🧬│ai-chat

[upbeat trail]

[sterile salmon]

hihi

[vague schooner]

hi guys

This is one of my very first websites, and I’ve put a lot of time and effort into making it the best I can.
I’d really appreciate your feedback — please rate it out of 10 and let me know what you think. Is it good? What can I improve? 🙏

https://thelittledev1.github.io/support.com/

[paper flare]

if dataset was subjected to stretching, how model will be affected? pretrain will correct this or the graphs will be weird

[gray rover]

Def some issue on their side OR they have changed their queue or waiting policy? not sure

but it should absolutely never take that long

to give you an idea, even inference on cpu ( which is considered slower by at least 2 to 5 times than on gpu ) takes 2-3 * the length of song
( so, 3 min song would generate in 6-7 or 8 mins, say, on i5 10400 - at least that's what I remember oof )

So yeah, def something is up

[dusty grotto]

hello everyone, can someone give a name to really high-quality models?

[gray rover]

well, unfortunately " high quality " is very subjective and quite often biased

so I'd really only recommend trying some out and judging on your own

[dusty grotto]

I've tried more than 50 voices, and all of them are very often not of high quality and not even close to the real ones.

That's why I'm asking right here.

[gray rover]

well, can you send an example sample of what you deem " not high quality "

could very likely be you have too high expectations and what you see as low quality, is quite often a consequence of current limitation of the tech
( potentially, that is

[dusty grotto]

wait

https://huggingface.co/AIEnhanceVoices/strawberryliaVA/resolve/main/strawberryliaVA.zip?download=true

this

[gray rover]

one sec

[dusty grotto]

It breaks off very often, and all my words start to stick together.

[gray rover]

Oh, but isn't this a model

I can't infer rn, mid training

's why I asked for samples ( outputs from rvc / applio

[dusty grotto]

you need souud?

[gray rover]

Well yes, that's samples yeah

[dusty grotto]

ok wait

[gray rover]

sure, no rush

just make sure to send .flac, avoid any lossy codecs such as mp3, aac, ogg etc

@dusty grotto ^
( also ye, saw the dm

Alr lemme see

[dusty grotto]

i send watch private messages

[gray rover]

Ok yeah, that is indeed low quality

quite " resonating " and smeary sound

But then, not all models are like that really

@dusty grotto Aside, 32khz model is not what you can call high quality to begin with

So first and foremost, do not aim for 32khz, 40khz is your minimum ( good 48 models are rare anyways as they're more difficult to train right

Also please use this chat and not DM, I want others to benefit from such information as well

[dusty grotto]

You can't send screenshots and the like here.

[gray rover]

And as for your question on the voice changer, you should ask in #✨│ai-help

[gray rover]

Because you are low level, that's 1 and 2, you don't ask for help like that in here anyway, instead on the channel for support

[dusty grotto]

Okay, I realized that I would not be able to achieve the perfect voice, thank you.

[gray rover]

Well, you can, but as I said, not with 32khz models

• not 32khz but 40khz ( 40k ) or 48khz ( 48k )

[dusty grotto]

and where can I find a model for me?

I've been around so much and there's nothing

[gray rover]

No idea man, just search around

Well, and for what kind of model you aim for anyways

if it's some crappy e-girl or whatever, I am not surprised

that kind of crap is notoriously known to be over-produced and rarely is good unless certain experienced people make it / or if it's a comm model

tl;dr, it is harder to find " HQ " e-girl or whatever than actual characters' / actors voices

[dusty grotto]

I'm just looking for some cool voice to laugh with my friends, I'm not going to pretend to be a girl at all.

[gray rover]

I ain't implying that, instead I say that a lot of such " low sassy fry ish breathy " e-girl like voices are low quality

So that's that

[dusty grotto]

And I'm interested in playing with voices.

[gray rover]

Well that's cool yeah

[dusty grotto]

good

Thanks, bye

[gray rover]

ig, kinda funny to see people say a lot / too many of models are low quality when in reality, 80% of what they tested is under or overtrained 32khz

lolz

[night lake]

prob a settings issue if words are being messed up

[gray rover]

nono, the model is indeed low quality

[night lake]

like so bad is slurs words?

[gray rover]

check it out:

very " blank " in color / tone, 32khz, has flute-ish / water-chamber overtone to it

that's sadly a model I'd easily give 3 or 4.5 out of 10

[night lake]

oh yea

it sucks

[gray rover]

yuh.. so that's a rip I suppose 👀

maybe you could recommend em some?

or uhhh, idk, I think there was someone into e-girl voices? I believe? don't remember who tho

[night lake]

ive made like 40 of them

[glad nebula]

sounds like someone applied 30 denoisers

[night lake]

but unless you talk like the model its gonna sound bad

[glad nebula]

razer

[gray rover]

oho

[night lake]

e girl models are so hard to make good because all audio of them is so monotone

[glad nebula]

32k + monotone asf audio

[night lake]

so unless you talk and act like the model its going to be bad

people expect so much from egirl models

its wild

[glad nebula]

this applies for every rvc model

but idk my model sounds just fine and i dont talk like him

[night lake]

thats why i gave up on them

:P

also im so tired of listening to asmr

[gray rover]

bruh
look at this nice line
( orange )

nice as in, it's shape and all

[night lake]

[gray rover]

kinda doing some tests with noise injection

and uhhh, if overlapped

🤔 yes.

better dynamics mh mh

[night lake]

noobies did something like that a bit ago

[gray rover]

ye but it's probs a lil different

iirc, he tried the adain thing taken from refine
and if not, we probs have different params and placement in the forward

Also, I don't really have his findings so, wanna see on my own if I can make anything cool

[night lake]

hope you find something good

[gray rover]

I mean, it's a bonus thingy I thought of today, only a part of final param / features test pre seoul training ( potentially

[spark saddle]

<@&1159293204038955078> <@&1347412420498292827> <@&1159293140440723499> <@&1361336153764266025>

goodnight

[still sigil]

[stark scarab]

https://tenor.com/view/ban-banned-sakura-anime-spray-gif-22585378

[gray rover]

welp, I don't think it was a smart move yk

in any case, if it's some rage-bait, won't work mh mh.
if anything, you'll get warned or muted perhaps

[dapper drum]

Question, which model do I use to put the OG backing vocals into my AI cover is it anvuew2

[night lake]

w anime btw

[gray rover]

wait what's that

[hollow oasis]

Hello, is there any reason u pinged us?

[gray rover]

gimme gimme

[night lake]

zombie land saga

very good and very goofy

[gray rover]

oh, might check it

is it similar to zoom 100? vibe wise or nah?

[night lake]

i havent seen it but from what i know of zom 100 its not like it at all

this anime is about zombie idols

[dapper drum]

Dereverb vs Karaoke: which first?

[gray rover]

oh, so comedy I assume

[night lake]

the girl here is sakura

[night lake]

yea

[gray rover]

well, zoom 100 was like 60-70% comedy so ain't that bad, thx for the title anyways

[dapper drum]

?

[night lake]

👍

[craggy ice]

also not everything is anti ai

thats a brainless term only npcs use

[sterile rampart]

good night too, but please don't do it for no reason as it's needed for emergency action

[ionic pumice]

💀

[tight lion]

Does anyone know an ai camera?

[magic moon]

hey

how is everyone doing?

[deft surge]

I want tô match you madre

Pau nu respects to padre

Myamore

@magic moon please found YoloTheCeoFromMoms in Roblox

[ionic pumice]

mi amor

[carmine flume]

Español

[red lintel]

Is there a better voice chat program than W-Okada now?

