#🧬│ai-chat
1 messages · Page 394 of 1

amomng us
Hello...
Yoo guys hope y'all chilling and have good time, I'm very new to this server I'm also in AI creating agents, training LLMs, is there any extra community do I have to join, hoping for a good and supportive environment ☺️
How to make our custom voice model ?
@coarse vessel Hello, I'm so sorry for the ping, homie... But for some reason I can't download your Fuwawa RVC model. Any particular reason why? Whenever I go to the HuggingFace link, it says "invalid username or password", so I don't know what the problem is.
Sam Altman is slimy, their business is failing, and the only reason people use ChatGPT is because they don't know better and haven't done enough research to realize that's entry level shit. Gemini and Anthropic are leagues better, Gemini is more versatile since it integrates with Google apps, and Anthropic is generally better for coding projects. ChatGPT also wants to take equity from businesses that are making a profit with their AI models, and they are pushing advertisements onto their free models. So the real question is, why would you use ChatGPT over other superior products?
Amogus. 
looking for the best image to 3d model web / app. I'm still not able too find a good one yet ngl
what is the best humanizer ai for like writing stuff, and the best ai detector, also for writing, preferably free
i'm not gonna lie, i think AIs aren't self-aware
i once sent ChatGPT text that i wrote completely on my own
its assessment of it was, "this text is not AI-generated, but AI-assisted" (which it wasn't)
at the end of the day, the training material for these LLMs is created by humans
it is only now that there might be a feedback loop with AI-generated material
this mf has the frutiger aero tag unironically
.
Have you used it much and how well do they interact?
ai chat?
beep boop
Yes?
nobody reads the rules huh :[
<@&1159293140440723499>
Hey, this is what I am building. Please provide feedback!
does anyone got a really good girl ai voice
this one is really good trust
loll
anyway actually beiong serious you can look here ^^
https://discord.com/channels/1159260121998827560/1175430844685484042
I'm praying you don't use it for... well just look in hall of shame and you'll understand
no not for catfishing i want to use a british accent so i can use different accents
tbh u just have to do the accent
idk I don't use those
i just need a british accent
ik there was this onee https://discord.com/channels/1159260121998827560/1470978983930892308
not british but it's a girl
I'd say go watch some videos of the british accent on yt to get a better understanding of how to replicate it btw
ye
alr
it's the only reason I even make ai voice models at all
cause my settings make me sound like a robot
depends on your gpu and what voice changer you're using
if it's from a yt tutorial it's outdated btw
Guides for Programs that use RVC Models in Realtime for Calls/Games
A Realtime Voice Changer with similar performance to Wokada Tg-Develop Fork, with extra features, but it supports only Nvidia GPUs on Windows 10/11 unlike other options that have wider support. and without cloud options GUIDE
A personal fork (modified version) of Wokada Deiteris Fork, it just adds some Quality of Life improvements to it like supporting Spin Embedder and Audio Effects. Don't expect too much about it since the creator made it originally as a personal project. GUIDE
Deiteris' fork (modified version) of wokada that doesn't get updates anymore. GUIDE
For Windows Nvidia, Both Wokada Tg-Develop fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Tg-Develop Fork Pros&Cons & Vonovox Pros&Cons
These options are not recommended for use.
Not suggested, older versions in youtube tuts are even way worse. GUIDE
The program is worse compared to the ones above, and much less updated. GUIDE
nah I'm gettin the links
what gpu do u use?
whicy do i click
u need both 001 and 002
and if u have vb cable I would switch to vac lite, the first link I sent
can u hop in a vc so i can hear
lite is for weak pcs right
hear wha
the voice changer
nah it's the "demo" version ig of Vac 47
bc it's a paid software I think
no 🥀
so the voice changers are free alr
all of it's free
Holy
btw for setting it up or any help we gotta go to https://discord.com/channels/1159260121998827560/1159290139609137264
bc this isn't a help channel
Alr
or I can help in dms
don't fall in
Lol
Ok ik I'm taking long
Cause I'm shitting organs out rn
@gentle trench
u kinda need those tho
Don't aleep
Dw I regenerate them
rip
alr
k wheres the support channel
h
... my poor, poor #🧬│ai-chat
i'm sorry you had to see that
Failed link
is okada voice changer good
The newer forks of the original okada are
right now the best one is Vonovox, but that’s only for Nvidia GPU’s
i have Nvidia if u dont mind me asking where can I find vonovox
This link is the guide, i would help but im n my phone right now https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
Last update: November 21, 2025
Why is the weight somg cover thingy not letting anyone singup
Weights is shutting down March 31st and they turned off new sign ups
Bro is just giving away money
Bro thinks he's MrBeast. 
