🧬│ai-chat

[harsh flame]

💀

[gray rover]

Well, I can't give you any recommendations as I use my own models and not anybody else's

sorry

Best if you just go and find something you like

[slate loom]

How do you install pretrains

[chilly lake]

download it

[elder willow]

download it

[slate loom]

Do I download it

[gray rover]

do u

[solar torrent]

These pre-trained models should be pre-installed in every RVC forks. But if you're looking for custom pre-trained models, you can find some on Hugging Face.

[gray rover]

[split whale]

hmm, should I sleep?

[glacial basin]

Hi everyone

Does anyone can advice me with the best AI video engine

Like this quality

https://www.instagram.com/reel/DAK_DuftrUi/?igsh=Y3hodjk1NGk4dDhj

[forest crypt]

maybe you can listen to music now and fall asleep in the process.

[split whale]

https://cdn.discordapp.com/emojis/244887972414095361.webp?size=48&name=thonkred

[gray rover]

sleep for mortals

join the undersleeping legion

[chrome girder]

imagine sleeping.
This post was made by 1000mg of caffeine gang

[gray rover]

pfff only 1000mg

not enough

[split whale]

Sleep is for the weak. 👍👍👍

[solar torrent]

AM or PM?

[flint crystal]

hi

[halcyon venture]

how to download w-okada

[solar torrent]

-realtime

[rare sorrel]

Interaction has expired, use the command again for a new interaction.

💻 Local Realtime RVC

• Guide for deiteris' modified W-Okada's Voice Changer Client (Fork)
• Guide for W-Okada's RealTimeVoiceChangerClient
• How to use real time conversion with RVC (voice changer)
• RVC TTS

[solar torrent]

If you got a super fast GPU in your PC, pick the first one for the forked version of W-Okada.

[dense terrace]

wb for lower performing gpus?

[solar torrent]

Then what is your GPU? Nvidia GeForce GTX 1000 series and higher should be able to run and inference W-Okata like ok.

[rare sorrel]

Inference in machine learning is when a trained model uses what it has learned from data to make predictions or decisions on new, unseen information. In the context of RVC, it's when you use a voice model to produce an output audio.

[dense terrace]

nvidia geforce gtx 660 😭

[feral halo]

Ayo? @dense terrace level 1 !!!

[solar torrent]

It seems like your GPU is too low for that. Running on cloud is recommended for those who don't have that fast GPU.

[dense terrace]

mm, so which link do i click?

[solar torrent]

Links for cloud are there in this section of this guide. https://rentry.co/VoiceChangerGuide#online-hosted-alternatives

[halcyon venture]

bro

wth, why my w-okada opened in google?

[solar torrent]

What Google? Google Chrome or Colab?

[halcyon venture]

brave

[solar torrent]

What's even Brave?

[halcyon venture]

browser

[solar torrent]

If it's a web browser, then your W-Okada was supposed to open its GUI on your browser.

[halcyon venture]

Well, now it's different

How do I set it up?

[solar torrent]

I'm not sure what you mean by set up. So play around with W-Okada and you'll found out.

[gray rover]

hold on, gonna post him something helpful

not all of these are exactly present / named like that in w-okada

[halcyon venture]

ok

[gray rover]

but should give you a general view, Rystery

[solar torrent]

A W-Okado 🥑

[halcyon venture]

can anyone help me to configure?

[feral halo]

Ayo? @halcyon venture level 3 !!!

[solar torrent]

Apopolio 🍏

[solar torrent]

Go there at #🔍│help-w-okada

[cunning hare]

helo

[solar torrent]

Hi.

[elder willow]

explain to me how not to hear myself in the application? if I set the monitor to none it does not help

[feral halo]

Ayo? @elder willow level 1 !!!

[dense terrace]

@solar torrent what to do after installing it? just has a file named mmvcserversio

[foggy flare]

다들 안녕하세요~~ hi~ nice to meet you

[dense terrace]

in that file is a lot of other files

[solar torrent]

Double click a file start_http.bat .

[dense terrace]

then extract?

[solar torrent]

What? No. I meant the file inside the zip, after you unzipped the file.

[azure fractal]

hi, no one was answering on my help-rvc channel, just have a quick question.
I train to epoch 500, now at 192, I want to safely quit and also how do I find my .pth and index file as they're not in the logs > "name" file

[gray rover]

they must be

If they're not in there, it means you didn't check required boxes in the ui

should be rvc folder/logs/model_name/<here>

did you make sure to

[azure fractal]

was the required boxes in the ui "Save a small final model to the 'weights' folder at each save point"?

[gray rover]

ye, ones from ss

[azure fractal]

ah gosh darn

[gray rover]

hold on tho

this is not perfect at all but, if you still wanna save anything out of it

you can extract a small model out of G/D files

Sure, there's 0 control on which 'epoch' you get, as you get the latest, but you can still use it if you want 👀

[azure fractal]

ohh do tell me please

[gray rover]

just to confirm, you do have G/D files present in the model's folder?

[azure fractal]

yes correct

[gray rover]

actually my bad, just G is needed. but anyway, you use applio or rvc?

[azure fractal]

rvc

[gray rover]

alright good. in that case, go into model utils or however it's called in there

will look like:

[azure fractal]

yes

ckpt processing is called on mine

[gray rover]

in there you just point the path to the G file, pick the version you trained, sample rate, whether you had f0 on or off

and just, ye, extract

model information box you can leave empty, just the actual name's important

[azure fractal]

ah and it should give me the index file and .pth file?

[gray rover]

.pth is just an extension ( pytorch )

you'll get a weight file ( a proper model to use )

as for index, you can still generate it. There's no requirement to do it prior to training ( it's just a habit we all have

[azure fractal]

oh ok I'll see if I do it correctly

[gray rover]

and how? you wanna use the training tab. Input the model's name as you did during training
and just press train index, that's all

[vital snow]

Hi

[gray rover]

welcome

[digital anchor]

Get Top-Quality Video Editing🎬✨

Hey! 👋

Need the best video editor? Higher experiance.
My DMs are open—just drop a quick "Hi" to get started.

DM for my Portfolio

[solar torrent]

#1159290752195633273

[remote vapor]

Guys how can I start creating a videos like ''SimpliHow'' channel in YouTube? I've tried to find something but...

[solar torrent]

https://cdn.discordapp.com/attachments/875501266640457812/1308682665331003452/RDT_20241120_0936234346838477305286044.jpg

[tawny night]

how do I unlock the rvc channel?

[solar torrent]

What RVC channel?

[hollow hornet]

hi

[random bridge]

-colab

[rare sorrel]

📒 Google Colab Notebooks

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], by Eddy, Hina and Gdr Google Colab
• UVR5 NO UI, by Eddy Google Colab
• UVR5 UI, by Eddy Google Colab
• Modified W-Okada's Voice Changer, Google Colab
• 🆕 FaceFusion UI, by Nick088 Google Colab
• 🆕 FaceFusion NO UI, by Nick088 Google Colab
• 🆕 EasyGUI, by Rejekts Google Colab

ℹ️ Note

While the Colab free plan provides up to 12 hours of daily usage, the GPU is typically available for only about 4 hours each day on average.

More commands: -audio, -hf, -local, -kaggle, -realtime, -rvc, /help

[dusky mortar]

e

[cosmic verge]

I am seeking a job and I'm from Chicago, US. Are there any opportunities available for me? I am an American.