[covert lake]

Wokada deiteris fork, use #1192011222023950368 and tell ur PC GPU

[weary moth]

I have access to Manus AI. If anyone is interested.

[covert lake]

fine wbu

[covert lake]

wdym

[soft elbow]

can someone upscale a video for me?

[covert lake]

if you tell your pc gpu in #1192011222023950368 helpers can help you into know how to upscale one

[supple stream]

many of us do have access to manus ai agent

[soft elbow]

too much work bro i jus want a video upscaled by one off yall

[queen kernel]

Looking like a promo 🧐

[covert lake]

@mental nova promos ain't allowed

[ionic pumice]

philo ??

fuiro mentioned

wuis

[cerulean warren]

.

[jovial patrol]

Whats up

[weary moth]

that's why I asked if anyone was interested, not if they already had access.

[atomic surge]

where is the help channels?

i don't see them anymore

[covert lake]

#1192011222023950368 #✨│ai-help

[sage moss]

hey guys is there an ai that generates music based of off other music? Im tryna change genres of a song

[low igloo]

hey can anyone assist me in the support VC?

[elder willow]

Where to sell my code

[late condor]

okay weird question, but how would i go abut making a voice changer ai tof my friends voice? like i want to be able to speak in their voice. (i do have written permission to do so)

[minor blade]

Get good/decent quality recordings of your friend's voice (40-50 mins would be ideal for realtime) and then clean the audio and train it on Applio.

[late condor]

thank you!!!!!!

[minor blade]

You're welcome.

-rvc

[pine acorn]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[minor blade]

There you got the docs too.

[late condor]

thank you!!!!!!!!

[minor blade]

🐢👍

[vestal robin]

what programs do i use the voice models in? are okada and rvc the main ones or is there better ones cus theyy are kinda laggy for me

[magic moon]

@tepid basin

get on vc people are wild

i cant control this all by myself

[tepid basin]

No

[gray rover]

[ Info for those with no access to model maker chat channel ]
My applio fork's got updated ~

[still sigil]

@rain iron this server isnt the place to find a job

[languid whale]

Please could someone tell me a name for a free AI upscaler like waifu2xGUI that I can use? Not sure if this has been mentioned in a chat or somewhere but this is the best server I know of that would know this sort of thing... And I've been here for a while to know that people in here are quite knowledgeable in AI shit.

[magic moon]

@tepid basin

[languid whale]

For video... Forgot to mention that... Like what Waifu2x can do

||https://github.com/AaronFeng753/Waifu2x-Extension-GUI||

[still sigil]

@inner nymph this server isnt for advertising

[still sigil]

please speak in english

especially if you decide to insult people here

[stark scarab]

lmao

[strange shard]

ey

[strange shard]

hola amigo

puedes ayudarme con algo

[dense ginkgo]

Speak english i guess,you know how to use chatgpt ?

[dense ginkgo]

For image or video ?

[dense ginkgo]

Damn what a great startup idea,like pick a song and it auto generates another happier/sadder/etc.
I don't think there is a specific one,but there are many IA helpers to edit video.

[dense ginkgo]

Running on pc is heavy, this server has a different version if you want to try. Deuteris fork is the name (fork means is not made by the same author- okada). Mods here say it is lighter and better than "traditional"
Running on cloud is another option.

[snow delta]

Use tile upscaling.

[lusty oxide]

https://www.raspberrypi.com/documentation/accessories/ai-camera.html

@tight lion 👆

[paper flare]

assuming models are multispeaker (0,1,...), will training them for n epochs will be equivalent to: train 0 speaker to n epochs; train 1 speaker to n epochs; ... ?

[dense harness]

how do i check that RVC starter guide? i need the link to the newest realtime rvc voicemod for AMD GPUs but i forgor 💀

there's that bot command that lists some related shi

!rvc

[pine acorn]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[chilly lake]

multi-speaker dataset trains all speakers at the same time, it does not really produce x unique speakers, rather the model adapts to handle unique values better

[covert lake]

ask in support channels

[covert lake]

We aren't a service for paid apps

[covert lake]

tell ur pc gpu in #1192011222023950368 and explain what u want to do

[covert lake]

maybe ask in #1192011222023950368

[paper flare]

so for example i can make all video game characters models in 1 model but in multiple speakers at same time? will training time increase?

[chilly lake]

you're not going to get unique voices

you'll get a model that uses whichever voice is closer to the audio you're trying to infer

it is weird

[paper flare]

oh i thought its a shortcut

thanks for clarifying

[soft tusk]

hello

[covert lake]

Hello

[covert lake]

Wsp

[remote grotto]

Hey everyone! 👋

I’ve recently seen some examples of Gemini Flash 2.0 Experimental being used for image generation inside tools like ComfyUI, possibly through custom nodes or API integrations.
Does anyone know if there’s an official API, beta access, or any documentation on how to set this up?
Or maybe how to connect it via Vertex AI as part of a programmatic workflow?

Would really appreciate any insights or experiences you can share. 🙏
Thanks in advance!

[umbral aspen]

hi guys
i have just arrived here
i am doing my first machine learning portfolio project
which is about electric vehicle price prediction dataset
i am talking with google ai studio to get some guideline
but still i feel like i need human help from real ai enthusiasts
so thats why i am here

is there anyone who can help me out when i get stuck while doing the ml portfolio project?

[chilly lake]

make something useful, like predicting the rain or snow

[umbral aspen]

?
electric vehicle price prediction is not useful?

[umbral aspen]

i mean...ok...if predicting the rain or snow is more useful than predicting ev sales, then i am ready to do that

[chilly lake]

why do you need to predict the vehicle price when it is clearly determined by costs and profit margin?

[open tapir]

I used Ai and made a project can I share the project?

[vestal furnace]

Hello everyone, is anyone familiar with the NotaGen model for symbolic music generation? I’m looking for assistance fine-tuning it on a different dataset.

[supple stream]

like?

[sterile salmon]

hihi

[open tapir]

A GitHub project a mini game typo Html

[paper flare]

Model Name: None
Model Creator: None
Epochs: None
Steps: None
Vocoder: None
Sampling Rate: 40k
Dataset Length: None
Creation Date: Invalid date format
Overtrain Info: None
Embedder Model: None
Max Speakers ID: 0Hash: None

can i somehow set this data after model has been created?

[chilly lake]

yes, can use python, import torch, load the model, set the values

[paper flare]

o

[astral turtle]

How can i use AI models for Music?

[umbral aspen]

because of fluctuating costs and market dynamics?

[umbral aspen]

why do u think predicting snow or rain is more useful than predicting electric vehicle price?

[polar flax]

both of them require observing facts rather than merely blind guessing

i.e. the costs in the supply chain and profit margin

[dense ginkgo]

I want to use Applio cloud is that possible ?

Running local is kinda heavy,3 gb

Small llama i guess ? Ia is so fast my info is 1 year old so very old by now

Llama is decent for performance/result ratio and you have tons of parameters to choose from depending on your machine

[dense ginkgo]

Think with me

If everyone thought tesla would increase 50% share price in 10 days, people would buy now therefore the prediction itself raised the price potentially making it false

Snow or rain doesn't have this problem but are less profitable to predict ,stock market is easy a billion dollar a year profit if you can

I believe weather is much more useful but less money

[chilly lake]

-colab

[pine acorn]

📒 Google Colab Notebooks

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.