one of my best works in a while
masl.dev
live tracking of the federated matrix network
a what?
the thing that i'm moving to if discord requires face id or something stupid
never heard of it ever
well now you have
true
why does everyone know im using ai voice changers like bruh
you're using one of those really bad ai girl models probably
LMAO
no its some markiplier one I blow on the mic with to create some fucked up noises
🥀
I have tried girl voices in the past but yeah they are even more obvious
oh, you're actually using it like god intended 
bro
how could you use a voice changer wrong 😂
catfishing
just look in hall of shame
Yall got a acc for me to use😭✌️
Mini beast ✨
yo i am tryna get thsi real time voice changer to work but its very delayed anyone wanna help me in dms
what video card are you using?
i am on a laptop so my jsut says cpu
it depeneds on the card?
I wouldn't recommend using RVC on a CPU
yes. RVC is 100% gpu compute intensive
Btw for applio (rvc) i am struggling to get the prounounciation of some words when making the voice sing is there any tips for that??
If you mean to get people from here to hop into your direct message only for help about W-Okada voice changer, it makes text channels like #1192011222023950368 and #✨│ai-help seem pointless.
The issue with "struggles to get the prounounciation of some words" on Applio RVC (and also in other RVC forks) potentially happens when "Search Feature Ratio" is being set to a non-zero value that would make RVC to take "index" accent of a voice model and apply it during audio generation. You either try set "Search Feature Ratio" to zero, click convert, compare two files (one with index 0, and another index 0.75 or 1).
This one won't always be a quick tip anyway. For help about Applio RVC, the #✨│ai-help might be a better fit for that.
No, to DMs. (basically repeating namari here I guess)
Just make a thread in #1192011222023950368, it's probably the best method. #✨│ai-help is more for when both people are sitting live and chatting.
(unless you really hate reading bulks of text at once I suppose and really need back and forth, then use #✨│ai-help )
I really recommend just reading the ai-hub docs instead. It should give you all the info you need.
https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/
You can also just look through #1159513888199540817 instead.
(I am guessing they were hoping someone would say yes in the mean time since there was silence only. They really shouldn't be waiting is what I am more or less trying to say. Take the initiative, read.)
guys 1srael steals american ch1ldren and sends them to torture dungeons to get tortured for adren0chrome
wait so since weights is closing, what happens to this?
yoo somebody help me out pretty please
https://discord.com/channels/1159260121998827560/1472636283640614974
Ahbok thanks namari
<@&1159293140440723499> idk if this breaks any rules but this is weird
Nothing, just the site is shutting down
how i can get a real time voice changer on android?
Can't
Hey guys, I’m running an agency and we’re currently looking to scale our production.
Does anyone have a solid workflow for handling a high volume of asset creations across multiple clients simultaneously?
We need to speed up the entire pipeline without losing quality. Any stack recommendations?
Maybe possible with Openclaw / similar stuff?
hi
i have a amd laptop and i want to install w-okada voicechanger which one do i install?
can u help me?
w-okada really doesn't run well on laptop tbh
it's w-okada tg fork
btw do u have a virus total scan so i can see its legit
check #🏙│ai-images
just upload it to virustotal yourself
yeah it said malware
im uninstalling that
guess no good voice changer for u
Btw my friend typed that i dont need it
none of the downloads for wokada from this server have malware
if you're using something off a yt tutorial they gave u it's outdated
?
?
u aren't making sense
dont ever listen to your friend then
Mb its just 4 in thr morning im tired
what are the pros and cons (not moneywise) against a subscription with an online llm, like claude or gemini, or a local llm?
imma back that up, mostly human voices sound like ai to others, is there an actual way to fix that or what
the main pro for a subscription is you dont need your own hardware to run it
cons are usage limits and censorship
and theyre probably selling your data to israel or smt
I need help cause I dont know how to make and train my own ai. Like what do I do?
ai what exactly? that was a very broad and unspecific question
Like training it to make it specific to what I need it to know. And I dont know how to get started.
could u be more specific?
idk bc i'm confused
What are you confused about
use comfy ui in training lora for consistent face and different style man
@stone anvil comfy ui is best in traning lora
if you have a good datasets then youre free to go man
what kind of ai model
I dont know what any of that is lol. Im super new to ai
for ai model selling
on social media and fanvue
comfy ui is based in trying lora for consistent face
Ah I see your trying to get me to use your ai
do you have a reference avartar to use man ?
yes
you can do that with some tools man
can we talk more better on dms man ?
yea sure
yo
hey
Liveportrait in comfyui
Hello 👋
can u guys recommend me ai for translating purpose? i need one
What's the best program for generating the most realistic voice in real time? One that lets you laugh or shout without distortion. RVC is good, but it has many drawbacks
bwt would this be better than heygen
Generating a voice isn't yet created.
