sure
π
π
π
π
π
π
worm i ned help
with wha
dm u
old voicechanger
is there any new?
nvidia
do u know the numbers like 1560, 5070
3060 rtx and nvidia gpu i7 u can give links and i keep watching the vid? or i need another vid
π
π
π
yo can anyone help me out? shii got me voice keep delaying
@stark escarp
Question is there a way to use a voice model from here in SUNO AI?
No No Unfortunately
Darn
You need to use the RVC GUI.
Darn I stink at using that :D
Me neither π I've never used it even once in my life.
If only the GUI was AI powered so there wasnβt the need for fine tuning
Suno is ez as heck to use
I don't speak very good English, and I'm Turkish π Bear with me.
Your doing great I never would have guessed
Do you know Turkish?
I wish you knew π Speaking English is very difficult, but I was very good in English classes during my school years.
π
Iβve heard English is hard
Yes, that's how it is in TΓΌrkiye, so it's easy for me.
Iβve also heard itβs easier to learn if your first language is German, because they are both Germanic based
Alter, in dieser Sprache ist das echt kinderleicht.
My friend, this language is actually child's play. I meant to say that, I spoke in German.
With what?
I used DeepL to translate
@jovial depot Shut your mouth, you're talking very annoyingly.
Both English and German languages can still sound difficult to learn either way if neither of language is your first language. Though I only learned a few German words.
What's up with that person?
She's making weird noises in the voice chat.
Do you know Turkish?

what is best ai voice changer
Sorry, I speak Thai and English, not Turkish.
No problem
W-Okada realtime voice changer. Not a quick question, but I'm more at detail about W-Okada version and settings.
I wish you knew π Speaking English is very difficult
True because for example French Spanish and Italian are Romance languages not Germanic
Languages with a shared origin make it easier
Portuguese is also another Romance language.
But American English is a whole other level of crazy then British, because American English adopts words and some rules from many different languages. Adding a chaos factor to it.
That and it frequently breaks its own rules. For example βtheirβ breaks the βI before E except after Cβ rule
That still gets me and itβs my first language
On a different note, any tips on how to avoid voice peaking and artifacts while maintaining the models voice when using GUI?
And who are you?
I was in vc I'm a man and he is just a troll I think he wasn't even speaking or present
That's why you weren't speaking okay
xwhats the best sets for a girl voice changer dosent have to sound like a egirl
or model