• **Applio**

by IA Hispano
Google Colab

• **RVC Mainline**

by Hina
Google Colab

• **UVR5 NO UI**

by Eddy
Google Colab

• **UVR5 UI**

by Eddy
Google Colab

• **Wokada Deiteris Fork**

by Deiteris & Hina
Google Colab

• **Hina's Modified Original Wokada**

Google Colab

• **RVC-AI-Cover-Maker-WebUI**

by Shiro & Eddy
Google Colab

• **FaceFusion UI**

by Nick088
Google Colab

• **FaceFusion NO UI**

by Nick088
Google Colab

• **Music Source Separation Training (Inference)**

by Jarredou & Makidanye
Google Colab

[quaint dragon]

@pine acorn

[umbral aspen]

so should i skip ev price dataset and start working on weather dataset?

[modest wedge]

E

[bold rampart]

Does this sound natural, or should I edit it more to improve it? Please give me your feedback

Btw, this was made with Weights

I just added auto-tune, compression and reverb using FL Studio 21 with the Antares Auto-Tune Pro plugin

[polar flax]

that sounds like room reverb

mostly plate reverb would sound better

[mental bough]

nah why did litsa rejoin this time

[polar flax]

https://tenor.com/view/i-saw-what-you-deleted-cat-gif-25407007

[mental bough]

Looks like your model is not trained well because of the artifact problems, reverb or something like that idk

[timber saddle]

hey i have a problem with the make AI covers RVC can someone help me with that ?

[tawny wharf]

ハハハ

[worthy coyote]

/\/\/\

[inner pivot]

I want to create a phishing detection system by converting website screenshots into text using OCR. Then, from this OCR-extracted text, I want to search for keywords that might be impersonated by the attacker, such as 'PayPal' or 'Amazon'. What kind of AI model should I use?

[chilly lake]

is it like "take a picture of a screen with cell phone, upload it to itunes cloud, download it from the browser to usb stick, go to staples to print it" kinda route?

like... HTML source is not enough?

[timber saddle]

True

[inner pivot]

no i think i take a picture and then i upload it to discord bot and i will check if its phishing or not @chilly lake

like this paper https://research-repository.st-andrews.ac.uk/bitstream/handle/10023/17853/A_NOVEL_METHOD_TO_PREVENT_PHISHING_BY_USING_OCR_TECHNOLOGY.pdf?sequence=1&isAllowed=y

@chilly lake

[covert lake]

Elaborate in #1192011222023950368

[timber saddle]

Ty

[rustic crest]

what can you do

[hybrid abyss]

How do I make sure I'm training in rvc v2?

[chilly lake]

what are you using?

you really have to go on an archeological adventure to find an app that trains exclusively v1 models

[supple stream]

self promotion is banned here

[mild tapir]

Hey I'm new here so how can i get the voices

[minor blade]

Use the #1175430844685484042 channel.

[dapper drum]

Which one do I use first de reverb model or karaoke then de reverb model

[inner pivot]

How about this @chilly lake ?

[hybrid abyss]

Applio colab

[chilly lake]

then it is v2

we've removed v1 training a while ago

[raven epoch]

i think im blind cuz i cant find the link to do the voice changer, not the rt one

[still sigil]

@molten acorn no advertising

[molten acorn]

Ok

[molten acorn]

Im not scamming trust me

Would you recommend me a server to advertise in?

[still sigil]

not this one thats for sure

doesnt matter.

[molten acorn]

Ok

Would you be interested?

[still sigil]

no.

[chilly lake]

Things like that were done many times before, OCR, heuristics, etc

[inner pivot]

@chilly lake do you have resources or project like that?

[chilly lake]

no, sorry

[burnt sand]

Anyone have the workin rvc link for cloud ??

[tall dawn]

anyone speak portuguese but help me?

[covert lake]

Elaborate in #1192011222023950368

[covert lake]

This server is English only, and for help there is #1192011222023950368

[woeful marlin]

how do i get rid of the demonic whispering

in my ear

when using a voice model lol

[covert lake]

😭

Elaborate more the issue in #1192011222023950368

Gonna need an exorcism for this one ig 😭😭

[still sigil]

silence the devil on your shoulder

[midnight solstice]

Hi guys

Wanted to know if w-okada was safe and not had a malware/virus

Because i am scared to install a malaware or virus inside my new pc

[swift berry]

is safe

[solar torrent]

It's completely safe if you download it from AI Hub website. W-Okada is a free and open source program.

If one of your antivirus programs deleted the program, make sure to allow it back.

[charred gust]

Yo hi fellas

I'm back

In 2025

Remember me?

I hope united shoes will

[queen kernel]

Sorry but I don't.

Btw welcome back

[charred gust]

Some do

And some dont

[queen kernel]

Anyways. Welcome again

[charred gust]

Thaks

[dapper drum]

Im confused, in what order are you doing this? Getting Vocals/instrumentals, DeEcho-DeReverb Vocals, Get lead/backing vocals, then DeNoise those?
Using the new latest models

[lucid copper]

Huh

[ionic pumice]

kar before derev but other than that looks fine

[ancient phoenix]

Guys, some AI that can read text then voice and output mp3 file?

[ionic pumice]

elevenlabs

edge tts

uhh

coqui

gpt sovits

[ancient phoenix]

This costs too much money, is there anything cheap or free?

[ionic pumice]

look at the other ones i mentioned

[ancient phoenix]

thanks

[frosty bluff]

hey wait, is this app not usable anymore?

[ionic pumice]

?

[frosty bluff]

was AI hub taken down?

like, is the app still available to download?

[arctic tinsel]

2 years ago?

[frosty bluff]

I see I see, so what about the people that already had it installed, can they still use the service or is it kaput for them too?

[arctic tinsel]

app? wdym?

Did we have an app?

[frosty bluff]

I'm sorry, I have no idea how this works. is voice.ai where I'm supposed to use the models?

[arctic tinsel]

there were a few forks developed by some of the developers here, but I don't think there was ever an official AI Hub app.

oh..

I remember there was some kind of partnership, but I'm not sure what happened with it. The admins would know about that.

[frosty bluff]

Yeah, I'm just starting out, heh. is there like, some kind of voice changer I'm supposed to download to use the models on the server?

[arctic tinsel]

rn ai hub collaborating with weights.gg. you should be able to create or download models directly from weights.gg without needing any extra setup.

[frosty bluff]

wait about to actually use the voice models? What app should I use for that?

like, a specific voice changer

I thought it was voice.ai, but now I'm hearing its apaprently bad

[arctic tinsel]

It's not a voice changer… as far as I know, most voice changers use W-okada.

try asking in the ai-support channel.

[frosty bluff]

thanks, I'll check out W-okada

[covert lake]

the ai hub server got taken down in october 2023 but been up back since a long time

also, ai hub isn’t an app

[covert lake]

tell ur pc gpu in #1192011222023950368

[frosty bluff]

oh, alright done https://discord.com/channels/1159260121998827560/1371772592280571977

[kind wind]

Hi

That real AI hmmmm

[elder fractal]

Hey everyone!
Does anyone still have a ChatGPT Plus invite link for the free trial?
I’d love to test GPT-4o, but can’t afford the subscription right now.
I’d really appreciate it – thank you so much!

[idle fable]

hey everyone

can i post some songs i made with AI here in this channel to know people's opinion?

[wise yarrow]

What type of developer is suitable for startups?

Many clients today are stepping forward with their own unique ideas to launch startups — not just to raise funds, but to shape technology, push boundaries, and build the future. Startups aren’t just small companies — they are civilization builders and innovation leaders in their fields.

But no startup can succeed without the right people.

The ideal developer for a startup isn’t just technically skilled — they must share the same vision and determination as the founder. They don’t just “work on” a project; they live and breathe it. There's no set clock. No rigid playbook. It’s about adaptability, intensity, and belief in the mission.

Startups are rocket engines — and every developer onboard should feel the same momentum.