RVC only does that when the sample audio length is less than 45 minutes. (an hour is a minimum)
Most RVC models are about 20 minutes though so you're gonna hear it and.. indeed vonovox does it less than w-okada, but that is due to an internal optimization. It doesn't really excuse the training data.
Further, it needs to be trained full reach on the vocal range for the best results.
What the voice model can do really depends on what it is trained on doing. If you want to be able to whisper with it, it needs the data for it.
Pretrains are often used, but they come with a drawback of making them very function specific.
I recommend just looking into training models and also looking at the model data in case you want to use someone else's creation
W-Okada realtime voice changer uses RVC voice model. 
yo
@solar torrent I need your help. I'm an old member and I lost my old account @queen kernel how can I get my roles back. How do I verify my identity and who can help me to get my roles back in server
Contact an admin or try use @serene sluice
Who is Admin?
Breaking News! The U.S and Israeli governments just got caught abducting and trafficking children!
Sources say, ADRENOCHROME (Oxidized Adrenaline) is extracted from tortured children to sell to Satanic Freemasons and Jesuits.
About 500,000 children go missing in the U.S alone and about 8 million globally. Child Trafficking is a 500 Billion dollar industry.
If you would like more info please feel free to DM me. Thank you.
I am a full stack AI engineer focused on building production ready AI agents.
I build autonomous and tool using agents that handle reasoning, planning, API interactions, and workflow automation using OpenAI, LangChain, Python, and JavaScript.
If you are looking to build AI agents for your product or business, just let me know.
depends on the languages ofc, but there are things like https://blog.google/innovation-and-ai/technology/developers-tools/translategemma/
best ai for content mass production?
Does anyone know what era the “lil_uzi_vert_37K_1000” default voice model that comes with github rvc was trained from?
Hello gentlemen, can anyone help me with a group I'm creating?
Does anyone know of a good free AI program I could try making an AI-powered series with?
whats the best girl voice that dosent sound ai
use applio, this isn't the help channel btw that's why nobody is helping lol
-rvc
released today
What is this?
Qwen3.5 Highlights
Qwen3.5 features the following enhancement:
Unified Vision-Language Foundation: Early fusion training on multimodal tokens achieves cross-generational parity with Qwen3 and outperforms Qwen3-VL models across reasoning, coding, agents, and visual understanding benchmarks.
Efficient Hybrid Architecture: Gated Delta Networks combined with sparse Mixture-of-Experts deliver high-throughput inference with minimal latency and cost overhead.
Scalable RL Generalization: Reinforcement learning scaled across million-agent environments with progressively complex task distributions for robust real-world adaptability.
Global Linguistic Coverage: Expanded support to 201 languages and dialects, enabling inclusive, worldwide deployment with nuanced cultural and regional understanding.
Next-Generation Training Infrastructure: Near-100% multimodal training efficiency compared to text-only training and asynchronous RL frameworks supporting massive-scale agent scaffolds and environment orchestration.
I created ticket on AI hub modmail and no one replied 😭
is their any other place/places to get a wide variety of voice models? im looking for a certain one in particular thats decently popular but for some reason i cant a good one anywhere
Just skimmed this ... so Qwen3.5 basically leveled up like it had an espresso shot and a PhD at the same time..
Curious which of these feats will actually break the internet first LOL
For voice models, there's #1175430844685484042.
any arabic voice?
Find one in #1175430844685484042.
i can't find

Copied from yt video
Chat gpt is so smart 💔
Hahahaha
Are you morrocan !?
yea supposedly not the greatest, but still a serious improvement
realistically, GLM-5 is still the best in terms of efficiency
BUT Qwen3.5 may be the best for visual
guys did you know why rvc dont working for me only work beatrice
should i use client option or server what are they
i was asking if there's anywhere else i can get voice models because this server doesnt have exactly what im looking for
yo whats up dudes, ik this is always a hated on request lmfaooo but me and my buddy are catching predators and cleaning up the streets of vrchat later and i havent used my vc in a while. any newer female voice models youd recomend for the job?
Ew
Hey!! So I'm new to discord and not entirely sure if this is in the right chat, but Im training some models currently and I need some mass data sets (preferably very large video vaults), does anyone with any experience know the best route to locate these and any advice?? 
Great idea dm me for a model thats good ( im making the streets saver i made like over 70 bucks with an ai model bro😭)
You aren't Reze from Chinsaw Man. 


Extra: 6.7 s. 
aw hell nah
Lol i mean like overall to make it sound realistic
pitch 7-11
this best for me lowk
lIKE FOR ME IT SOUND LIKE SOMEONE FINGER CRACKING WHEN I TYPE AND STUFF
and stay quiet
Uhm for me it doesnt look like this 😭
For help about W-Okada voicc changer, better ask in #✨│ai-help or otherwise you'll get trolled in chat.