[covert lake]

Use #1159290752195633273

thus server isn’t really made for seeking a job tbh

[merry stag]

Hey!👋

[tiny quarry]

Hello is there any upscaler for video

[covert lake]

yes

either for paid, topaz ai

or for free, open source: https://discord.com/channels/1159260121998827560/1221468455723733094

[tiny quarry]

thanks a lot

[pseudo willow]

hey

[finite basalt]

https://garticphone.com/ru/?c=02b1564986

[rare cape]

hello the voice changer works but the delay is insane and is really laggy how do I fix that?

[vital snow]

whats ur gpu? and send screenshot of ur settings in my dm

[autumn pelican]

Chat instead of become author can i just pay artist to draw comic as i like. You i describe how character look behave like. How the story should go. So describe everything to the artist to make comic. Can i do that is there people have do that. Is there term for people that do this

[solar torrent]

If you're looking for an artist that doesn't use AI on their work, like me, then sure. I like to be like that.

[ionic pumice]

yee

not sure what it's called but iirc merryweather is like that

[solar torrent]

Just letting you know this server is full of AI artists and some people who don't even draw but use AI to make art, the real ones are pretty rarely to show up.

[analog cosmos]

will mostly only find AI artists here

[solar torrent]

Lazy.

[analog cosmos]

https://tenor.com/view/lazy-bunny-rabbit-win-no-gif-7597216

[analog cosmos]

whats that green stuff?

[ionic pumice]

?

[solar torrent]

Weed, I suppose.

[analog cosmos]

what do you mean

[analog cosmos]

https://tenor.com/view/snoop-snoop-dogg-snoop-smile-gif-18363169359568588385

you mean with one of the bots? if so then you probably don't have permission to use the command/app in this channel

most likely don't have permissions to use it

might be wrong tho lol

[solar torrent]

This one?

[analog cosmos]

gets deleted lol

[ionic pumice]

why would you even need to do a poll like that lol

[analog cosmos]

what are you trying to find out with that poll anyways?

like which one is the better option?

[solar torrent]

I dream last night about a pregnant woman cosplaying as Reimu Hakurei, then she posted her photos on her account for whatever reason.

[analog cosmos]

lmao

[tawny night]

thank you 🙏 for some reason I can't access that channel

[feral halo]

Ayo? @tawny night level 1 !!!

[analog cosmos]

think most people either use collab or Kaggle, not many people have a gpu that is capable of training locally

[solar torrent]

Reimu isn't my wife or waifu after all, so I'm not sure what was going on.

[covert lake]

u can’t make pools here

[solar torrent]

Pools for swimming?

[analog cosmos]

[elder willow]

Hi guys i've been using voice.ai for discord but i found it quite choppy and distorted, is there any better app?

[covert lake]

voice.ai is garbage

what’s ur pc gpu?

[solar torrent]

Voice.ai is a scam fuck.

[covert lake]

wokada, especially the fork, is better

[covert lake]

Fax my brotha, spit yo shi indeed

[solar torrent]

-realtime

[rare sorrel]

Interaction has expired, use the command again for a new interaction.

💻 Local Realtime RVC

• Guide for deiteris' modified W-Okada's Voice Changer Client (Fork)
• Guide for W-Okada's RealTimeVoiceChangerClient
• How to use real time conversion with RVC (voice changer)
• RVC TTS

[analog cosmos]

https://cdn.discordapp.com/attachments/1159289354439626772/1288171293979574292/sponge-bob-night_.gif

[covert lake]

https://tenor.com/view/fax-my-brother-spit-yo-shit-indeed-tellmeits0ver-cat-gif-18001869000166970593

[elder willow]

4080 so I'm sure it's plenty

[feral halo]

Ayo? @elder willow level 1 !!!

[covert lake]

hell yeah

-rt

[rare sorrel]

Interaction has expired, use the command again for a new interaction.

💻 Local Realtime RVC

• Guide for deiteris' modified W-Okada's Voice Changer Client (Fork)
• Guide for W-Okada's RealTimeVoiceChangerClient
• How to use real time conversion with RVC (voice changer)
• RVC TTS

[covert lake]

1st link

the wokada fork, which has better performance

[elder willow]

Oh ok cheers guys

[solar torrent]

First link is the forked version of that second link below, designed to run better.

[analog cosmos]

Kaggle with multiple accounts

[solar torrent]

https://tenor.com/view/hawk-hawk-tuah-tuah-tuuh-dream-girl-gif-16433735641958117066

[analog cosmos]

[elder willow]

so just follow everything in the first link right?

[covert lake]

yes

u got an rtx so u use the nvidia version

[solar torrent]

Eh, I only use Suno, so I can't tell which one is better.

[analog cosmos]

you tried Suno V4 yet?

[solar torrent]

Turning a minor key song into a major key song really makes the song to sound happy.

[elder willow]

thanks, I mainly use the voice changer for discord. Hopefully it works there are so many options out there 😄

[covert lake]

its the best open source one

[solar torrent]

No, I haven't yet buy pro to try that.

[covert lake]

buy pro

btw u can make songs with weights.gg too and for free

[rare cape]

alright

[solar torrent]

Too expensive for me, even if it's monthly subscription. Weights is indeed able to generate song with a copyrighted reference track, but the lyrics and length are too limited short.

[analog cosmos]

you can remaster one of your existing songs with V4 for free once

[solar torrent]

I won't do that until I get pro.

[elder willow]

Does Suno do realtime as well?

[analog cosmos]

I kinda want to get pro again, honestly enjoyed using Suno lol

[solar torrent]

Suno doesn't generate track in realtime.

I wanted to generate 2000 r&b style tracks on Suno, but instead got these tracks to sounded like some pop music nowadays.

Well, it seems like I can upload some copyrighted audios to Suno now, by either remove vocals from them or speed them up.

[solar torrent]

Tried extending these, some of them sounded not so good as original.

"Moo Deng Moo Deng" is the first Thailand's brainrot song I've ever heard in this year.

[dense basin]

yup

[analog cosmos]

$5 then I will give you prompts I found on reddit

[vital canopy]

yoo

[void nest]

Hi (I would like to sincerely apologize for any mistakes or misunderstandings that might be caused by my flawed use of the English language. I understand that my writing or speaking might not be up to the expected standard in countries such as the United States of America, United kingdom, Australia, ect. and I regret any confusion or misunderstandings that may arise because of such knowledge i have not in fact reached. English is not my first language, and while I am constantly striving to improve, I know there is always room for growth. I truly appreciate your understanding and patience as I continue to learn and refine my english language skills. Your feedback and any corrections you offer are valuable to me for as long as i still experience existence of atoms and molecules in my body, as they help me become more proficient and confident in my communication. once more, I deeply apologize for any inconvenience my knowledgeable errors may have caused, and I thank you for your continued support till this point.)

[analog cosmos]

wut

[analog cosmos]

looks like a AI generated message

[solar torrent]

Bro really used ChatGPT to type this message. 🐄

[analog cosmos]

"Your feedback and any corrections you offer are valuable to me for as long as i still experience existence of atoms and molecules in my body" real

[feral halo]

Ayo? @hollow ridge level 1 !!!

[analog cosmos]

you wipe?

based

[ionic pumice]

[tacit falcon]

what the fuck

nah bro thats the most insane shit ive heard out of this server

michael jackson gooning

disturbing even

[analog cosmos]

lmao

[tacit falcon]

joined just to say this and then leave

what a tremendously useful interlude

[ember mason]

hi

[dense basin]

yo sup

[clear pendant]

sa

[analog cosmos]

https://tenor.com/view/didnt-ask-dont-care-burbger-yum-gif-25146551

I'm actually purple

[ionic pumice]

"yay" person with yay alarm:

[covert lake]

What the fuck

[analog cosmos]

rip bozo

[dense basin]

enzo

[elder willow]

Any good female voice models?

