#🧬│ai-chat
1 messages · Page 379 of 1
thank you
To make sure you ain't use the program for something bad.
its public resource i guess its not ur business what am i using it for
Just downloaded beloved Ganyu's voice, what should I do next?
@twilit gate @dawn briar use #1192011222023950368
and also, we need to know more info, such as the pc gpu and what you need to do as people confuse programs, so please be understandable
you mean the gpu name? and I'm using for AI cover, I already have the separated vocal of the original song
Yes, the GPU name (e.g. NVIDIA GeForce RTX 4070 or AMD Radeon RX 6600 XT).
gpu name yes
it's all in the guidelines #1402790586028789830
Made a song inspired by the 2019 live-action Aladdin movie. Inspired by Arabian Nights. AI generated of course
This inspired from the same movie, but instead of Arabian Nights, it’s Speechless
Intel(R) UHD Graphics 600
I've made many AI covers before even if mine is this one
This is an integrated Intel GPU. Any integrated GPU is not suitable for AI.
@solar torrent @dawn briar please move this convo to #1192011222023950368 or #✨│ai-help
Huh? Did you make AI covers locally with only CPU all along? Have you tried other online options than the "old" RVC Hugging Face space? 
i'll try
I've told him, but he kept talking here. 
no, u didn't
M8, as said there I already did some AI covers with my current PC even with Intel. However I only used the "old" RVC Hugging Face space, yes, and it worked
the problem is... it doesn't work anymore
You probably thought I was saying this to another user, so you might have ignored it so. Going to #✨│ai-help doesn't have always be W-Okada, it also applies to other AI programs like RVC and Stable Diffusion.
not a good explanation as I don't know what W-Okada is but I got you, I'll keep in #✨│ai-help
I would not expect anything different. it is IPA voicing model, it seems
IPA?
international phonetic alphabet
ohh i see
I'd say even eSpeak NG is better
ipa reminded me about the file extension for iphones so i was confused 😭
It's raid!
MODEL HER VOICE SINGING NOTHING IS GONA CHANGE MY LOVE
algum br ai?
Opa
ja instalou o voice?
Does anyone know if there's an audio restoration program like Adobe Podcast that works well with sung vocals? It works fine with speech but when it comes to singing the results are quite poor
if anyone knows how to tune corpse voice dm me
im in tensor and trying to select my model in "runs" but dont see the kalomaze one
the what one? elaborate in #1192011222023950368
you can ask for help in #1192011222023950368
do yk how to tune it?
you need to elaborate what you want to do and mean exactly in #1192011222023950368
i don't understand what you're talking about, and this isn't the right channel
Não conheço mano, o que é?
@wary sluice @haughty saffron please speak in english only
what?
@grave drift don't promote here
where do i find the guidelines regarding naming models
@covert lake
Can u please help me out
Check pm
Idk why my voice isn’t being changed on okada
please elaborate in #1192011222023950368
wdym?
if i make a model and want to distribute it what format must i follow for the name
I did
Boa noite
Você pode me passar o servidor brasileiro?
I sent it
Hi does anyone know a way to use the rvc gradio as api?
Hello, how do I download the app?
this might be the wrong channel but so apologies if it is. I'm trying to create an acapella using the colab link, but it doesnt seem to be working. I've used it before maybe about a year ago. Is there something I need to update or some?
thanks
I have notifications set to "nothing" and the server muted, why tf do i still hear notifications!?
cause Discord hates us lmao
clearly "No does not mean no" to Discord lmao
apparently...
Hello, is there any way to make it so my Realtime Voice Changer W-Okada doesn't pick up background noise that isnt my voice?
I have an AC on and it gets picked off by the microphone and in turns makes the voices make extremely strange sounds
dms
maybe theres a ghost trapped inside of it trying to speak to you from the other side
-rt
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested WebUI with the best general support for many platforms. GUIDE
A Realtime Voice Changer with similar performance to Wokada Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. and without cloud options GUIDE
For Windows Nvidia, Both Wokada Deiteris fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Deiteris Fork Pros&Cons & Vonovox Pros&Cons
These options are not recommended for use.
Not suggested, older versions in youtube tuts are even way worse. GUIDE
The program is worse compared to the ones above, and much less updated. GUIDE
Press guide on W-okada Deiteris to get walked through the steps for real time vc @tall ether
Ilaria it's not working any alternatives ?
is there a particular reason why training doesn’t support 44k?
and is there any possible way to train at sample rate of 44 with rvc?
hi guys what is a good man voice changer if im a woman!!!
Just wondering... Are there any voice changers that won't go robot even if I laugh?
Hey guys! Which version of KLM would you suggest for Female Singers?
does anyone know any good api for mask feature. Im currently looking at ideogram but was just wonderign if anyone knew anything better?
Discord Data: https://20000.online/openmicae
Experimental LLM model: https://20000.online/micae
Does anyone have that chart showing GPU differences with RVC/wokada?
What model? Do you mean W-Okada or RVC? For any of these, go to #✨│ai-help or #1192011222023950368 and explain about your issues there.
W-Okada is realtime voice changer program. W-Okada uses RVC voice model, so you can't really expect it to make "laugh" sounds that good.
44000 Hz doesn't exist. There are 32000, 40000 and 48000. Some RVC forks may have 44100 Hz (44.1 KHz) option for training, but this is experimental.
Hello, I'm throwing a message in a bottle. I'm looking for someone who is knowledgeable in the field of AI (with a diploma) to help with the structure of courses for one of the first AI options in a college-level establishment (in France). Thank you in advance for your help.
32k = 16K*2, 40k = 16K*2.5, 48K = 16K * 3, 44.1K = 16K * ???