Yes, many founders choose full stack developers to save on costs — and that makes sense early on. But with that comes a trade-off: broad scope often means reduced specialization.

In the end, the right developer isn’t just a coder. They’re a co-creator, a problem-solver, and a partner in the journey.
It is me

[hollow oriole]

anyone want to help me with voicechanger?

[quick sequoia]

hey y'all, what would you reccommend for a good zero shot tts

[opaque jacinth]

brooo

vvhataə

[queen kernel]

F5-tts ?

Ask in #✨│ai-help

[queen kernel]

The help channel is #✨│ai-help

[calm solar]

okey

thx

[chilly lake]

depends on the language you need

chinese or english - many options

romanian - zero

[late storm]

gdzie piwo?

[crude tinsel]

Ok, it's been almost 6 months since I last checked anything related to RVC, has anything actually improved? I keep about hearing Applio improvements and some newer pretrained models, but does that noticeably improve the overall quality of audio outputs?

[light girder]

lots of testings, but improvements are on the way

hopefully

but rvc boss ? Bro forgot RVC

[gray minnow]

improvements are on the way

2027

[polar flax]

perhaps

but fellows in https://discord.com/channels/1159260121998827560/1159290193619189821 are still testing lot of stuffs

may not be stable for most regular users

but you're free to test and contribute in that channel

[charred gust]

Wassup

Another GTA 6 delay

People who waited 1 year

[polar flax]

im sure rvc v3 won't come before GTA 6

[charred gust]

...

Ok

Me bored

I really want to know if GTA 6 will work on PS4 too

[polar flax]

perhaps half life 3 could instead come

[charred gust]

Gawd damnit

[polar flax]

as they showed some progress as some leaks say

[charred gust]

I also heard it's worth is going to be more than Burj khalifa

Is it?

[queen kernel]

It only supports 30FPS in PS5 and maybe slightly better FPS on PS 5 pro. I don't think it will work on PS4. If it does maybe low resolution or huge frame drops

[polar flax]

console 30/60 fps moment

I'd rather run 60 fps in lower graphics settings

[queen kernel]

60+ are only possible only in PS6

[queen kernel]

Yep. I'm taking about 4k

[polar flax]

that's why upscaling exists

ppl may reject it but things may change with DLSS 4/FSR 4

even some ppl are fine with mediocre FSR 3

[queen kernel]

If we talk about RAW performance. Upscaling in sometimes bad

[polar flax]

upscaling is a new norm

where native 4k needs high end gpu for 30-60+ fps

[fleet iron]

Hey, how do I get around image rights restrictions? chat gpt isn't generating Goku for me.

[polar flax]

for just image gen, consider Flux

also weights site allows using it

for creations

[fleet iron]

Thx 🙏🏻

[sweet cedar]

So. What now? That Model Masters are gone

I mean. Wow. JUst like that

[covert lake]

everything just continues

[patent zinc]

Hey guys you have voice wussup my boy?

[lusty nova]

yoo how do i get and upload voices to vcclient

[queen kernel]

Ask in #✨│ai-help

[lusty nova]

figured it out

[still sigil]

@wise yarrow not the place to advertise for jobs (or find a job)

[wise yarrow]

got it, sorry

[elder willow]

can i speak russian here?

[still sigil]

https://discord.com/channels/1159260121998827560/1159346439424573440

[elder willow]

thanks

[versed tree]

deleted message

[polar flax]

please go to https://discord.com/channels/1159260121998827560/1159290139609137264

[idle eagle]

is it okay if my models sound glitchy

soemtimes

do i train them?

[heady oriole]

Has anyone tried this for enhancing speech/vocals to sound better, clear, or more high quality

https://github.com/modelscope/ClearerVoice-Studio

[bold rampart]

Has anyone tried this? This is a new text-to-video and image-to-video generation model, but can generate “infinite-length videos”, as said by the GitHub repo: https://github.com/SkyworkAI/SkyReels-V2?tab=readme-ov-file

[crystal breach]

can it make a video as long as grahams number

[shell crow]

hey i need big help

xD

this is the answer to everything

[chilly lake]

a modified frame pack can do infinite length video

[polar flax]

so basically fork of framepack?

[chilly lake]

Skyreels is a finetune of Hunyuan

just as framepack

but more disappointing

[covert lake]

promos ain't allowed

[hollow silo]

Hello, would this channel be the right place to share a master thesis AI-related study? I am looking for participants 😊

[solar torrent]

Your project or just some "promotion"?

[hollow silo]

Its my friend's project/master thesis study

Just thought I'd find people on this server willing to participate

[solar torrent]

To share your project, there is #1359898289335566570. But anything promoting or advertising your stuff is not allowed in this server.

[hollow silo]

It is not advertising in any case, it's for the purpose of passing a master thesis, that's it. I'll take a look into the projects channel, thank you!

[cloud ginkgo]

Are voice models on weights.com checked for malware and malicious pickle imports as good as on huggingface.co?

[ionic pumice]

if they werent they wouldnt be there

but applio dont execute code anyways so who gaf

[median falcon]

yo how can i make frutiger aero AI?

[chilly lake]

AI what?

[covert lake]

i guess he means ai images in a frutiger aero aesthetic? (like old pc)

@median falcon if this is ur question, then elaborate ur specs in #1192011222023950368

[chilly lake]

there are loras for that

for sdxl and flux

[hollow kiln]

Is there any way of cloning a voice and then have a real time voicechanger of that voice for free?

[fringe ibex]

procuro pessoas que queiram trabalhar online como freelancer (pago por cada trabalho a ser feito)
pode ser de qualquer parte do mundo
Mensagem privada

[minor blade]

If you make any voice model on RVC/Applio, of course you can use the same model on W-Okada/voice changer

-realtime

[pine acorn]

💻 Local Realtime RVC

Guides for Programs that use RVC Models in Realtime for Calls/Games

• Wokada Deiteris Fork

Most suggested. GUIDE

• Original Wokada

ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are way worse. GUIDE

• **RVC GUI Mainline Realtime**

Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated

[vapid oriole]

@gray rover (Sorry this is machine translation) I just installed and tried the released version 3.1.5 and RefineGAN learning doesn't work properly. In version 3.1.4 the learning that used to work normally is RAdam or
When using Ranger21 optimizer, Mel Spectrogram Similarity is shown as 0 and the loss value on the tensor board is measured high. However, when I set the optimizer with AdamW, it seems to work normally. (Not sure)

[dapper drum]

Has anyone test the?
New one

Apparently it use Vocal big beta5e and mel roformer becruily karaoke together

[sudden bone]

can samone tell what are the best settings like for in out moon pitch index

i have a e girl voice and need good settings

[nocturne timber]

Does anyone know a good place to order a commission for an RVC model now that they're banned here?

[wise kettle]

Anyone know a free ai that I can send unlimited images to, need it for maths homework

[thorn hornet]

what program is this

[gray thorn]

What is the best voice changer?

[dapper drum]

X Minutes

[vale sierra]

Guys, does anyone have a model of the alliance voice from Half-Life 2?

Or civil defense dispatcher

[hexed ore]

Hi

[mint granite]

My model maker role is gone 😦

[lucid copper]

Huh

[proper nova]

Anyone know of good text to ai vidoe generators

I made a few songs I want videos to

animations etc.

[verbal shale]

guys who has udio cuz I need a audio of a song that I wanna change the wocals to (plz)

[queen kernel]

Framepack

[queen kernel]

Hello

[queen kernel]

Ask in #✨│ai-help

[onyx zodiac]

Hi!