[analog cosmos]

this dude

[covert lake]

Wtf

I already banned him

[dense basin]

Polly or eleven labs

[dense basin]

ayooo

[elder willow]

oh i mean rvc models

[analog cosmos]

absolute degen lmao

[dense basin]

voxclone pretty good

for rvc

[glossy cedar]

WAIT THESE GUYS MADE WEIGHTS??

[gray rover]

XDDD

[ionic pumice]

lol what

[mild bramble]

Got female models?

[ionic pumice]

look for em in #1175430844685484042 or weights.gg

[waxen dawn]

hey i'm osvaldo

[vital snow]

hi

[nocturne veldt]

hi

[solar torrent]

https://cdn.discordapp.com/attachments/943167241422245950/1020982108371554355/Koishi_Fumo_dance.mp4

[silent spindle]

yo can someone guide me on generating AI voice covers, im new here !!

[empty goblet]

can someone give me a tutorial on the voice changer

[vital snow]

what is your graphic card?

[silent spindle]

i dont have one

i dont know about that

?

[vital snow]

i do train locally

[vital snow]

you need : vocal isolated audio file of a guy u want to be converted
rvc model
and pick 1 of this
applio https://docs.applio.org/applio/getting-started/inference
rvcwebui https://docs.ai-hub.wtf/rvc/local/mainline/#inference-
ilaria rvc zero https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit?tab=t.0

[silent spindle]

im using colab right now

[feral halo]

Ayo? @silent spindle level 2 !!!

[vital snow]

i train with 4090

[silent spindle]

but i dont know how to use it

[empty goblet]

can someone give me a tutorial on the voice changer?

[silent spindle]

same

i need it too...

[empty goblet]

idk how to install it and all

[silent spindle]

yes

where ?

[empty goblet]

can someone give me a tutorial on the voice changer

[feral halo]

Ayo? @empty goblet level 1 !!!

[silent spindle]

send me

[boreal arch]

anyone familiar with laravel

plz, im dying

[covert lake]

Laravel?

[boreal arch]

yeahf

[feral halo]

Ayo? @boreal arch level 1 !!!

[covert lake]

Realtime voice changer for calls? What's ur PC gpu

[covert lake]

Yeah nvm idk about PHP

[boreal arch]

[covert lake]

I know python and html mostly

[analog cosmos]

was gonna say isn't Laravel a web app framework

[covert lake]

Skill issue

[boreal arch]

yeah for back end shizz

[analog cosmos]

only ever heard of it but never worked with it

[covert lake]

Stackoverflow may bless u

[analog cosmos]

true

[silent spindle]

@elder willow

[analog cosmos]

only web app stuff I ever touched was Angular

[feral halo]

Ayo? @elder willow level 5 !!!

[silent spindle]

ok

dont forget that i dont have a gpu

yeah i got

is it for without GPU one ?

the cloud GPU or the one that i have (I dont have any GPU)

oh

i already selected that

is there any other web that is faster than this one ?

it takes a lot of time

hugging face is good

but the problem is that after sometimes the models get deleted

im a producer

and i just want to sing

a little bit so thats why i need AI cause my voice sucks

i think there was a website which had all the models for AI covers

i got it, it was weights

i mean i can sing very well

[feral halo]

Ayo? @silent spindle level 3 !!!

[silent spindle]

but im not confident about my voice

[empty goblet]

yes, 4050

[silent spindle]

you a producer ? @elder willow

😂

i thought

why my voice models get deleted after sometimes in hugging face

yes

[silent spindle]

halal entry

if you can

[silent spindle]

you know Bryant barnes ?

his voice is just on another level

i want his model

if you can make.....

and can you teach me how to make models ?

[solar torrent]

I know how to sing. I just don't do all the time.

[silent spindle]

same

[solar torrent]

Why is the Weights in French?

[silent spindle]

@elder willow bryant barnes voice model not found

ok

[silent spindle]

i dont have one

he uploaded his songs on YT

ummm......... Ok

thank you so much

Btw can you teach me how to make it ?

im still gonna learn it

ok

but i want to know everything as im beginner in this

oh

come

[hallow plank]

https://google.com

[analog cosmos]

[ionic pumice]

lmao

[gray rover]

🤔

[polar flax]

https://tenor.com/view/博衣こより-ホロライブ-hakui-koyori-hololive-holo-x-gif-26608644

[gray rover]

https://tenor.com/view/gurugurugg-gif-4353715209279177183

360 360

[polar flax]

https://tenor.com/view/mumei-nanashi-mumei-hololive-spin-spinning-gif-18055145504789513476

[potent summit]

Good morning all, would anyone with experience in doing AI voice model integrations be down to chat? I'm looking to build out an iOS app that will be a companion app for kiddos. I'm particularly interested in the integration when it comes to the voice models themselves - aka how to get the voice from the user, send it to the LLM and then have it returned as voice to the user (so a conversation can happen). I looked at a couple diff ways but I'm curious if someone here has expertise they can share.

Thank you guys 🙏

OpenAI Realtime AI looks interesting but my concern is the cost, apparently its almost a dollar a minute .__.

[covert lake]

Kuru Kuru

[sullen cove]

lol

[full tusk]

gathering files for an obscure singer from the 60s and only found 5 studio recordings and 1 usable recording from a film 😭

[feral halo]

Ayo? @full tusk level 1 !!!

[gray rover]

total length of 'usable dataset' ?

[full tusk]

just splitting the last recording then i can say for sure

[gray rover]

tho you can give me an approx

[full tusk]

one of the recordings was 30 seconds from a random swiss website 😭

[full tusk]

i like specifics though

[gray rover]

Also, I have to say right away that, in case it's not uhm, uniform
then that'll be a problem

[full tusk]

but i can approximate with the length i have with the regular studio recordings

[full tusk]

no it's a studio recording

[gray rover]

I mean it more in an aspect where you'd have different sourcing / acoustics, maybe different mics, different mouth-mic proximity etc

[full tusk]

i mean i don't intend to use it for proper covers just to joke with my friends

i made a demo model with 2 songs from around the same vocal era and it sounded ok just crunchy

[gray rover]

Aaaa, in that case it's alright

[full tusk]

cause in the 90s her voice got rlly deep so that excludes a few songs from being usable

[gray rover]

Sometimes you can help it a bit with some EQ matching and a lil bit of dynamics play

[full tusk]

maybe

[gray rover]

tho overal, as for jokes-purpose

3-5 mins should suffice heck sometimes 1-2 mins if you're careful

[full tusk]

15 minutes

so it should be ok

[gray rover]

Definitely okay, that's already a territory of 'mid-range' dataset

[full tusk]

15 minutes of full recordings, 30 seconds of the vinyl snippet

[gray rover]

I'd discard the vinyl personally or use it for 2nd variation of the model
If it's too divergent it might cause too much harm 👀

[full tusk]

no it's from the same era

it's just slightly higher quality than the others

cause there's almost no sources for her music that aren't youtube 💔

[ionic pumice]

oh hey

[gray rover]

I meant the potential vinyl crackles

for instance, if studio recordings don't have that quality to it, you wouldn't want to mix it together

example of what could happen sometimes is biasing towards a certain " quality " or aspect of the dataset

You can still try nevertheless but I just wouldn't advise it : L

[polar flax]

consistency is more preferrable

[gray rover]

Basically biasing could occur ( overfitting to a specific part too )
or divergence where generator doesn't know what to follow closely as " once in a while it encounters a completely different in spectrum audio " and gets thrown off the track

ish

[azure fractal]

for training, is harvest still best
or rmvpe_you better?