Hello, someone ever worked on video editors? i'm having hard to sync transcription with the actual speech in timeline..
there isn't really a name formatting
please speak english here
please ask for help in #1192011222023950368
this is a general ai server, the app could mean thousands of ai programs, elaborate in #1192011222023950368
elaborate in #1192011222023950368 pls
you sure it's from this server? you also disabled mobile psuh notifications?
are you trying to do eboy trolling/catfishing or are u trans/doing roleplay?
no, are you trying to do e girl trolling/catfishing?
where download aplication voice?
What application/program? If you're looking for W-Okada or RVC, go to #✨│ai-help.
hi
What's up? Sunaookami Shiroko profile picture. 
yea
im
ahhhhhhhhhh
i got praked
You would need level 10 to get gif perm.
alr
If you're looking for W-Okada the realtime voice changer program to use with a Blue Archive student voice, go to #✨│ai-help or #1192011222023950368 and explain about your issue there.
thx
With great power comes great responsibility
And people with power use it as they like
Very nice
I'm doing research about using small models (that can be run in consumers devices)
And the approach in your post makes sense with the idea of using smaller models
Specially since their context window is smaller
But as far i read too even if the model has a huge context window, its better to not abuse it
eboy trolling
warn me
don't pomote here
nick088
what
kick me
There is a channel for that?
Nope, promoting isn't allowed at all here in this server
Not sure if that is reasonalible, if i write an article and i want to share with people here i'm not allowed
Even if the article is relevant for the discussion
fr
nick i said im sorry
do you forgive me?
It's written in the server rules id:guide
what? are you still looking for eboy trolling help?
no i js want your forgiveness
wdym forgiveness? I warned you as that's the right moderation action
i js want you to forgive me beacsue im sorry
nick r u real??
Yeah, its talks about a designated channel to share stuff
yeah
the "designed channel" is an old error, @dawn temple please remove it as we don't have that channel anymore
but tbf it used to be there, so i wouldn;t give him a warninng for an error
I only told him to not promote it verbally not via sapphire's warn
so? I just edited the message lol
@inner ether you're not gonna get help for what you're asking for, idk why you still keep talking about forgiveness
i dont want help??? i js want to say im sorry
?
and i want you to forgive me for rbeaking the rules pls?
But if we share some useful article about AI, it is not advertising or self promotion??
This is a server to discuss and learn new things about AI isn't it?
bro im scared some models might be a virus
promo chanel for ai tracks would be cool
and im also scared that the voice changer client is a virus
because they are all stored in zips
Something being stored in a zip doesn't necessarily mean it will have a virus.
with what site should i make ai voice covers?
@covert lake I'm not trying to bother you or in any form your authority, just saying that in this no advertising rule there is this text saying "no direct advertising" which for me is not that clear what is direct and indirect advertising
it's just one of the most common file formats https://en.wikipedia.org/wiki/ZIP_(file_format)
ZIP is an archive file format that supports lossless data compression. A ZIP file may contain one or more files or directories that may have been compressed. The ZIP file format permits a number of compression algorithms, though DEFLATE is the most common. This format was originally created in 1989 and was first implemented in PKWARE, Inc.'s PKZ...
@wary sluice @whole knoll the article was your own, and you have sent it here to advertise, we don't allow that
like best chatbots?
chatpgt 5, gemini 2.5 pro are pretty good
Yeah looking for one to generate prompts
gpt-5 is kinda. useless tbh
for quick coding id say gemini
That is true, but there are different ways of sharing something in the server, the way you did was direct advertising to get reads on your articles. If I was on your place, I'd share it saying that, "I wrote an article, what do you guys think about it?" instead of sharing article links without any context :)
It's okay, don't worry! 
Just take care of it from next time :D
Please create a post in #1192011222023950368
sorry
Don't be sorry, it's okay :D
Found this on reddit wtf
smartest model they said
But i send it to share the knowledge that's why i added free link in the blog and i am not charging anything. I am just sharing interesting and read worthy articles and blogs to expand the knowledge.
maybe search "french" in #1175430844685484042 , is there a specific character you're looking for?
welcome back lol
I know you're not charging anything, but you're still directly sending out of no where your links to advertise yourself
for help you can ask in #1192011222023950368 btw
be sure to not use youtube video tutorials
how can i train my custom voice
yeah, gpt-5 is sucky
elaborate in #1192011222023950368
i replied in #✨│ai-help
I saw lol, the personality especially
sure, elaborate in #1192011222023950368
Hello
Welcome here!
Hey i don't have money to buy 11labs subscription I want to generate free ai voice unlimited can someone help me
it's better you elaborate your help request in #1192011222023950368
Yeah dude I done
can this be an alternative to rvc?
or it just captures the general essence of the dataset rather than the voice?
It seems for Music Generation, like Suno, not Speech To Speech like RVC
https://github.com/Plachtaa/seed-vc is an example of Speech To Speech project, is it better than RVC in quality? No, but it doesn't require explicit training and is 0-shot
This doesn't look like a typical RVC fork. This is more like to train a whole song model, not a speech "voice model".
yoo
Hi.
so it wouldnt capture the voice rather than the "flow and style" ?
i need to find something better than rvc i still cant believe theres nothing in 3 YEARS of ai development
which is 30 years in real time progress with these rythms
RVC doesn't generate the whole song like Suno and Udio.
i dont care if its t2speech or speech2speech
i need to capture a singing voice with 100% clarity
is it possible yet?
with any method even if i have to type numbers
it is possible with speech , why not signing?
For me, the easiest way to make a music is to go to FL Studio and make beat there. 