[gray rover]

you should look in:
#🔍│find-models
#1175430844685484042
or on weights.com

[onyx zodiac]

👀 been playing with offline voice cloning again

got it running locally with a tkinter GUI + encrypted engine
no cloud, no login, no torch dependencies. just run and meme

curious if anyone else here builds AI tools without huggingface or web calls?

[delicate tundra]

Documentations

🇬🇧 󠁧󠁢󠁥󠁮󠁧󠁿English

Docs
FAQ

🇫🇷 Français

Docs

🇮🇩 Indonesia

Docs
FAQ

🇨🇳 中国（简体字）

Docs
FAQ

🇮🇹 Italiano

Docs
FAQ

🇷🇺 Русский

Docs
FAQ

🇻🇳 Tiếng Việt

Docs
FAQ

Can't find your language?

Help us translate for 3 months of free VoiceMaster+!
Translation Form

VoiceMaster+ · May 15, 2025, 4:10 AM

[topaz veldt]

Hey folks! I’ve just joined the Hub — great to be here! Quick question: is it okay to share a video link here to ask if anyone can identify the AI stack used to create it?

[solar torrent]

So #🧬│ai-chat just become another Q&A channel about AI programs. That's crazy.

[ionic pumice]

?

[solar torrent]

Any idea you're curious about?

[tranquil lantern]

Fv5 is SO GOOD bro I'm cumming from how good it is

😭

[ionic pumice]

relatable

[tranquil lantern]

Yeah screw RX11 and its Spectral Denoise

I'm switching to this

[gray rover]

lol

best of luck pal

[polar flax]

for real

[topaz veldt]

Anyone?🥹

[heavy slate]

Hi, do you guys use replay.ai?

[charred gust]

No

[hollow oasis]

i think some people do, but u can find more replay users in weights server

[lament charm]

hey guys just wanted to say thanks for helping me be gendered correctly in vc while also using a cool unique voice aha

Been using the yuna from ffx filter a whole lot its so clear and breathy

what's y'all's favorite?

[dusk mirage]

trying to create my own model and none of them work so far

[lapis raft]

this not about fivem?

[river verge]

DIlly ding, dilly dong! Two new RegalHyperus drum models just released!
Il Vento d'Oro (v2) & Hes Biya (Drum models no. 602 & 603)

[lament charm]

good luck!

[drowsy quarry]

Does someone know which is the best place to find a client related to AI chatbot or neural networks or a job of AI engineer

[simple flax]

help me pls

IPAdapterAdvanced
Error(s) in loading state_dict for Resampler:
size mismatch for proj_in.weight: copying a param with shape torch.Size([768, 1280]) from checkpoint, the shape in current model is torch.Size([768, 1024]).

[covert lake]

elaborate in #1192011222023950368

be sure to elaborate though

[still sigil]

@plain spoke no advertising

[viscid pilot]

Guys. Im having a problem and that i need help

So im currently using google gemini

For validating my business concepts

Before i opt out i gave google gemini my business concepts and a few weeks before today, there was this instagram account that is new and was just created like a few days ago that has a similar business concept like mines

I have also been selected to try out google gemini’s new features

Plus the instagram account that had a similar business concept like mines has accounts that are new and are alts in their followers and are following just a few new accounts and they’re in a different country

Whats happening here?

[ripe spade]

that's why people use local LLMs for sensitive data

[viscid pilot]

Can you give me links please?

Thank you

[last shell]

what are best options for local LLMs? Llama?

[viscid pilot]

like ollama

[bold rampart]

Used, https://uvronline.app, got this result:

[still sigil]

clearly your business concepts are rather common

[viscid pilot]

what makes them common? give me brutal harsh truths

[still sigil]

what makes them unique

theres 8 billion people in the world, at least 1 will come up with similar ideas

[viscid pilot]

okay i mean i have stated my unique selling proposition in my business structure

and also my competitive edge

[still sigil]

so pretty run off the mill strategies

[viscid pilot]

okay im wrong with my business structures and the majority of whoever wants to go into business can come up with my ideas easily

[viscid pilot]

thanks for the reply. i'll learn

[chilly lake]

there are millions of people who all come up with different ideas and not every idea is unique

ideas themselves are pretty worthless, especially if they require significant investment with questionable payback

[viscid pilot]

so its whoever can execute them first

[chilly lake]

cheaper and more efficiently

[viscid pilot]

didnt sam altman said "start with the hard ideas first because that'll get the attention from investors more"

[chilly lake]

investors jump on things that offer better than the regular investment returns, considering the rate of startup failures it has to be a signigicant number

[grave eagle]

whahts the best API key to make a combat AI

[pure ice]

May I put images models here please

[opaque mango]

man

[final shard]

I love ai

So much slop everywhere

[delicate notch]

Can someone help me to find a checkpoint/lora with a simillar style for ai picture generation? I can show you examples in dms.

[queen kernel]

Yeah please

[queen kernel]

Which model images?

[dull snow]

how do i use the voice models as a voice changer?

[queen kernel]

Ask in #✨│ai-help

[minor ravine]

Gm

[spare torrent]

Good morning to you.

[vagrant fern]

Hi

[leaden solar]

where can i find beatrice v2 voice models

[polar flax]

go make one if you can

[leaden solar]

i cant

i dont have a gpu

[polar flax]

if you follow some video tutorial recommending beatrice models

you don't

beatrice lacks support esp for training

also yeah the models can run on average cpu

but the quality is mediocre

compared to RVC

[leaden solar]

rvc runs awfully slow on my machine

ive tried finding beatrice models but its like impossible

[polar flax]

speed only matters in realtime

and even for vst usage, you can always do offline pre-conversion first

[solar torrent]

If your PC CPU is old and low, that's what you can do.

Also, no one in this server has ever trained a Beatrice voice model, so that'll be a rare find I guess.

[viral dragon]

hi guys

where I can make the voicemodels online

without downloading anything

[polar flax]

it lacks support

for sure

[covert lake]

#1367522742735343758

[covert lake]

hello

[covert lake]

elaborate in #1192011222023950368

[covert lake]

you don’t, beatrice v2 models are lower quality but more lightweight compared to rvc v2, so no one uses them bc of the quality

dont use video tutorials

tell ur pc gpu and what u want tod o in #1192011222023950368

[covert lake]

tell ur pc gpu via task manager in #1192011222023950368

else if it has no gpu and its a 2010 laptop, u can use cloud (remote good pc)

but i can assure u it’s extremely rare to find a beatrice v2 model

[covert lake]

tell ur pc gpu in #1192011222023950368 first

[pure ice]

Yes but I need permission

[mellow siren]

I need help with Virtual cable setup

It worked earlier when I was using w-okada windows app

[covert lake]

elaborate in #1192011222023950368

[solar torrent]

For Virtual Audio Cable to use with W-Okada, go to #✨│ai-help or #1192011222023950368, not here.

[covert lake]

this isn’t the support channel

[arctic tinsel]

how to make realistic e-girl voice?

[solar torrent]

Goofy.

[covert lake]

he’s joking

[solar torrent]

Might make a sticker for that later, the one that asks for W-Okada in chat.

[dusk mirage]

man, Vonovox is just moah

dr87 really cooked this

[frosty elm]

hiiii

[covert lake]

hello there mikasa

[frosty elm]

im testing out a few voice settings, wanna help?

[covert lake]

elaborate in #1192011222023950368 for any help

[tawny parcel]

hey

are there any AIs that allow to translate audio files (or videos) between different languages, but in such a way that I can specify the translated text yourself, and the AI actually speaks the text with the same intonation as in the original?

[solar torrent]

I ran out of anime girl voice model ideas to make them sing BBL Drizzy.

[solar torrent]

W-Okada or RVC voice model training? For any of these, go to #✨│ai-help or #1192011222023950368.