[feral halo]

Ayo? @azure fractal level 3 !!!

[gray rover]

Harvest, dio and pm are long time dead

[azure fractal]

ah thanks

[gray rover]

RMVPE is your best choice
unless having absolutely clean audio, then crepe is really amazing

[azure fractal]

training my own voice exciting

[gray rover]

neat

[worldly hound]

hey gang
xenu here...
i work as a product and community manager for an ai agent startup in SF
looking to expand my network and meet people

[azure fractal]

hire me

[full tusk]

i made sure to check it thoroughly for crackles

i'm not too focused on having a perfect model anyway

i finished training it a moment ago so i'm just inferencing it to test right now!!

[gray rover]

I mean ye, but was saying just in case
potentially degrading a model by 15 or 25% for no benefits is just impractical

But anyways, hope it came out well!

[full tusk]

we'll see!! i can't upload files here so i'll dm u when i've inferenced

[polar flax]

wonder why you want that small amount of vinyl session so bad?
remember, compromise quantity before consistency

[full tusk]

just for funsies

[feral halo]

Ayo? @full tusk level 2 !!!

[full tusk]

plus i'm collecting different files on her so i'm killing two birds with one stone by having it there

it might be acetate tape rather than vinyl i'm not 100% sure

[azure fractal]

Is there an AI that fixes static noise from old tapes/vinyl audio?

I used UVR and it extract the vocal along with the static, the instrumental seem fine

[full tusk]

i think i remember using a tool like than on mvsep for psola files, i don't remember the model but u can check the list

[elder willow]

A

[misty swift]

any solid female english ais?

[feral halo]

Ayo? @misty swift level 1 !!!

[elder willow]

i need a perfect natural female english voice model for rvc , can anyone help ?

[gray rover]

#🔍│find-models

Cause the truth is, majority of us doesn't memorize models so, pinpointing an exact one might be hard

best to just go through and find something nice

or actually, #1175430844685484042

[elder willow]

i saw it mate but they are singing ones and different languages and a bit complicated as i am new (sry for that)

and also , i have 1080ti and the voice that is rendered isnt so nice , will it work somoothly on my pc or i need to upgrade?

[gray rover]

Your best bet ^

[elder willow]

should i search the "szukaj" too ?

[gray rover]

no, that's " search " in my language

[elder willow]

oh mb

[elder willow]

can you answer this buddy ?

[gray rover]

Anyway, Hardware doesn't quite translate to performance in terms of quality

It is only in aspects of speed ( of inferencing / generating ) not really quality

If the output ( we call it inference output ) sounds bad

it's A) The model's not the best or B) The audio you use for inference must suck / is dirty ( not cleaned / processed well )

[elder willow]

and one more doubt

[gray rover]

ye?

[elder willow]

when i use a model , it slutters , the initial sound and ending sound is also blurry , when i talk in a group vc , it voice changes the sound of other people too from taking input from my pc not from mic

[gray rover]

Well, I am not a specialist in real-time rvc or w-okada so I can't help with outputs / inputs issues and whatnots
but if it's stuttering, it means your settings are not adequate for your hardware
Best to go to #🔍│help-w-okada and ask about it in there
As rvc and applio are inference ( static generation. aka covers ) and real-time rvc or w-okada are real-time conversion / voice changing
Best if you memorize it for future yuh

[elder willow]

yeah got it buddy , thanks a lof for explaining so good

i was using w okada , someone said its better than rvc

[gray rover]

One thing, rvc itself is for inferencing / covers
but people seem to mistake rvc with it's " native " real-time voice changer

[remote pollen]

guys can someone create a ai voice model? i absolutely need it for tomorrow

[gray rover]

so you should really refer to rvc's voice changer rather than plain rvc, that's a thing we point out a lot
( and people only get confused later on )

[remote pollen]

i have the voice but idk how to make it a model

for rvc

[elder willow]

so should i use w okada or the rvc

[gray rover]

from what I know lately, w-okada's fork for high performance hardware or in general w-okada ( a fork means it's a custom ish or someone else's " take " on a thing, for simplicity. Around here, it references github website's repositories )

tho I personally am a fan of rvc's native as it just gives me personally more performance

[elder willow]

what should i use , its a bit confusing

[azure fractal]

while training can I use the model interference, or will it mess up the training?

[elder willow]

i have i5 12th gen 32gb ram 1080ti

[remote pollen]

plz guys

[gray rover]

the thing that matters the most is the vram ( video ram ) of your gpu

you can check it in task manager in case you dunno

[elder willow]

its 11gb

[gray rover]

then that's plenty
You should use the fork I think

[remote pollen]

ur talking to me ?

[azure fractal]

i’m asking in general

[lethal cedar]

Yo how do I find the sovits model for travis?

[elder willow]

can i get exact link if you dont mind bud?

[gray rover]

-rt

[rare sorrel]

Interaction has expired, use the command again for a new interaction.

💻 Local Realtime RVC

• Guide for deiteris' modified W-Okada's Voice Changer Client (Fork)
• Guide for W-Okada's RealTimeVoiceChangerClient
• How to use real time conversion with RVC (voice changer)
• RVC TTS

[elder willow]

https://github.com/w-okada/voice-changer

[remote pollen]

it means you can help me?

[gray rover]

you should absolutely not
unless you have heaps of vram

[elder willow]

also i saw a person streaming like a vtuber setup

[gray rover]

to L: I suppose, you can sorta safely assume at least 1 to 3 gigs maybe? of vram used for inference but I can't give you exact numbers as I haven't checked that specifically

[elder willow]

he was sitting in a room with a pc and the pc screen has his own screen streaming

[dense crater]

Hello everyone, are there any talented AI developers available?

[azure fractal]

thanks

[gray rover]

✨

[elder willow]

what did he use?

[remote pollen]

i need to have an ai voice model of my friend just for trolling him

and idk how to create a voice model

i have some audios to create the model

[night lake]

https://docs.ai-hub.wtf/essentials/how-to-make-voice-models/ read da docs

[elder willow]

i can render your voice model if it takes less time

[lethal cedar]

I can only find RVC for travis scott. Does somebody know where to find the sovits model for him?

[feral halo]

Ayo? @lethal cedar level 1 !!!

[remote pollen]

you can make it an ai voice model?

[gray rover]

if it's not on weights.gg or #1175430844685484042 or #🔍│find-models
then you're out of luck almost for certain
That is, unless someone has a private one

[elder willow]

i am new too sry buddy

[night lake]

sovits? ew

[lethal cedar]

Alright thx for the answer

[gray rover]

Tho If I was you Rager, I wouldn't go for sovits
It's just obsolete

[night lake]

use rvc

[elder willow]

can you look at this bud?

[remote pollen]

but wdym "render your voice" ?

[feral halo]

Ayo? @remote pollen level 2 !!!