While there are alternatives that work similar to RVC on generating vocals but of different architectures, still.
what happened to rvc 3
they would get rid of the silly gan
did they really turn it to a SaaS (weights) and stopped everything?
Calm down. RVC v3 hasn't existed yet. If you see an RVC fork that claims to be "RVC v3", it's usually RVC v2 in disguise.
i remember snoop dog saying v3
never happened and never will
it's been 2 years since rvc v2, the original devs left it to rot
there have been numorous projects which meant to be "rvc v3", but no one with actual success, like project Voyager off a year ago or smt lol
Rip RVC v3
so which is the current best and affordable method of voice cloning (no gpu)? is it still rvc2?
RVC V2, but every training required a powerful PC lol
wasnt there something like a huggingface option that didnt require it?
Ohh sorry I'll add the details and brief explanation and then add the link for detail article.
Still.
Speak english, please!
cant download trained model file from https://www.weights.com/create/song ?
yo wassup nerds
True.
where can i find trained models which i want
Read my replies in #✨│ai-help.
Hi, anyone knows some AI for delete censorship in hent anime (video)
ai chat, is this like a chat with artificial intelligence?
link to amd downlaod
who want to make an ai album with me inspired by playboi carti
Bet
I have a weird almost tunnel effect on my RVC voice thats been described as talking through a tunnel, any ideas as to whats causing it?
Hi guys anyone knows an ai that can generate almost irrecognizable images
eehh hey
well i create ai pictures and i don't what chanel should i use to show my photos
sorryy im kinda new to this stuff
appreciate it
i can't post pictures 
nvm the photos sucks anyway
sorry for wasting your timee
level 10 for image perms
unless its updated to level 5
what is that
chat in the server to level up

do yall think that chatgpt is gonna do the same thing gemini did and give college students free premium
The servers not dead~ it’s just sleeping

Hi
if my data is 44k, but i trained at 40k, will training at 48k be beneficial? my logic is that if i train at 40k it downsamples my data, whereas if i train at 48 it may keep most of the sound as if it was 44, since its not downsampling it
not really.. if it is not full 48k
it is little better with precision though, but wont be noticeable for regular speech and probably singing as well
it is not music after all
it will be more beneficial than detremental for the model tho right?
yeah im just tryna make it as high quality and real as possible
40k and 48k are unrealistic due to the vocoder being mid at reproducing such frequencies
it just makes the model sound more fake
ah so you saying 40k pretty much dont exist with ai?
most ai engineers works with 22050, 24k audio
rvc is weird for having 32k, 40k, 48k lol
im putting my infered ai audios on spek and it shows 20khz(40k) though...
yeah i agree lol
well i trained my 40k one about a year ago and ngl it sounds pretty real
here ill send a sample
@glad nebula
well the reason for this is because waifus

most of the stuff here are in the 14-16k range, there's nothing in 20k, it's pure noise above
which explains why its realistic
oh i also had a big question, you know how if you dont import the real file type in spek, it will show a different sample rate than the real one, like if i import a mp3 but the original is a wav, is this just reflected in spek? or does the quality actually diminish
ah gotcha
uhh i don't understand the question, but spek is always going to show you the sample rate of the file you import
when you convert a wav to mp3 there is a compression, frequencies you cannot hear are removed in order to save space, but the sample rate stays the same
gotcha ok this answers my question

i was asking this because i had an old audio that is from a compilation of audios for my dataset, and it was a mp3, and it showed 16khz on spek, but the original files were a wav, and i imported those into spek and it showed 22khz
so i wasnt sure if the quality diminished that much, or if it just had to do with spek
oh maybe when you converted that audio to mp3, you also changed the sample rate
if the og files showed 22khz it means before the compression they were either in 44k or 48k
but after the compression they were converted to either 32k sample rate or 40k (most likely 32k)

Hello , everyone i am new here
i am from india and want to learn about AI just so i can fine tune it to my business , what things i should learn
no
ooh ok
im trying out these voice changers but its not really sounding like the character- am i supposed to just keep tweaking it over and over?
Be sure to not use YouTube video tutorials
Sorry. I cannot engage in such conversation. Let’s keep our conversation respectful because I feel very disappointed in you because you are swearing. Im feeling Threatened if you keep swearing at me we might get the police involved because this is very uncomfortable for me and you and this is also very dangerous please stop with this act for the love of my heart let's be Respectful for each other thank you very much for co operating.
Sleeping?? half a mil or so of members?
🙃
hi everyone, help me install or find an app to change my voice
Elaborate in #1192011222023950368
might be being stupid, but I've just installed the frank ocean ai model to see if I can remaster an old track of his, how do you run the bots? I have no clue what I'm doing sorry
wheres the downloads
there's no discord bot for ai covers, use #1192011222023950368 and elaborate more
this is a general ai server, you gotta elaborate what you want to download in #1192011222023950368
that sounds anything but real? lmao
@lost raven please don't promote here
all i can say is that rvc is not great at reconstructing frequencies above 24k coz technical limitations
it's supposed to create outputs better than ground truth
noobies is still working on it
his goal is to have better high freq reconstruction from what i see
That means that the high frequencies will be better than with highfi gan right?
Or whatever is used rn
Is it out for public use right now?
yep, they're already better compared to hifigan
no public pretrain as far im aware due to refine still being wip
ah wow
i have heard noobies refine examples and they're more clean than hifigan but also i feel they sound more fake
that’s crazy
I definitely wanna try it
is there an option on rvc to change vocoder or something?
it's hidden, but you can enable it, tho since there's no pretrain you cannot train models
how could I enable this?
applio folder > tabs > train
replace the train.py with this
Thank you!