[bold forge]

how are yall creating ur covers?

[ionic pumice]

from thin air

[covert lake]

tell ur pc gpu in #1192011222023950368

[covert lake]

#1192011222023950368 or #✨│ai-help

[stable shell]

hey guys anyone know whats the best e girl model

[minor blade]

It's matter of testing with models you can find on #1175430844685484042 channel to see which one fits your voice the most

[vapid moat]

can someone teach me how to download this stupid voice changer

[dusk mirage]

it's in #1159513888199540817

[vapid moat]

yeah found it. thank you 🙏

[spring loom]

Lookin for a tool that does tts via an api, any suggestions? Local ey

(I've got models)

[foggy beacon]

yeah that stupid voice changer

[stark idol]

hey

i have some questions

is anyone online here?

wow nobody here?

[chilly lake]

nope

[echo socket]

shape so cooked rn

[stark idol]

what is a realistic voice in #1175430844685484042

im trying to find a realistic one but so far i havent found one

can someone help me? thank you.

[vapid moat]

why is it so slow? Like… i say something and it’s 3 hours later

[foggy beacon]

adjust your settings

and check if your pc is even able to handle it

the guide should have requirements there

[fast seal]

wsg

Seems like I need to chat more to get perms

[dense ginkgo]

Wokada on collab, mix it up (for streamers). Maybe they work,no idea

[dense ginkgo]

They sound robotic ? Sometimes the problem is you,you need to adjust the settings depending on your voice and the model,not plug and play sadly

[dense ginkgo]

Check dackus on YT, guy is a professional on pretending to be women on dating servers

[dense ginkgo]

The most cutting edge stuff transcribes,translates and does tts similar to your voice on another language AND ADJUST MOUTH MOVEMENT on the video automatically , never tested it but amazing if it is legit

[lament tapir]

Yo what’s the best ai I can use online that’s free and you don’t need a subscription, and I can make those trippy vids like catsoup ai

[solar torrent]

If you wanna be an E-girl, become one.

[solar torrent]

You called W-Okada stupid? That's crazy. For more help about W-Okada, go to #✨│ai-help.

[gray parcel]

[solar torrent]

It was a joke. I don't help someone to set up W-Okada for catfishing.

[solar torrent]

Sorry, but I can't pick the most realistic voice model for you. To check which model is high quality and realistic, click on a voice model post. If there's a sample audio, click play.

Also, make sure you don't desperate to the point of asking anyone to ask you back on this one.

[solar torrent]

I'm not sure about one that has API. Applio has a built-in TTS, which uses RVC voice model.

[solar torrent]

Have you ever done these by yourself or you just watched some tutorial videos on YouTube? W-Okada doesn't seem to do TTS in itself.

[dusty atlas]

what happend the the ai model request channel?

[solar torrent]

This could be your answer. #✨│announcements message

[spring loom]

yeah i might just modify applio, unless there's a faster alternative (performance wise)

[bold rampart]

Used Clear-Voice Studio locally for this

[supple stream]

[bold rampart]

To install Clear-Voice Studio locally, here is the installation guide: https://drive.google.com/file/d/1s3NmU_g6LSmlKAyCichLTn7LPjjs-EQG/view

[glass maple]

Does anyone know of good AI research tools like OpenAI's deep research? I like using it because it will access information online automatically and generate well sourced long-form reports. However I'd like to find another tool where I can customize the sources that are used. Oh yeah, and doesn't cost $200/month lol

[chilly lake]

so you want to use hundreds of KW of power and AI accelerators paid by someone else to get something done for free?

[solar torrent]

So is it some kind of LLM?

[glass maple]

Yeah it’s a function for ChatGPT.

[glass maple]

Where did I say that it needed to be free. It just needs to be a reasonable price.

[solar torrent]

[glass maple]

I just want to try a different method for doing research

[bold rampart]

The Mel-Roformer by unwa v1e+ is really good, it has no artifacts, this is one of the examples, vocals only

I used https://uvronline.app

[solar torrent]

Don't. This is not where you promote your giveaway contest.

[supple stream]

<@&1159293140440723499>

[gray rover]

Careful there ~

or you might get striked by discord randomly, anytime

✨

[bold rampart]

Bruh, there's no instrumentals in it

[gray rover]

yet vocals

we don't know what kind of libraries and algorithms they have access to

and as they say, better be safe than sorry

You see, I thought I was safe too yet got flagged / got violation registered, and I don't recall posting any instrumentals in here

hence why I am letting you know

[hollow oasis]

👩‍🍳

[queen kernel]

Hi

[stable shell]

THATS WHERE I FOUND THIS SERVER 😭

[quiet oar]

sa

[polar flax]

both instrumentals and acapella can be copyrighted

even though it's extracted

[gray dawn]

Hello everyone, can someone please help me? I would like to have a short voice clip from Fatman Scoop or DMX, about 10 seconds long. I'm having trouble making the voice sound right. If anyone can help, please get in touch with me. Thank you very much

[polar flax]

please go to #✨│ai-help

[graceful stump]

Hii everyone

Can anyone help me to build my ai startup

[warped dune]

Hande Ünsal Ses Modeli kimde var zip dosyası olarak paylaşan olursa muhteşem olurdu

[solar torrent]

Please speak English in chat.

[warped dune]

I'm search Turkish Star Hande Ünsal vocal model as zip

[vernal estuary]

hello all im here for like minded people irl im with a bunch of sheep tbh

and ive already made myself rich from future leverage trading

whos into this and wants to talk? and who is interested in video creation A.I's???????? hit me up im tired of having conversation irl with empty brains lmfao i be in the wrong rooms....so im here now lmfao

[tame crypt]

Yo do you know Werenoi?

[dense ginkgo]

With enough Duct tape i guess you can do TTS in wokada, by API will be even harder.
Like virtual microphone hooked to google translate audio or smth.
Type -> google translate audio -> wokada changes the audio > download audio output ???
Maybe this can work

[dense ginkgo]

Paid model requests are banned bcs they are 99% illegal but free ones i dont know.
Can't find them too.

[dense ginkgo]

Bruuuh anyway duckus hides his "secret sauce" so i don't know what he uses in most of his videos.
Just pick some cute anime girl and change settings until you find something compatible. Takes a while but you will find it. Not everyone can hit very high or low notes,just find something compatible with you. Zero two from darling in the franxx as a first test for example.

[sharp vortex]

hi

Can someone help me with chunks and extras?

[boreal cargo]

I am running a local ai on a 3060 ti and a ryzen 3 series cpu what can I rent to render faster someone recommend me vps at a good price na servers I wanna be able to open windows on vps and run my own programs

[minor blade]

Actually i'm not sure if they was referring to the request-channel to just request free models.

[chilly lake]

I know a guy who does video production

[austere karma]

can you introduce me to him please

[empty hazel]

Gents, I have a question

What is currently the best model you can use to train voices and how can I get it?

I don't mind if it's paid

Because after seeing what ElevenLabs' Flash 2.5 and Turbo 2.5 can do I am jealous

[empty hazel]

just to clarify I am willing to pay for models I can use to train better voices and not ElevenLabs itself

[solar torrent]

When you say "duct tape", I already know you must be joking. You would route audio of an entire tab of Google Translate to virtual line cable (Line 1) using a specific extension in browser, set input on W-Okada to Line 1, press record button, type anything into Google Translate and click on speaker icon there, and there you got the recorded audio.