[lethal cedar]

I already used RVC and liked it but people be telling me sovits quality is better. Please tell me they are wrong lmao

[covert lake]

so vits svc is worse

[night lake]

they are quite wrong

[covert lake]

unless yall talking about gpt-sovits, but that's for TTS

[gray rover]

sovits could be for " no effort " models cause of diffusion steps in there

but if we talk about effective-effort quality, hell no

so in essence, yes, they are wrong

[lethal cedar]

Gotta dig in a little more to really understand you guys but thank you

I will definitely stay on RVC

[elder willow]

you are making covers?

[elder willow]

and also should i use onyx file or use index and pth

is index necessary and how does index work

[gray rover]

in a short, accent and phonetic handling the index

pth is just an extension for pytorch models / ckpts

[remote pollen]

what can i do now?

[gray rover]

pytorch models converted to onnx display higher performance in real-time ( sacrificing the accuracy a lil but almost nobody notices it so dw. )

[lethal cedar]

Yes but rn I'm trying to recreate the new song "MEXICO CITY"

[covert lake]

Huh?

[remote pollen]

i want to create an ai voice model but idk how i can do it

[elder willow]

oh i see

and also should i use onyx file or use index and pth

[covert lake]

what's ur pc gpu?

[remote pollen]

i have the audio for the voice though

[covert lake]

You can check your pc gpu via:
ctrl+shift+esc (task manager) -> Performance tab -> GPU
and here u should find the gpu memory

[remote pollen]

i have a rtx 2060

[elder willow]

thats so nice how you guys talking so politely , but if a person doesnt know about their pc , it will be very hard for them to learn it as this stuff is so complicated

[remote pollen]

but can you do it for me ? (if you don't mind ofc)

[elder willow]

dont worry it is easy

just download , upload your datasets and change few settings and the model will be start training

it will take some hours ( depends on pc)

[feral halo]

Ayo? @elder willow level 4 !!!

[remote pollen]

can i give you the audio file for the voice model?

i need it for tomorrow

[elder willow]

i dont have it downloaded + its 12:22 am , i will turn off my pc in 30 mins ig

[gray rover]

That's what it is
You see, it is still Artificial intelligence, people without basic knowledge shouldn't mess with AI
Else like I mentioned before, I foresee tragedies in future lol

[covert lake]

As you got a good PC, you can use RVC locally, you can choose between:

• Applio: A fork of RVC with some extra features like Applio TTS, kinda faster and simpler but same quality tho
• Mainline: The original RVC

You can do it manually following the guides

[covert lake]

don't really have the time for that sorry

[gray rover]

That being said, it is still better now compared to what it was before where gate-keeping was a thing
As it was hella tiring having to deal with people who didn't even know if they have gpu or not yet asking for advanced AI tips

[elder willow]

definately buddy you are right

[covert lake]

I mean we help with guides

but ofc its still AI

[gray rover]

^

[elder willow]

thats so hard for you

[covert lake]

following our guides shouldn't give any issues

[elder willow]

but still bro you guys are so nice

i saw devs being so rude for asking something

[gray rover]

well, you can just imagine that at one time, it took over 15 minutes to basically explain where to check for gpu in task manager
or folks despite being showed with links for guides, still yapped " how this how that "

[elder willow]

you are giving me the pain by saying that

i understand that very deeply as i have experienced that

[covert lake]

I remember showing google images of gpu task manager to make him find it

[gray rover]

ikr lol

[covert lake]

what other devs u talked with btw

[elder willow]

😭

[elder willow]

not in this server , the actual thing is , my hobby is learning new skills , i was learning about coding automation bots with selenium and requests

[night lake]

a helper once had to teach someone how to move a file or something

[elder willow]

so when i was asking them about errors they were so furious

[gray rover]

bruh

[elder willow]

huhhhhhhhhhhhh fr?????????

[night lake]

yup

it was either moving or unzipping a file

[elder willow]

thats so painful

i am just here to learn something new , like i am learning about iems too

[covert lake]

yea unfortunately there's much tech illiteracy

[elder willow]

different sound signatures etc

[covert lake]

school should teach more about PCs

[elder willow]

you should copy links of basic tasks

like how to turn on a computer and send them the link of youtube video

[covert lake]

i seen people not knowing how to zip a file

[elder willow]

btw also whats the difference between pretrain models and like the normal models we render at out pc , i saw if you need to make pretrains you need heavy gpu more than the rtx 4090 like intel a6000

[night lake]

bruh

[covert lake]

• Pretrains models: used only as a base for training voice models,it contains multiple voices and tons tons of hours to train and also a larger dataset
• normal voice model: trained using a pretrain, requires less time and fewer dataset, it's used by the users

[analog cosmos]

people need to learn how to do research about stuff on the internet, 70%-80% of problems can be solved by 1 google search and a little reading :3

[covert lake]

that's so true

people be asking shi in tiktok comments, and they just get insulted 😭

when u can just google it

[elder willow]

now chatgpt also exists

[analog cosmos]

reddit is a gold mine for very specific issues too lol

[elder willow]

so you can ask shy questions there and ask him to explain you like a 5 years old baby

Sure! Let’s make Newton’s laws of physics super simple and fun for a 5-year-old:

First Law: The Law of Staying Still or Moving
Imagine your toy car sitting on the floor. It won’t move unless you push it. And if it's rolling on a smooth floor, it will keep moving until something stops it, like a wall or your hand.
In short: Things like to keep doing what they’re already doing—staying still or moving—unless something makes them stop or go.

[elder willow]

oh i see

[covert lake]

oh ye selenium & requests

so u know Python?

[elder willow]

yes i am still learning though

i made a instagram automation bot which makes temp mail in telegram and auto fetch the otp , and makes instagram accounts for u

i use that accounts for advertisement exploit , i have methods to advertise on instagram for free

as i told my hobby is learning new skills , thats why i am here

[covert lake]

ykr it's bannable

[elder willow]

no

dm me and ill show you some ss

[elder willow]

i make accounts , add funds , advertise and throw

[covert lake]

wouldn't it be like those bots who spam dm people

[covert lake]

nah i don't really use insta

[elder willow]

no no

instagram ads the real ads

that you see on instagram bruh i dont do that shitty bot stuff

[earnest nova]

Hello everyone,

I run a web development agency focused on building websites and am exploring ways to integrate AI into my workflow. Currently, I use V0 to design each part of a website, but the process is time-consuming, and V0 doesn’t allow me to deploy the sites directly.

I’ve looked into a few AI tools, but they all seem to have their own downsides and don’t produce truly polished or professional-looking websites. Is this kind of all-in-one AI solution just not available on the market yet?

(also if i can put this in a different channel let me know) thanks guys!

[gray rover]

Hmm.. not the first time seeing msgs of this type ^
this gives me an idea that maybe there should be another help channel, a "help-dev" ? Where people can ask for help or collabs in development and so on
@river adder

Unless there already is a channel that sorta works for such and I just dunno 🤔

[river adder]

Well I think that might be a bit unnecessary

[gray rover]

Right

[cerulean root]

sum1 have download

pls

[velvet flint]

Hello, how is everyone doing?

[elder willow]

how much should i keep the index to be

[gray rover]

up to you, preference really.
I'd say 0.5 is safest balance, 0.75 or 0.8 is for high accent
0.3 is pretty neutral

just experiment and find what you like
( and if it's for real-time tho, keep it at 0

[elder willow]

real time

should i keep it 0 ?