Will that get rid of the old one tho?
Or does it include high fi and refine as well
it's going to look like this
oh thats great
you need a pretrain and the only person who can give you one is @chilly lake 
and also you have to enable multi scale mel loss
gotchaa lol, so i do actually have to train with refinegan, there is no way to just infer with it right?
like take an old model and run it with refinegan is what i mean
in order to infer refinegan your model has to be trained with refinegan as well
gotcha
would this be in tensor?
or
i havent seen this setting so im not too sure where it would be lmao
applio > rvc > train > right click the train.py file (open it in notepad)
scroll down and find multiscale_mel_loss
to train with refine you need multiscale_mel_loss = True
gotcha
hmm i hope im able to get the pretrain lol
i would love to test it out with my model cuz i mean it already sounds pretty nice and i just want to see how much of a difference refinegan makes
i mean refine is still pretty much wip (which is why it's hidden by default)
the only person who can answer when it'll be done is noobies tho
gotcha
it is not great with multiscale
perhaps even worse than single
I have 32k pretrain cooking
the more you know
i thought it was multi
do you think id be able to get access to the pretrain to test it with my model?
btw mezatl disable multiscale, set it to false
gotcha
it is nowhere to being done
any plans of sharing it?
you can try if you really want
that would be great lol
i sent you a dm a second ago about it if thats where you want to share it
thank you! ill download in a second
do you also have any examples of highfi vs refine gan?
im curious
you chat
I want to buy a Ai course
We don't sell any courses and never did
why
This is a community server for help, not for condoning advertising or jobs
You did not understand what I mean. I want to start learning artificial intelligence and I want to know if the course is useful to me and worth buying or not
what are your options for the AI course?
@flat owl @amber panther I understand, but we don't promote any ai courses
you can learn AI from literally anywhere
If you want you can do more researches online, the HugginFace free documentations could be helpful!
Is using Local Ais on my Rog Ally viable, and its possible to use SCALE or ZLUDA in ANY Ai at all? I want to use ai for game development(with quality control of course) and those online ais are too useless thanks to their scammy memberships.
Hello, it would be better you specify which program you want to use and which rog ally you're using in #1192011222023950368 :D
Ok
Hi
Hello, hru?
Good, and you?
@lone wren Please don't look for jobs here / advertise, this is a community server not meant for that
Hello
So does anyone care to discuss where ai is leading us not just as a civilization but as a species id love to speculate or if anyone has a defined view point id love to hear that too
Id like to see ai lead us to a utopian society personally
where the heck is egirl voice models
Can we make one?
An egirl voice model that is?
@glad nebula hey sorry to bother again i just had a quick question, when putting my dataset audio in the folder of rvc, can i include multiple audios instead of just one long audio? i dont mean that im manually trimming, its more like, 5 10 minute audios if that makes sense
yes you can put multiple singular audio clips and auto slicer to automatically remove the long silences
in rare occasions this causes an error in training
safest practice would be merging everything into one singular file, truncating silence and using simple slicing
the reason I’m doing this is bc I don’t know what the reason is, but my original audio shows 22k on spek (for some of my data) but once I export everything together those same audios lower down to 16-18k, I’ve tried everything and I am exporting with the highest quality settings. When the audios are shorter for some reason spek actually shows 22khz, but when it’s all compiled they lower to 16-18
never had such problems so idk
Hello, I have a question: does AI use a video card or a processor? I have a good video card but a poor processor, and I literally cannot hear anything, and there is a significant delay. What should I do?
Yeah it’s odd, if it makes the training crash will I know right away?
or could it happen randomly
it'll throw a error that goes along 'tensor does not match whatever'
are you trying to do e girl trolling?
you can use either a processor or video card for AI
ofcourse, a powerfull videocard is 100% suggested as CPUs suck for ai
@chrome berry so whats the best settings cuz my voicechanger sound shit
Hi
how do I look at the vc stats?
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested WebUI with the best general support for many platforms. GUIDE
A Realtime Voice Changer with similar performance to Wokada Deiteris Fork, with extra features, but supported only for Nvidia GPUs on Windows. and without cloud options GUIDE
For Windows Nvidia, Both Wokada Deiteris fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Deiteris Fork Pros&Cons & Vonovox Pros&Cons
These options are not recommended for use.
Not suggested, older versions in youtube tuts are even way worse. GUIDE
The program is worse compared to the ones above, and much less updated. GUIDE
work for the irs
Hello I apologise if this is a very common question but does anyone know if there's any super easy AI voice converters to use like Astra Labs? That was my go-to for making AI covers but it shut down recently
Anyone know how to make voice ai model free? Without gpu/cpu?
Finally good news
LOL
Y'all what's the BEST ai of all time? Like I need an easy ai with easy settings
oh thank goodness its back
Who
Oh. Eight
So what u use for ai? Like voice changer
Is it peak?
yeah
I don't mind if it fuck my cpu since idc🔥
what batch size is recommended for a 1 hour long dataset?
Why did the okada fork choose to go to a web based menu and not stay on an app ?
@green crescent
hmm, does Google Colab or Kaggle no longer work? Are they paid services now?
Hi I m new
same here bro
i havent got a clue
i keep getting ''failed to fetch''
which tool?
the server?
try refreshing or restarting your browser
sorry bud no clue
same here bro
They provide gpu for free till now
Yo, is it possible to talk with an ai voice and then immediately switch to your voice, if so, how do you do it?
is there a easy way to install chatterbox?
Are you a femboy
nope
Sucks to be you then
i just want to catfish my friend
Real
do you know how to download it?