[polar flax]

we mostly support rvc training

however it is STS not TTS

you can try using framepack that can run on 6 GB gpu/higher

[polar flax]

don't ask to ask and go to #✨│ai-help

[solar torrent]

However, if you wanna use TTS W-Okada with Discord or voice call, you would need another VB-Cable or full version of Virtual Audio Cable on this one, as it somehow needs at least two virtual cables in order to output that to another program, and this would be more complicated to setup than what you think.

[empty hazel]

The reason why I ask is because I've seen other people use like...Titan, I think it was? For model training and some other people use RVC while I use ApplioRVC's built in models for training

[solar torrent]

Anyways, Applio eliminates the needs of virtual line cable, as it has TTS built into it, and you can do TTS there with RVC voice models.

[empty hazel]

Yeh

[rain isle]

Hello,

I’m looking for an AI that chant a football anthem and simply replace the team name with another one, while keeping the original rhythm and melody.
It's Real Madrid one, i wanna change Madrid to Hortaleza

Does anyone know of an easy tool or method to do this?

Thanks in advance!

[chilly lake]

you need something that does audio2audio

example https://github.com/ace-step/ACE-Step

it is a big ass model

[sick heron]

what to do if the RES is large?

[dense ginkgo]

You need IA to convert it ?
Maybe do low tech : record yourself saying what you want in 20 points of the same room (microphone has to fixed position to get "crowd singing vibes") then do voice change in like 19.
Maybe this can work, crowd singing IA is so niche so i don't know if it performs well or not.

[dense ginkgo]

Deuteris fork has TTS ? Guess not,just confirming.
Applio is kinda heavy.
The duct tape comment is outdated but guess it could work if someone had no other option,like if the internet went out and people had to use what is in the hard drive.

[solar torrent]

What are you talking about? Detris' fork W-Okada never has TTS built in. Applio does.

This is my best I could understand, because what you say confused me a lot.

[calm wadi]

Sorry, what was the name of the voice changer?

[solar torrent]

W-Okada

[hollow dagger]

Goodmorning

[dapper drum]

Y’all think AI cover maker is gonna get a new update for the new models

[gray rover]

no

Unless you don't refer to " AI cover gen "

and in that case.. you can elaborate on which specific ' maker ' you mean

[mint summit]

how do people get such clean AI changed voiceovers for animated videos (zzz or genshin)

is it w-okada or something else? is it free?

[west quartz]

No it's not free bro

[queen kernel]

W-okada is free

[astral turtle]

Can anyone tell me the best AI tools for generating videos wirh less subscription. I am paying a lot to many websites but not getting the right context yet

[dapper drum]

Anyone know what I shoud use for in RX 11?

[chilly lake]

look at this mister money bags

[dusk mirage]

[supple stream]

veo 2 Google Ai studio
qwen 3 ect

[chilly lake]

there's no config.json file in the model's folder, this file is created after feature extraction is done

[polar flax]

skill issue

[dapper drum]

what we think?

[formal plume]

how I get a RVC to make AI cover

[polar flax]

please read the guide https://discord.com/channels/1159260121998827560/1301535386572427364 or explain more your problem in #✨│ai-help

[dusk mirage]

man called me noob with a simple line

[foggy beacon]

AI go boom

[foggy beacon]

Sounds like it’s underwater

[sinful mulch]

hi

[upper atlas]

https://www.youtube.com/watch?v=KRRQnIdNZx0&t=191s
can someone find the girl voice in this video

[ionic pumice]

is this the cover ? https://www.youtube.com/watch?v=82E86MBj35k

[minor blade]

That voice is probably unavailable.

It's very likely a private model

[cunning holly]

no ones joining

the event

oh nevermind

they finished it

[sick heron]

wtf

hahahahahaha

pizdec

[ionic pumice]

lol

oh uh

the dataset sounds muddy btw @dapper drum

[tiny patrol]

Hey!

[dapper drum]

https://tenor.com/view/surprised-face-gif-6506754539819949799

[hushed gust]

why cant i use gifs as model images on weights anymore (RVC)

[rustic atlas]

Hello, someone will help? How to change the photo into animation for free.

FREE

[robust geyser]

could anyone tell me if my ai cover sounds realistic? dm

[drowsy swan]

how can i download the software?

[minor blade]

You can download RVC/Applio by checking the docs.

-rvc

[pine acorn]

📚 RVC Documentations

AI HUB Docs

https://docs.aihub.gg

🍏 Applio Docs

https://docs.applio.org

[minor blade]

I suggest you the first one

[drowsy swan]

thanks mman :)

[minor blade]

You're welcome buddy 😄

[coarse valve]

I realized the same way, too, which caused my post to have an error when I added in a recent model I made.

[dense ginkgo]

if i knew i might be compatible with you,you need to adjust the settings to find something non robotic, not plug and play sadly

[molten acorn]

I discovered prompt to make ai crash

[late bolt]

what app do people use voice changers on now?

to put ai models on

[upper atlas]

anything like it here?

[dapper drum]

Which modules should I use? Rx 11

[ionic summit]

Hello everyone

When will the release of AGI take place?

[nova mist]

I am Heavy

[polar flax]

not before GTA 6 release

unless you explain what you mean

[ionic summit]

Gta6 will have AI as bot players

[wet drift]

Hmm

[craggy ice]

when will we solve world hunger?

god ai images look shit

5 years later and they still look like ai

[ionic pumice]

ai inbreeding 💖

[ionic summit]

By becoming a cyborg and existing in the digital realm

This world is nothing but shaping your soul

The second we can escape from this world and exist elsewhere such as a digital realm that is when true life begins

You come into this world pretty much powerful the second you start to get older and older you get weaker and weaker , many get sick

Doesn't make sense the whole point of life should be getting stronger and stronger, and more powerful

Merging with machines can cure all of the problems we face today

Fucking (sex), hunger, damage to the body and consciousness and much more

[polar flax]

just be machines

as they can't be hungry

[polar flax]

developing androids is more feasible

though they'd need to sustain and evolve by themselves

also the human consciousness is too complicated to research scientifically

[raven knot]

hi everyone

is anyone here into automations with ai & stuff?

[polar flax]

https://www.techspot.com/news/107956-openai-launches-codex-new-ai-coding-agent-software.html

currently available to ChatGPT Pro, Enterprise, and Team subscribers

[craggy ice]

oh good, more AI slop

"why do i suddenly have a bill from AWS for $20,000?"

[sleek valley]

any good girl voices ? i wanna troll my friends for laughs and giggles

[minor blade]

Check the #1175430844685484042 channel and test which girl voice fits your voice the most

[gusty heath]

Holy

💀💀

AWS

Anybody working in open source drag me in 💀

Lemme help ya people out

[solar torrent]

W-Okada the realtime voice changer.

[covert lake]

please elaborate in #1192011222023950368

[hardy mauve]

What are the parameters I should change to get the response faster in real time voice changer

[covert lake]

please tell ur pc gpu and show a screenshot of the program in #1192011222023950368

[modest horizon]

Hello guys

[weary temple]

https://tenor.com/view/mahjong-mahjong-soul-jyantama-ichihime-im-mah-bout-to-jong-gif-5894450925248783497

[maiden kestrel]

.

[quick gorge]

hello

Are any of you guys interested in Elevenlabs?

[supple stream]

wha

[weary temple]

mahjong

[queen kernel]

Hii

[ionic summit]

No it's time to evolve, biology is outdated

Machines, technology and AI is the answer to all of our problems

Make sure you are keeping your body healthy by exercising daily 7 days a week, get lots of rest, take care of your mind.

[peak pine]

Why when I use /find it don't responds?

[drowsy quarry]

Does someone know which is the best platform to sell projects and models?