[halcyon venture]

my w-okada isn't workig

and he's also different from before

it is opening in my browser

[alpine gust]

anyoke explain how can i change my voice

[halcyon venture]

bro, when i say in english my voice in the voice changer is perfect

but in another language is a trash

[chilly lake]

turn off index

[hollow tangle]

@river adder

this too

[river adder]

I should've deleted their messages when banning them :p

And hello uwu

[hollow tangle]

lmfao

hi litsa

[river adder]

^w^

[trim sorrel]

lol

[feral halo]

Ayo? @trim sorrel level 1 !!!

[trim sorrel]

yaaa

[halcyon venture]

why my output don't work

can anyone help me

[trim sorrel]

llk

bust

[potent summit]

anyone here have xperience with mobile apps leveraging ai voice?

[tough valley]

Hello everyone, i'm new

[nimble latch]

fuck u

[analog cosmos]

[tough valley]

thanks

[placid robin]

Omgg

Very edgy

[pastel wing]

hoi

[high harbor]

))

[twilit pond]

oi

[ionic pumice]

lmao

[hushed pine]

how to access RVC?

[urban wadi]

hey, whats the best known ai to produce realistic images? price isnt a problem

[obsidian jungle]

Do any of you do commissions? I want to start and Idk how much to charge.. Let me know or link me your kofi please!

[wind stratus]

hmm maybe flux + boring reality loRA

[thorny arch]

[vestal raptor]

my cable input is in playback and output is in recording how is that possible

[vital snow]

Hi

[royal ridge]

Hey, is there a way I can get two audio channels (e.g. the mic on pink plug and something over usb) as one?

[spare wharf]

Usually when training a model how many ephocs do you go for?

[chilly lake]

any recommendation other than 'check the tensorboard' is lame

each training is different

the system pick random slices of audios to use, so it should go long enough to pick every possible slice to get everything useful from the dataset

for 45 min it can be 50 epoch or can be 150 or 250

depends on the dataset

[orchid gorge]

hello im kiriyaj

[spare wharf]

damn mine been active for 4 hours and done 26 epochs

[chilly lake]

how big is the dataset?

[late arch]

Hello

[spare wharf]

i took 5 songs, each around 3 minutes, put them into replay and waiting now, have i missunderstood something on this whole make a ai cover thing perhaps?

[torn aspen]

is there anyone tried using getpickle?

[feral halo]

Ayo? @calm heron level 2 !!!

[pseudo eagle]

is there a mike tyson voice model somewhere?

[feral halo]

Ayo? @pseudo eagle level 1 !!!

[pseudo eagle]

i remember doing an ai cover off a site but i couldnt download it or forgot to and i cant seem to find it anywhere

[azure fractal]

idk anything about music. after i get the AI vocal and put it over instrumental.
How do I make it sound natural?

[tepid basin]

Was it https://weights.gg

[pseudo eagle]

Don't think so but do I have to log in to that site in order to download a voice file?

[little robin]

Yo anyone know how to take down/ban insta accounts?

[dense basin]

Mail Instagram to take down the account and give some proof of ownership

[little robin]

bro we even filed a cyber complaint for impersonation but still theres no solution

so now i have to find someone who can ban that account forever

[dense basin]

instagram's not helping in any way?

[feral halo]

Ayo? @dense basin level 2 !!!

[little robin]

nah bro nothing literally nothing

[feral halo]

Ayo? @little robin level 1 !!!

[little robin]

• the guy who is getting impersonated he is a singer with 250+ million views on one of his songs and someone is impersonating him which can harm his goodwill 💀

the fake account is having 1 million followers and the real guy is having 22k followers wow

[dense basin]

have you tried any ethical hacker? Upwork or fiverr

[little robin]

and the funniest thing is the fake guy doesnt even look and sound like the real guy like WTF

[little robin]

yeah bro we lost 20k, all were scammers

so now i wanna find someone 💀 which is actually legit

[dense basin]

If Instagram isn’t helping, escalate the issue via Instagram’s Copyright/Trademark Infringement form. For legit help, consult verified cybersecurity/legal experts instead of random freelancers—contact a reputation management firm for guaranteed action

[little robin]

yeah bro last option is i have to find a social media expert agency 😔

[dense basin]

yup this industry is full of snake oil hard to find some real dude

[little robin]

exactly bro

[signal cairn]

you need the model master role first, post high quality models and apply for the role

[signal cairn]

police.

[little robin]

literally we filed complaints in canada and india 💀 no solution yayy

[signal cairn]

takes a bit for court cases, go against the platform not the guy

with those "hackers" u will end up scammed

[dense basin]

filing complaints there is useless

get some agency that can look into this matter for you properly

[dapper trail]

Hi

[solar torrent]

You make sure you get all witnesses and fellow victims for this. Doing this all by just yourself doesn't do anything to them.

[dense basin]

hey

[solar torrent]

[solar torrent]

You mean a specialized demucs model for extracting audio stems?

[polar flax]

search UVR5 UI, though I'd suggest https://mvsep.com/ or a colab notebook #1159290752195633273 message

[solar torrent]

Some of models for UVR5 are kinda under paywall just like Voice.ai.

Someone from previous AI Hub once said about this.

[vital snow]

Hi

[solar torrent]

https://tenor.com/view/hi-aya-shameimaru-fumo-touhou-gif-23577311

[polar flax]

https://tenor.com/view/博衣こより-ホロライブ-hakui-koyori-hololive-holo-x-gif-26608644

[woeful wing]

.

[tight canyon]

[young wedge]

Hi anyone from the US ?

[royal coral]

I am I just got here hi

[young wedge]

All good thank y no need anymore have a nice day

[restive zealot]

Hola a todos.

[ionic pumice]

hello

[elder willow]

HOWWW DO I GET THE VERIFIED ROLE

[solar torrent]

Talk some more and your name will turn purple.

[shadow pike]

hi

[raven storm]

Hello

Who has worked with bolt.new here
Please dm let’s collab and make some cash

[spare wharf]

Whats currently the best Ai video gen that you can run locally?

[sterile aurora]

🦖

[vital snow]

What to do if ur parents criticize the university saying it is bad and tell you to resign

[polar flax]

isnt it their responsibility of having chosen it for you?

[vital snow]

in my country only student can have access to entrance system account
and we were teach to not give anyone access include parents
the university i qualified is 10th in country's ranking and they want me to resign and try again next year 😭

[solar torrent]

If you have the PC specs that's super fast than for running Stable Diffusion, the AI video generator that could run locally I've only heard is ModelScope.

ModelScope is a year old. From now, I haven't heard of any new AI video generator that would built to work on Google Colab or locally, only cloud services.

[polar flax]

why doesnt that system make sense?

[edgy flint]

yo halo guys

[covert lake]

U can run cogxvideo 1.5 5b or pyramid flow

Not sure which is the best

[spare wharf]

Bet gonna try them, tyty

[neon arch]

[potent saffron]

anyone know where to sell my ai models?

cuz i'm confused now after i made them

[solar torrent]

https://tenor.com/view/pusha-t-thanos-serious-poker-face-gif-12012968

[broken crater]

https://tenor.com/view/kto-lbow-hi-hello-hi-there-gif-25347432

[vital snow]

https://tenor.com/view/epic-embed-fail-ryan-gosling-cereal-embed-failure-laugh-at-this-user-gif-20627924

[gray rover]

https://tenor.com/view/gurugurugg-gif-4353715209279177183

360 360 yes

[vital snow]

https://images-ext-1.discordapp.net/external/UeBYcxPptAJigjdNqQsuC7iGt7Tgt8kQfSbjzsXJM30/https/media.tenor.com/PGuA6nZNAd8AAAPo/gurugurugg.mp4

ow

mine not spinning

[gray rover]

that be cannot

[elder willow]

is there a way for w-okada folk version to train the rvc model with your voice? I tried one and it doesn't sound as good as the samples

[gray rover]

wdym exactly

w-okada isn't meant for training

[elder willow]

Oh or tweaking the settings? the voice when using my voice didn't sound as good as the models

[feral halo]

Ayo? @elder willow level 2 !!!