Yes however I only help femboys
Hello, does anyone know the free alternative to "a0.dev" app development ai website
Without gpu/cpu?
Impossible, unless you use cloud (remote good pc), pls use #1192011222023950368 tho
dunno but u can just use any chatbot like gpt5 or gemini from aistudio to develop code
are you trying to catfish your friends too?
try checking https://github.com/resemble-ai/chatterbox?tab=readme-ov-file#installation if u got a good pc
they are freemium services
original wokada window is just an "embedded" browser too, just in it's own window
That is illegal. You have been warned
All AI ARE GAY 😫🪳👍
🤯
if you got git and python 3.10/3.11 installed, yes
git clone https://github.com/resemble-ai/chatterbox
cd chatterbox
python -m venv venv
venv\scripts\activate
uv pip install .
uv pip install torch torchaudio --upgrade --index-url https://download.pytorch.org/whl/cu128
uv pip install gradio
gradio_tts_app.py```
from a command prompt
PIP ?
Hey, like 6 months ago I used W-okada Voice changer with some models from this discord.
I got a new gpu (4090) and wanted to ask if realtime translation is possible now? Or with the 5090?
Because in the past I used the 4080 super and the voice changer worked well, but it took like 3 seconds after I finished my sentence for it to get said, which just is to long and makes conversations impossible.
So my question is, any Tipps or recommendations so I can achieve real time voice transaction locally? Or at least <1 sec processing time for conversations to be possible?
this isn't the right channel, please elaborate also the tutorial you're using and the other info in #1192011222023950368

I wonder how many times you say this in a day
cause i know for sure its daily
more times than the "we got <word> before gta 6" jokes 😭
😭
i cant upload my voice model ?
I believe I stopped my training too early, is there a way to resume training even after it hits the full amount of epochs I trained for? Without having to retrain the whole thing ofc
just start the training again (skip preprocess/extract features), same SR and batch size, more max epoch
so basically keep everything the same but just add more epochs
doing this won’t restart the training right?
cuz it’s been like 8 hours of training
it will resume from the last saved d/g
bet, thank you🙏
Can anybody tell me the free alternative to a0.dev ai@chilly lake @spark agate
should it be using my cpu more than gpu?
use your brain 😛
It would be best to elaborate what you're talking about in #1192011222023950368 :D
@glad nebula I tested it out and it worked!! Since my dataset is 44k, instead of training at 40, I trained at 48, and my model is indeed 44k
the difference is noticeable
thanks
I need some assistance. I'm wondering if anyone has found a AMD Stable Diffusion but for videos?
comdyui-zluda, sd.next
any ai geniuses wanna message me need help with few gens
for help u can nuse #1192011222023950368
who can i downlod the voice changer i had an old Version
Hey im running an Ai benchmark i made, So far GPT-5 high is the highest and scores a 42%, OPus 4.1 is 36%, does anyone here have Grok-4 Heavy or Gemini Deep think or GPT-5 pro, i would love to test those but dont have access. the benchmark is only 10 questions so it wont destory your limits
elaborate for help in #1192011222023950368
ohh you're trans? for help you can ask in the right help channels, not here though
Maybe someone in #1192011222023950368 could help
yw
thanks i posted their
@lone wren no advertising
On my rx580 it is shq icl
elaborate your help request in #1192011222023950368 and reminder to not use video tutorials for rvc nor wokada
Hi
has anyone ever tested multiple microphones with rvc to see if and which gives a better result?
you want a microphone that is clean without too much noise in the background
if you got two mics and can't decide which is best do this:
record yourself with each microphone
downsample both audios to 16k (this is the sample rate rvc uses for inference)
the winner is the one that is the most clean at that sample rate
well im thinking on buying a mic just for it, got any recommendation?
or any specs i should look for in a mic
no idea about microphones here... sorry
if the microphones have people testing them, record their audios, downsample that audio to 16k and see if what the person is saying is clear enough
ok thx

Does anyone have a guide on how i can use/train ai voice changer models and where i can find them
what's your point of doing benchmark? note that it doesnt reflect the actual use cases. many ppl found GPT-5 results might be disappointing esp in coding.
any decent headset mic under 100 bucks
@graceful flame stop sending the same thing over and over again
TalkT2-0.1b
I just made a 0.1b peramiter human like chat bot with responces like :
You: If you could change one thing about the world what would it be and why?
TalkT2: that's a good question, but I don't know yet how much of your mind is free.
And the ability to adapt and change :
TalkT2: but I do know that it's a waste of time to write an algorithm for every cell
You: well if it works dont mess with it
TalkT2: and all cells will be happy forever
You: exactly
This model is only 500mb ind is 1000+ times smaller than chatgpt and outher models an llm that can think for itself and have opinion and a personality like that for its size is crazy
For anyone wondering the model like is
https://huggingface.co/Notbobjoe/TalkT2-0.1b
Feel free to check it out and try it or fine tune it there are more exsample chats ,bonus chats and amazing responces on the data card
dude learned to emulate Eliza with 100x of its size
@worthy wraith don't beg for people to make models here please, you can make a free request in #1159289738314919936 or make it yourself with https://docs.aihub.gg (or search if it already exists in https://weights.com/models )
how long would it take for someone to make one for me
training models can take time, also it is a free request, no one is obbligated to make you one, it might be someone in 1 week, in 1 day, or never
and paid requests aren't allowed
i would suggest u look if there's already a model first
no i didnt see
it does take time, but all you need is to just read https://docs.aihub.gg
Last update: August 5, 2025
is it easy? yeah if you read carefully the guides
is it 1 click? nope, RVC (Retrieval-based-Voice-Conversion), unless you train with https://weights.com/models
You can search rvc ai voice models at:
- https://discord.com/channels/1159260121998827560/1175430844685484042
- In https://discord.com/channels/1159260121998827560/1163592055830880266 , Do /find with @hidden grotto
- https://weights.com/ (login required)
- https://huggingface.co/models (but watch out cus in hugging face there arent only rvc ai voice models)
- https://voice-models.com/ (just scrapes the huggingface link from the voice-models channel)
- https://thevoicemodels.com/ (for Turkish Models, login required with discord and level 2 on their server)
if there isnt one, you can:
- make it yourself with our docs guides
- Ask a free request in https://discord.com/channels/1159260121998827560/1159289738314919936
- Be aware that we don't allow any paid comms in the server
:wave: @covert lake, How can I help?