[covert lake]

sorry but this isn't really a place for selling ai projects or models

[drowsy quarry]

Yes this is why I am asking for the platforms

[covert lake]

could you retry in #🔍│find-models

[peak pine]

It just didn't found anything that's the answer

[covert lake]

so it just said this?

[peak pine]

no it just said "not responds"

[covert lake]

if so, you can maybe try other ways
You can search rvc ai voice models at:

• #1175430844685484042
• In #🔍│find-models , 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/
• 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

[hidden grotto]

:wave: @covert lake, How can I help?

Available Commands:
• @weights find <query> or /find <query> - Search for RVC Voice Models
• /create - Create an AI Cover
• /image - Generate an Image

[covert lake]

have you retried? because the bot works fine on my end

[peak pine]

uhhh

[covert lake]

maybe restart discord

[peak pine]

Does it support " " in query?

[covert lake]

oh wait you're right

spaces seem to break it

i will report this to Weights, please try without using spaces for now

[peak pine]

[tulip gale]

how do ppl

use ai to create characters

singing

like mr krabs singin a song or smth like dat

[covert lake]

by using RVC, please tell ur pc gpu in #1192011222023950368

[polar flax]

that means yuh uh, not nuh uh

that you mean to evolve machines

not humans, genetically

don't expect anyone to research & develop on the advanced genetic evolution and mutants

[cloud sequoia]

How do I make an ai juice wrld song from my phone

[covert lake]

use #1192011222023950368

wassup everyone

[brittle lotus]

Hello

[covert lake]

how are u?

[brittle lotus]

So far so good

How about you

bro

[covert lake]

Im good too

[brittle lotus]

Hh

Where are you from bro

it's still afternoon here.

China

[covert lake]

ohh italy

[brittle lotus]

Bro i like pasta

Durian pizza... 🤢

Yk

[covert lake]

https://www.reddit.com/r/LocalLLM/comments/1kqz90h/i_trapped_llama32b_onto_an_art_installation_and/

[slender estuary]

So, when I talk to Gemini “Imagine if Ninja had a low taper fade”, it just went ahead and started generating an image, which is something that I do not want it to do. Thankfully, I cancelled it by going into a new chat and deleting the conversation.

[brittle lotus]

I just came across this video on YouTube, and it feels straight out of the future

https://youtu.be/MiKqprPSKx0?si=gGgZmTJUiauA1uDk

[tight current]

hi i'm working on project using AI predictive maintencae for electrical motors i have data from ADXL345 vibration for defrant faoult and thous data are so similar naw haw i deel with this problam

can any wone helpe me pless

[gusty heath]

I need a mentor , any experienced person to guide me 😭

[tight current]

i con help you what you need

[gusty heath]

Awesome damn appreciate ot

I m lost

I did all ai ml

Architectures

Basic of pytorch and like i know how all ai ml works now

But idk what to do

The thing is i cant do anything without a goal ,i need someone who can gimme a goal or guide me whats next so i can finally get into research or job

[cloud sequoia]

I can't send files 😭

So I can't make ai covers

[tight current]

in what you are going to ML to classify what

to use it

[gusty heath]

What do u mean??

[gusty heath]

Idk i wanna go to research

To create efficiency

In specific things in it

[tight current]

like developing the ML algorithms or just training the models to use it in deferent project like predictive maintenance

[gusty heath]

Both

Both are equally fun

[tight current]

now you need to start to understand whta is ML and thar type and haw that works than tray to train models of ML the developing of the ML models is the Lasst thing

[warm mural]

Is there any voicemod voices that make the voice stutter and glitch like some of the animatronics form fnaf?

[covert lake]

We don't support voicemod

We use wokada deiteris fork for realtime voice changing for calls

Tell ur PC GPU in #1192011222023950368

be sure to not use YouTube tutorials for realtime voice changers

@polar flax https://www.reddit.com/r/LocalLLaMA/s/dkInCRxpjZ

[drowsy quarry]

That's cool

[still sigil]

@proven ledge no advertising

[proven ledge]

Why

[still sigil]

server rules

[proven ledge]

I understand, but it is not abnormal for people to ask for support. Wouldn't it be great if you opened a separate channel for advertising?

[still sigil]

nope

[proven ledge]

I think you are a very toxic person and you are jealous of other people's achievements.

[still sigil]

server rules exist for a reason

[proven ledge]

Exactly, and there is no reason for this rule. That is why it should be removed.

[still sigil]

i don't make the rules

i enforce them

[proven ledge]

Okay, then what I am saying is only for those who make these rules.

[still sigil]

well until the rules are changed (if they are changed) follow them

[proven ledge]

I'll do

[polar flax]

advertising/promotions aren't allowed, period.

[polar flax]

any acceptable reason?

I don't think we're likely to consider it

[covert lake]

We don't allow any advertising it, we used to in the future but it's not worth at all.

[proven ledge]

I can't advertise. I'm very uncomfortable with that.

[covert lake]

This server is not made for advertising

[still sigil]

https://business.google.com/aunz/google-ads/

[covert lake]

This server isn't made for it

Look for other ways to advertise yourself

[proven ledge]

That's not an excuse. It may not have been done for advertising purposes, but advertising can be done, for example.

[still sigil]

the purpose of this server isnt for advertising

[covert lake]

It's the server rules, there's no need to keep talking about it

[proven ledge]

It's ridiculous to make rules for no reason.

[still sigil]

id like to see you join p*rnhubs server and start advertising there

every server has a purpose

and advertising isnt this servers purpose

[proven ledge]

I understand, and I'm not going to stay on a server where even the rules are pointless.

[covert lake]

@proven ledge this server is English only, don't speak Deutsch please

[covert lake]

The rules aren't pointless, you're free to look everywhere else to advertise your things, this isn't a place for everyone to just advertise everything

[polar flax]

this is like a community forum

we keep everything under moderation

[proven ledge]

There's only one thing we disagree on. I think the rules should be appropriate for their purpose and your purpose is absurd. This is my opinion and I expect you to respect it.

[covert lake]

@proven ledge please refrain from continuing this topic, that's our rules. No advertising.

The purpose isn't meant to advertise at all,that's all

Please don't self promo again, and respect the rules

[proven ledge]

Yes, not everything serves only its purpose. For example, does someone who governs the country only govern the country. Not everything is used for its intended purpose. Servers can also be used as an advertising tool for entrepreneurs like me. That's my opinion and you will respect it.

[covert lake]

This server isn't for entrepreneurs at all

And it's not meant to be an advertising tool

No need to continue, that's the rules

Just find another way to advertise yourself

[proven ledge]

I expect respect but I still get disrespected. There is no need to say anything to you. Keep thinking what you think.

[covert lake]

It's not an opinion, it's a rule of the server

If you keep insisting to promote yourself we will have to take action against it

[proven ledge]

I understand and I say it is an unreasonable rule in my opinion. You are still arguing about it. Anyway, never mind. You don't understand.

[polar flax]

btw you can share any open source project here: https://discord.com/channels/1159260121998827560/1359898289335566570

though it's not meant to advertise it

[covert lake]

That's for open source projects but not anything about advertising paid tools or other things

[still sigil]

he was asking people to sign up for an email ai thing

[covert lake]

Yup, for example that wouldn't be like sharing an open source project in #1359898289335566570

[still sigil]

watch him decide to advertise it there instead

[mild timber]

is Voice ai worth subscribing for?

[dapper drum]

no

[tranquil lichen]

please make made in abyss reg ai voice model

[drowsy quarry]

Can someone suggest to me a platform for selling projects or finding the bestest clients for AI projects?

[deep smelt]

i'm a verbal assassin

[deep smelt]

someone need lalals ai pro ??

[supple stream]

bro thinks rules are useless

[gusty heath]

I did all that