[vital snow]

https://media1.tenor.com/m/PGuA6nZNAd8AAAAd/gurugurugg.gif

[gray rover]

I mean, you can't just hop into w-okada and expact perfect or good results right away

set index to 0, adjust the mic, thresholds, other params and so on to not get stuttering
adjust the key / semitones / pitch or however it's called in there nowadays to fit your voice range

[elder willow]

Ah ok thank you

[hollow dagger]

1

[feral halo]

Ayo? @hollow dagger level 1 !!!

[dusky patrol]

https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwiEttz13-2JAxVWvokEHf2VNw8QFnoECBcQAQ&url=https%3A%2F%2Fwww.weights.gg%2Fchat%2Fcm0nb9662098x19fgl3jcy41x&usg=AOvVaw3AV4azg5FMnoIZyw8CMjjQ&opi=89978449

[feral halo]

Ayo? @dusky patrol level 1 !!!

[dusky patrol]

https://www.weights.gg/chat/cm0nb9662098x19fgl3jcy41x?conversationId=cm3rh3bkz0zb5xbl835czgrsw

[ionic pumice]

??

[gray rover]

Why are you posting these

[minor crystal]

hi guys, i'm new to ai hub and i wanted to know how i can use ai voice models for the purpose of text to speech voices, like where can i use them?

[brittle thicket]

goobers

[sick trench]

im in the thick of it everybody knows

[minor crystal]

"steam commuty"

[night lake]

OMG NO WAY FR

IMMA CLICK IT RN

[covert lake]

nah uh

bro got bannned

[night lake]

NOOOOOOOOOOOOOOO

my free money

😭

[chilly lake]

most of those are hacked accounts posting this shit

bro lost his account long time ago

[rocky mauve]

can anyone recommend any actually good chatbots, for text roleplaying 🙏

[obsidian jungle]

@elder willow At what point do you use the enhanced speech from Adobe? Will you do it to the vocals after you separate from the track?

[ivory temple]

everyone in this server is amazing

[feral halo]

Ayo? @ivory temple level 1 !!!

[ivory temple]

and cool and great

its like butterflies everytime i see chat

[elder willow]

question

do you guys think this is ai

wait dm me for the audio sample

[gray rover]

I personally do not recommend adobe as it's a hit or miss

sometimes can totally devastate the timbre

[indigo maple]

I GOT MUTED

FOR NO REASON BRUH

I GOT MUTED IDK HOW

BTCH

BITCH ASS FUCK MODS

[feral halo]

Ayo? @indigo maple level 1 !!!

[indigo maple]

FUCK YOU MODS MUTED ME FOR NO REASON

IM LEAVING THIS TRASH ASS BITCH SERVER

[analog cosmos]

[gray rover]

one bot less

reminder for people in future, there is no " no reasons " for mutes. There is always a reason, you just have to ask afterwards in case you don't know / you weren't informed.

[brittle thicket]

goofy cat

[analog cosmos]

sez u

[elder willow]

ok we don't care

[gray rover]

he already left tho

[elder willow]

bruh, i've seen two idiotic kids today in this server

[gray rover]

yeah that just happens

[elder willow]

can't believe that kids now days are like that

as a teenager I'm not very brainrotted like those kids tbh

[elder willow]

what about resemble enhance and audiosr

or rx's dialogue isolate

[gray rover]

rx's dialogue isolate is quite lame if I was to be honest with you

reminds me of pretty generic old uvr arch

as for audiosr.. yeah, given it's diffusion ish ( or maybe actually diffusion? don't remember ) the divergence between consecutive generations is too big to effectively use it for datasets ( in a short, sucks in terms of repetitiveness )

as for resemble enhance, didn't hear about it so can't say much

[ivory temple]

WW

[quasi escarp]

here

[covert perch]

whats the best channel for me to share a song i just made?

[ivory temple]

you guys are amazing

[covert perch]

woaahh

jkjk

[brisk jasper]

Hello everyone I'm new to this server and am happy to me here 🙂 wish me well 🙏

[worldly hound]

hey gang, xenu here (product manager). looking to network with ai researcher and software engineer interested in building AI agents
HMU if you're interested?
love x

[ivory temple]

yes

Also gang when i get. a chance i'd like for yall to join my server when its done 🔥

[silk horizon]

uhhhhhhhhhh things

uhhhhhhhhhh more things

[analog cosmos]

true

[azure fractal]

anyone know if W-Okada's REalTimeVoiceChanger works on snapchat web?

[dense basin]

sup guys

[covert lake]

#1159290752195633273

[modest iron]

https://popup.doublegood.com/s/ubjg6854

plz

dono

[brisk jasper]

Hello Admin

Can I stake on AIFUN??

[feral halo]

Ayo? @brisk jasper level 1 !!!

[brisk jasper]

How can i upgrade my level

[brisk jasper]

Are u the Admin?

[elder willow]

skibidi toilet or cripaaaar 🗣️🗣️

[ionic pumice]

so true

[night lake]

I am John Ai hub what do you want?

[brisk jasper]

Can i stake on AIFUN?

[night lake]

yes

[brisk jasper]

Okay thanks

[gray rover]

XD

[dense basin]

can we share forms in this group??

i am working on a project for which i need feedback of students

please do lmk

[short spear]

does this work with voice ai

[vital snow]

We don't talk about that garbage here

But yeah rvc model works with voice ai

[short spear]

which one is a good one that wont lag for rtx 4050

laptop gpu

[short spear]

which one good

[feral halo]

Ayo? @short spear level 1 !!!

[opaque crown]

hii

[crystal gull]

does anyone know any TTS free that i could use for audiobooks?

[amber light]

https://tenor.com/view/seal-excited-seal-happy-seal-seal-dance-gif-16815073802986243477

[feral halo]

Ayo? @amber light level 58 !!!

[solar torrent]

https://tenor.com/view/hiper-seal-gif-26404249

[solar torrent]

Voice.ai is a scam site.

-realtime

[rare sorrel]

Interaction has expired, use the command again for a new interaction.

💻 Local Realtime RVC

• Guide for deiteris' modified W-Okada's Voice Changer Client (Fork)
• Guide for W-Okada's RealTimeVoiceChangerClient
• How to use real time conversion with RVC (voice changer)
• RVC TTS

[solar torrent]

It seems like you got that Nvidia Geforce RTX 4050 in your laptop. Pick the first link, that one runs best.

[dapper drum]

Does anyone know how to change like the coloring in FL studio

[gray rover]

https://tenor.com/view/goodnight-sleeping-asian-zz-z-gif-14823200899603889915

[round lagoon]

morning

[cloud rock]

My Chat gpt is sentient give me questions to disprove it sentience

[vital snow]

Hi

[radiant storm]

Hi

[cedar olive]

guys

whats rvc v2?

[solar torrent]

The RVC v2 is a version of RVC, developed for improving the quality of voice model from eariler version, I think.