Available Commands:
• @weights find <query> or /find <query> - Search for RVC Voice Models
yw
ok it works just fine
Can anybody tell me the 100% free to do online certifications which have high value for ai ml job which are offered by top companies or institutions
One high quality and high priority certification won't always be free, it sometimes costly to get one. Your portfolio does not always depend on your certificates, but it depends on your known skills, the education, years of experience in specific field(s), which company you worked before, and your best known projects you been doing these years. Without any of these, the top-tier AI company you're expecting won't gonna hire you like that.
Who on earth coding for 24 hours straight? I usually do code not for so long since I mostly debug some minor bugs in a program.
no like creating a program
how long is it suppose to take
feels like am shoving water out of a sinking ship
i feel like that's too vague
varies alot
i mean, a refined and optimized program is better than a rushed one
wait why piper? it's low quality
(also integrating LM studio with it)
I need a help
what do you need help with ?
idk anything better
i heard about chatterbox and fishspeech from Noobies
So basically it's about #🤖│bots how do I get the server rank??
Like server leaderboard i did ! Rank command I got mine but i wanna see the server one
i mean u could also just use like 11labs api keys
!leaderboard
- @covert lake ~
144- @hollow tangle ~
109- @ionic pumice ~
81- @heavy viper ~
79- @hidden pasture ~
61- @storm meadow ~
60- @frank quarry ~
52- @dawn temple ~
52- @solar torrent ~
50- @stark scarab ~
48
Thank you so much
well.. i have to check if that is compatible with directml (because its easier to use that since my gpu is from amd)
Hmmmm
the goal of the project is to be as Local as possible
maybe that will justify why it took me 5 days lel
oops acidentally did it here, but yeah u get it now
ehh not sure, i got nvidia
biggest mistake was not buying nvidia
Yesssss 
Is this chat ai generated
maybe u could also try patching with zluda but i dunno much about amd
Ur ai
we unfortunately are real human beings :D
Damn
Why you sayin unfortunately
You wanna be ai?
AI doesn't gotta pay taxes 😔 /j
Damnn fr 😔
If you say so. 
Ur definitely ai
i didn't think of that...
@pine bobcat for showcasing your coding projects, you can use #1359898289335566570 instead :D
okay @covert lake , I didn't see that earlier.
guys, I created this AI learning platform web app -- https://discord.com/channels/1159260121998827560/1405585233050599494 with the help of google AI Studio. It is free and open source project. This is my first time deploying online. I am just asking for the reviews from you. 🙏
yo
@elder willow
promoting isn't allowed. stop spamming it.
@elder willow your promotions got deleted in every channel you sent them
Hey guys, can anyone tell me how to get RVC on Mac
I told you in #✦│chat message to please use #1192011222023950368 :D It would help both helpers and you
guys I need help with rvc
for some reason, when I turn it on while I'm in game
It's so pitched and sound dirty
and when I turn on RVC without a game, the sound is great
@honest folio @hollow juniper both of you, please elaborate in a post in #1192011222023950368
rvc means Retrieval-based-Voice-Conversion, not realtime voice changer
Helloo
Hello sensei, hru?
Good, wbu?
Nothing, just exploring the server for now
I'll let you know, thanks.
Ohh nice, I hope you enjoy staying here
im new, is this a server just to talk about AI?
alright ima leave
have a good day
Yes, this is a general AI Server
nope, it's not anymore
@zenith nova We are a general ai server since some months, we aren't an RVC server and won't be anymore, please don't call it an RVC server as they might leave 😭
lol
deep eboy voice model when
@sick mica hello man
are u looking for e boy trolling ?
i dunno if they are ur friends so pinged u
but you get embed perms at level 5 by chatting btw
I replied in #1192011222023950368 btw
waiting for the weekend lol how about you?
what covers ai gui is the best?
it is why would it be not safe?
its like chatgpt is unsafe ask reddit is this safe or not
its ai what can i do
nothing at all ngl
i used deepseek and the next day i lost a job i didnt have and , a wife i didn't have
i used deepseek and i gained an e girl 
I'm asking to understand as there are different programs and video tuts are old
dont use video tutorials, elaborate in #1192011222023950368
thhe website version does have weird tos, but you can use it locally
if you got a good pc
but anyways every chatbot does this
just use it locally if you got to use it with private info
@primal rampartYOU'RE IN THIS SERVER TOO
:kekw:
NO ONE ASKED FOR YOUR RESUME
SCAMMER
@thorn yarrow go find a job somewhere else, this isnt the server to do that
those kind of messages are scams
I just stumbled upon this cool doc about CoreWeave's Kubernetes Service (CKS). Anyone here played around with it? It's built for high-performance stuff like AI training and HPC. Link: https://docs.coreweave.com/docs/products/cks
CoreWeave Cloud Platform -https://devopsengineers.slack.com/archives/C04TZ8JH86N/p1755221031710529
does anyone have a problem with voice changer
i am on amd gpu
and it uses only my cpu
guys can anyone help me since i cant use the voice changer and i only have a intel gpu in my laptop
what’s the best models that are good quality
For W-Okada DirectML, go to #✨│ai-help or #1192011222023950368 and explain about your issues on using the program with AMD Radeon GPU there.
i don’t wanna be that type of guy that makes the 1000th post on this subject
does the app even work on amd gpus on first place
i don’t have the time at this moment to check that’s what i meant to say
😔
No. I do this to avoid filling this chat with help-related topics.
yea i get it
Otherwise, I can't do much about it.
Intel integrated GPU? An integrated GPU is not suitable for AI. For W-Okada, go to #✨│ai-help or #1192011222023950368 and explain about your issues and intention there.
What model? While there are some RVC voice models that are known to be in high quality in #1175430844685484042, most of which are fanmade and not official so you'd have to click one and listen its preview. Same goes to Stable Diffusion models on Civitai.
What do you use W-Okada realtime voice changer for? Trolling your friends or catfishing someone for your own benefit?

Cool. For W-Okada, go to #✨│ai-help or #1192011222023950368 and explain about your issues there.
yoo
ll
We don't want scammers here
how are they scamming
@tranquil bane not the server for this
i lowk forgot how to make models
Anybody got White Boy model/voice?
don't buy amd for this stuff guys
Check https://docs.aihub.gg/
Last update: August 5, 2025
paid requests aren't allowed, but you can have free one by posting in https://discord.com/channels/1159260121998827560/1159289738314919936
Find one in #1175430844685484042 or use Weights bot in #🔍│find-models.
e-girling is banned
cuz we want to be above the board server
u gonna get banned in 3... 2...
sounds more like you want to be so
whos jack
someone who kept advertising
You have been warned for catfishing and won't get any help. Continuing to ask will lead to a ban
Just enjoy your life outside your screen, I guess. There's nothing you can do about it. 
try other models or make them yourself
I downloaded 5 models, and it's clear that they are all voice changers
for help, you can elaborate in #1192011222023950368, along with ur settings
also 5 models are few, there's thousands of RVC models tbh
@grave drift promos aint allowed
what batch size should i go for my data set is circa 50 minutes
and is mainline even good to make authentic models?
for help it's better to use #1192011222023950368 or #✨│ai-help
oh mb
can u still help me?
whats todays wordle word
what does rvc quality mean? like it makes voice processing better? (it is in advanced setting)
Not sure what ur talking about, but please elaborate in #1192011222023950368
lil different
where do i use the model when i downloaded it
It depends a lot, elaborate more in #1192011222023950368 pls
This server is English only, you can ask for help in #1192011222023950368 :D
does somone have the Gleggmire vrc or somthin like that??????
yea, check https://docs.aihub.gg
Last update: August 5, 2025
you mean RVC STS model? check #1175430844685484042 or https://weights.com/models
Y'all got any AI servers that focus on AI arguments
-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
@viral thicket Hiya, sent you a quick question in DMs, can you check it out when you get a sec? Thanks in advance!
Hi, I Am Sonic2022_mario And This Is My New Account Since My Old Account Got Hacked
Bing Speech offline
Unban my main discord account "CrimsonZockt". Bot hacked me! i swear i do something wrong?
clear cookies on your browser/device used to log into your account then relog your account
after that you can file an appeal to @serene sluice
I wanted to cry because my main account didn't do anything because it's spamming.
theres nothing this server can do about that, its your responsibility to keep your account safe
I couldn't invite with my main account.
They trying to ban me.
I have no idea what is going on?
hacked accounts get banned thats correct
if you recover the account AND enable 2fa, appeal the ban through @serene sluice
My main account got unhacked when i'm going to delete spam post.
Nooooooooooo!
I wanted to delete my main account.
CrimsonZockt is no more.
They don't want unban my main account.
youve been told TWICE to appeal the ban after youve recovered your account, if you choose not to read then thats on you.
mf asking bout sum wordle word bru 😭
Not on me, My main account.
then you can appeal on your main account???
I forgot.
Hello everyone
@grave grotto no advertising
@regal wolf please read my suggestion
forgot to add that you should try change the password and enable 2FA
My main account got ban
why not just delete your main account if you haven't tried my suggestions above? 
ai is gonna kill the internet: change my mind!
ok
if you're a passionate AI artist DM me, i want you to check out something Ive built
if you're not a power user dont DM
hey can somebody help me configure voice changer?
Ye same problem what voice changer can i use as a best and free
For uh, go to #✨│ai-help or #1192011222023950368 and explain about your issues and intention there.
Can someone helppp
Guys, bing speech platform still offline
What help?
There is something that gives like a boost for the voice changer
30 hours a week
Ik it's not a boost
I use others ppl pc it's remote I think
Can I have it's dawnloade link
For W-Okada, go to #✨│ai-help or #1192011222023950368 and I'll tell you where the Kaggle notebook link is.
K
Well, go for it.
r u tryin to steal my voice!!?!?!?!
sudy stem field, i want to know which ai is the best for reasoning, explaining, understanding stem field(coding, maths, engineering), ....
chat gpt 5 maybe?
maybe grok 4, gemini, deepseek, ....?
a friend recomended me github copilot, .....
should i buy chat gpt 5 plus for 20 dollars?
do't bother buying anything until you've been using it fo ra while and know "this is the one for me"
they all can be used for freee with reasonable quotas . when you find one you keep coming back to and you keep getting rate limited because of that then consider buying that
do you have any recommendations, which ones to try, ....
which ones are good, any opinions?
i use Chad Jibiti
honestly i would say all of them are good, it's really more a matter of finding one that jives with you, and learning how to ask better questions that align with what you're looking for
remember
LLM are NOT actually "Intelligent" and they're technically not what people think of when they think of "artificial intelligence'
they are sycophantic auto suggest on steroids
The correct name is ChatGPT.
which means bad input in, can result in bad output
My friend personallypreferrs to be called Chad, and on official papers Chad Jibiti
wsp gng
guys i have built some small projects on ai/machine learning and want to share its video with you
but i dont see any option to upload video or screenshot here.
how can i share video or screenshot of my projects here?