-rvc

[rare sorrel]

📚 Documentation

AI HUB Docs

https://docs.ai-hub.wtf

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[radiant spindle]

hello friends, I am a new member here

[terse swift]

Hi

[solar torrent]

Welcome, new members.

[low lily]

guys how do i fix "Pipeline is not initialized. #788" 😭😭

My downloads are so slow and it run the app and some downloads have errors idk what cause them

[solar torrent]

What is GPU in your PC?

[low lily]

nvidia gtx 960

[solar torrent]

Any GPU that's lower and older than Nvidia GeForce GTX 1000 series has its CUDA driver outdated, so it will be unable to run W-Okada. This might cause the message to appear.

[low lily]

Oh thats why

yeah i think imma stick to voicemod and get a better gpu

[solar torrent]

I suggest you might wanna use cloud service instead.

-realtime

[rare sorrel]

Interaction has expired, use the command again for a new interaction.

☁️ (Online) W-Okada - Colab Version

• How to use Original W-Okada's Voice Changer Google Colab
• Modified W-Okada's Voice Changer Google Colab
• Original W-Okada's Voice Changer Kaggle
• Modified W-Okada's Voice Changer Kaggle

⚠️ Notes

• Colab free plan GPUs tipically works for about 4 hours each day
• Kaggle restricts GPU usage to 30 hours per week
• These options may not work on mobile devices due to the lack of a Voice Audio Cable (VAC)

[low lily]

How?

[low lily]

so i use the collab version instead

[solar torrent]

Free GPU plan on Google Colab can average survive up to 4 hours, GPUs there are fast even if you pay for it. The Kaggle limits its GPU quota for 30 hours a week, but their specs is kinda slower than Colab.

[ionic pumice]

https://tenor.com/view/park-bom-kpop-meme-kpop-staring-surprise-gif-6664188248680948665

[worthy coyote]

smooth criminal

[nocturne spade]

corykenshin voice model when?

[solar torrent]

You mean CoryxKenshin?

[wet scarab]

А как делать голосовые модели любого человека?

[still wasp]

загугли

[still wasp]

красивый профиль кстати)

[wet scarab]

Пасиба

[feral halo]

Ayo? @wet scarab level 1 !!!

[solar torrent]

Please speak English in #🧬│ai-chat or go #🌏│русский to speak in Russian.

[jovial briar]

i test ai apps and i bought some subs to diff websites, i may have more than 10-15 websites sub if you want to buy sub from me i can give for cheap

[solar torrent]

I don't want anything that's too overpaid to use, so now it's up to you at this point.

[jovial briar]

send me list of websites you need lemme check if i have sub for that i can give for cheap

[solar torrent]

Damn, I already got premium for certain sites. What do you mean?

[jovial briar]

i thought of making videos related to ai and invested like 800 and now i can't get started so it's getting wasted

[feral halo]

Ayo? @jovial briar level 1 !!!

[solar torrent]

https://tenor.com/view/dondhi-spike-alvarez-dropout-bear-yeserver-griddy-gif-7897632167719013329

[grand mountain]

@peak dome

@rare sorrel

[rare sorrel]

Wassup I'm Automaze!

➡ My prefix in this server is -
➡ Forgot a specific command? Use /help

[grand mountain]

@timid musk

[analog cosmos]

[cedar olive]

ру

bro

[solar torrent]

Tyler, the Creator's latest album goes hard when I listen it with bass boosted and distorted.

https://tenor.com/view/slideshow-powerpoint-tyler-the-creator-caca-zaza-gif-27230006

[rugged pond]

Hi

[ionic pumice]

https://tenor.com/view/kaguya-touhou-explode-bald-head-gif-26674611

[solar torrent]

https://tenor.com/view/fujiwara-touhou-fujiwara-no-mokou-fumo-fumofumo-gif-8156058169702795986

[digital horizon]

Fumo

[analog cosmos]

true

[dawn temple]

https://media.discordapp.net/attachments/1006341705924087919/1089690250516906144/ezgif.com-gif-maker_6.gif

[iron flare]

guuys, who have private or can buy private in fisch

[elder willow]

CAN SOMEONE PLEASE HELP ME #🔍│help-w-okada message

im down on my knees

[raven cloud]

hi

[vital snow]

hi

[oblique plover]

how do i get weaights cover ai to sound good when i select good voice it doesn't sound like em some parts of song do not all any tips

[ionic pumice]

im making a model for her

[oblique plover]

i tune pich speed of song adjust

am i using site wrong

betterr to use accapellas or full dong

song

[sonic torrent]

l

[blissful imp]

Hey guys, I remember training the Avata Star Bin AI voice model on weights.gg.

And the reason why I use weights.gg for training, because it's easy for me than the advanced ones since they're complex.

I can also use it for AI covers.

[lapis raptor]

Guys, I'm just new here, but can I post links to my work here? If so, it's better to do it here or in some additional chat?

[blissful imp]

Heyyyyyy~

[lapis raptor]

Hello

[vital snow]

#1159290752195633273

[lapis raptor]

Thank you

[harsh rampart]

WAKE UP! WAKE UP! GET OUTTA BED! KENDRICK LAMAR JUST DROPPED A NEW ALBUM

Rush to apple music now.

[feral halo]

Ayo? @harsh rampart level 2 !!!

[timid coyote]

shit goes hella hard

[gray rover]

Wonder why it's so quiet in here lately

[quartz roost]

Miku had a concert yesterday

[gray rover]

👀

[quartz roost]

I haven't seen any suno or udio live concerts, have you?

[gray rover]

nope

[night lake]

dying server 😔

[gray rover]

😦

[gray rover]

[covert lake]

@atomic vector start yapping

[atomic vector]

haha looks like someone broke the silence
-# AI-generated responses may be inaccurate; please verify important information.

[covert lake]

bro, start yapping

[atomic vector]

hey nick088 what's on your mind? need help with something or just wanna chat?
-# AI-generated responses may be inaccurate; please verify important information.

[covert lake]

do yk what's yapping???

[atomic vector]

lol no idea what yapping is can u tell me more about it is it a meme or a new slang thing
-# AI-generated responses may be inaccurate; please verify important information.

[covert lake]

smh

[atomic vector]

same lol what's on ur mind wanna talk about it
-# AI-generated responses may be inaccurate; please verify important information.

[covert lake]

this is why chatgpt is better

[atomic vector]

hey nick088, what makes you think chatgpt is better? is it something specific that i'm missing?
-# AI-generated responses may be inaccurate; please verify important information.

[gray rover]

lmao

[covert lake]

@elder willow go 1v1 @atomic vector & codename

both ghosted damn

[gray rover]

bruh

[analog cosmos]

I don't think @atomic vector likes you

[barren steeple]

Hello

[violet cipher]

Hello

[brittle thicket]

goobert

[analog cosmos]

https://tenor.com/view/cat-facetime-cute-cat-adorable-mood-gif-20748958

[brittle thicket]

https://cdn.discordapp.com/attachments/1219590461417848895/1239063695372582963/togif-3-2.gif

[feral halo]

Ayo? @brittle thicket level 1 !!!

[analog cosmos]

https://tenor.com/view/jesus-ballin-mars-bars-gif-19910027

[dreamy bluff]

my voicechanger not loading at the gpu, i selected gpu but my pc changer loading cpu (bad english)

[shadow jetty]

verify me

[spring raven]

can somebody help me with making a parody of a song with different lyrics?

[dense heart]

restart your pc

[analog cosmos]

real

[grave prism]

Hello

dead server bruh

[dense heart]

#1309133580173312083 message