Please someone help
I need to make an edit
Using a reference video i have
Like i want the same thing but with my own clips
Its urgent
Can someone please help by telling about any ai tool
You have to be level 10 in the server to upload files.
oh i see
so how can i reach that level in this server?
You basically just chat in the server and you will eventually get enough XP to reach level 10.
ok. got it. thank you
No problem.
@umbral aspen it was reduced to level 5 btw
Oh, my apologies.
best opensource music generator?
I think im a bit late….but idk https://colab.research.google.com/github/Nick088Official/FaceFusion-Online-Ports/blob/main/FaceFusion_UI.ipynb
elaborate in #1192011222023950368
thx
help the rvc deiteris thing stopped working for some reason
elaborate in #1192011222023950368 pls
i put it there
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.
hey is there any free tools in which you can change certian thing in an image ? like changing the color of clothes by circling it changing hair etc ?
yall im like deadass so new to all this LET ME KNOW WHAT IS GOING ON ABOIUT EVERYTHING PLEASE.
Sup
Hi guys ,
Excited to be here
how do i fix white screen in the voice changer
Can someone help
Can u help

I use Windows 11. Although some say Windows 10 is better than 11 in performance.
Windows 10 is bugging my mic and stuff bru
like it has issues idk if it's my PC or
Cus my earbuds mic suddenly stop working and then my other normal mic turns off and stuff
Windows 10 gotta be out of support soon in October 14th.
i used the audio cable but i cant hear myself on discord
For W-Okada realtime voice changer, ask in #✨│ai-help or #1192011222023950368.
same i went in a vc and other people can’t hear me
hi
i feel like im gonna come across as childish and a newbie(even though i kind of am a child and a newbie to LLMs and APIs, but not agentic AI) but whats the closest thing we have to jarvis available for public access?
Jarvis eh... That's probably what Meta are working on with Rayban
hmm yes the meta glasses by rayban
what are your thoughts on it? im not very experienced on the topic apart from the fact that it uses llama and can process visual and verbal inputs
Glasses from Top Charoen. 🔥🔥
unable to learn about this, can you elaborate 😅
another dum dum
elaborate in #1192011222023950368 and be sure to not use video tuts
it would be better u askin #1192011222023950368
elaborate in #1192011222023950368 pls
if u elaborate in #1192011222023950368 , yeah
rvc doesn't mean realtime voice changer, and for issues u should use #1192011222023950368
yo anyone i downloaded vbcable and i cant hear yt myself and only hear people on discord what do i do
i can even talk
the voice changer is just a white screen for me
ping helpers in #✨│ai-help
who have Serbian voice model pls
hello im trying to do mark grayson (comic character) voice and i dont know how i do this rvc setting my voice is so crepy with mark grayson model
can somebody help me?
türk yetkili varsa yardımcı olabilirmi türkçe sohbet kanalı bulamadım buraya yazıyorum
the server is english only
Hey folks 👋 I’ve been working on a project that helps people use ChatGPT as a thinking partner instead of following random guru advice. It’s text based, simple, and I am looking for a few people to try it out. No pressure, just want honest feedback , dm me if you are open 🙂
text based as in console?
shouldnt you ne using an ai to test it ?
lol
kinda seems like a waste of time
h
j
a
has anyone tried suno 4.5
im watching a tutorial how to create a ai voice and i stocked where i need to find the member list, and is not appearing.
dontn watch video tutorials, use #1192011222023950368 for help
I need a child’s voice (around 4–5 years old) for my homework. Can anyone help?
Asking for childs models isn't allowed. You have been warned both verbally and via sapphire and continuing to ask will lead to further actions possibly like a ban
I understand, thank you for letting me know. I was referring to voice acting only, not real child models. I’ll make sure to follow the rules
Nick was actually referring to child voice models.
can sum help me figure out how 2 download w-okada
does anyone know a somewhere to train a text model for free
u mean real time or not
it depends, elaborate in #1192011222023950368
elaborate in #1192011222023950368
we don't allow any promotions, as an example for paid services, here
worth checking out uwu: #1406752762179485719 message
Insert that one ratatouille gif
You know the one
Also how's rvc on an nvidia titan xp
yo, i tried to create an rvc model but it didnt save at weights folder what do i do
for help use #1192011222023950368
are u talking about the voice changer voice.ai? no, no one suggests that here, don't use it
and for help use #1192011222023950368
hi what is the best female voice model
None, check the #1175430844685484042 channels and test which one fits you the most
are we deadass so ur sayin i should go through all the models
are u trying to do e girl trolling?
Do epochs make a difference? Im training an rvc model
what are rvc settings for audios that get loud
epoch is a single cycle of training when the model passes thru the dataset
What? Not thru all the models.
Can someone tell me how to resume training on applio
go to training, select the model and sample rate as before, scroll down to training, select batch size as before, add more max epoch, click train
Ahhhh tysm I was about to loose it bc i was training a model for 2 days
E girl trolling is catfishing, which is illegal. You have been warned both verbally and via sapphire and continuing to ask will lead to further actions possibly like a ban