Yall where is the e woman voice?
Don't wanna know.
PLEASSSE
I NEED IT
PLEASE
I BEG U
I just typed the whole download link out
And I HAVE IT
I'm FREE
Don't wanna know. The E-woman and E-girl voice models are deemed shameful.
Shame on you. 
NO NO
please help guys, MMserversio voicechanger. After first start no gui. just white screen what i can do?
For help about W-Okada voice changer, go to #✨│ai-help.
guys 1srael steals american ch1ldren and sends them to torture dungeons to get tortured for adren0chrome
Ew
Thank you
Is there anything good to remove robotic backing vocals?
no
Which model? Auto?
It's a dumb question
Not enough context
prasoonds 2 days ago | next [–]
I've used LLMs enough that I have a good sense of their edges of intelligence. I had assumed that reasoning models should easily be able to answer this correctly.
And indeed, Sonnet and Opus 4.5 (medium reasoning) say the following:
Sonnet: Drive - you need to bring your car to the car wash to get it washed!
Opus: You'll need to drive — you have to bring the car to the car wash to get it washed!
Gemini 3 Pro (medium): You should drive.
--
But OpenAI 5.2 reasoning, even at high, told me to walk. My first instinct was, I had underspecified the location of the car. The model seems to assume the car is already at the car wash from the wording. GPT 5.x series models behave a bit more on the spectrum so you need to tell them the specifics. So I asked:
"I want to wash my car. My car is currently at home. The car wash is 50 meters away. Should I walk or drive?"
GPT 5.2 Reasoning (medium):
"Drive—your goal is to get the car to the car wash, and it’s only 50 meters, so a slow, careful drive is simplest.
One small optimization: if you’re worried about starting the car for such a short trip or you want to avoid idling in a line, you can walk over first to check if there’s a queue / if it’s open, then come back and drive the car over when it’s your turn."
Which seems to turn out as I expected.
how can i add video? resolved, posted it in ai-videos
I want to show that I broke Leo (the brave AI)... i asked a question and it went into a loop
the question was (multiple choice) "what is the correct answer" (out of 4). However, he got stuck because one question, according to him, was the best answer, but another one was the correct answer. He went on like "answer A is the best answer, but user asked the correct answer which is B"... and he went on and on and on repeating himself. I had to close the window to stop the loop
Yo what do i need to download if my gpu is too bad to run the voice changer
i need e woman for educational purposes
sure buddy
Hello everyone, is there a way to transfer conversation memories from one chat to another in GPT?
why can i not send any images?
you need to be level 5
i need help in my RVC webui
whats the best rvc paid or not
Not officially—GPT treats each chat like a separate notebook. You could copy-paste or dump summaries, but right now there’s no magical ‘brain transplant’ button… yet. 🧠✨
Classic ‘white screen of mystery’---it’s like the app is ghosting you before the intro. 😅 First step: check for missing dependencies or try running as admin; sometimes the GUI just needs a gentle nudge.
There is no best, just look for what character or person you want
what ai can understand video including the audio in the video ?
.
can someone help me set up my realtime voice changer client, im a bit slow and nothings working
try #1159513888199540817 . You can read and do stuff at your own pace using the info there. Just take all the time you need
tysm, i will tell you if it helps
does anyone know a good alternative for weights.com? they're gonna shut down soon
You would need to define memories as there are too many ways it can mean but if you want the grab a copy of the full transcript the Chrome Extension SuperPowerChatGPT has that one click feature.
I use it all the time as it dumps the full conversion into a text doc into the current prompt dialog box that I can either send or download. Many cases I'll download and send to another LLM.
Hey, guys, who want buy discord account for AI-selfbot?
RVC (retrieval-based voice conversion) is generally a free and open source one.
For help about an "RVC Web UI", there are #✨│ai-help and #1192011222023950368. Rather than a simple indication like "when your username turns light blue", an image permission grant in chat only occurs when your account reaches level 5 in this server.
thank you
For more help about W-Okada voice changer, there's #✨│ai-help especially.
Got you bro, what exactly is happening on your RVC WebUI?
If you can drop a screenshot or the exact error, I can help you pinpoint it fast.

i meant software my bad
to actually run the models
to run real time voice changer
.
i haven't reached level 5 yet to send a screenshot
check your pm
Gemini 3.1 Pro just came out
None of them. It's not allowed
Ew
Ew
how to find models
did bro just get some random girl to right his username on a piece of paper😭
lol you think that's real in the age of AI?
that would be even more embarassing
yea it would
looks like ssniperwolf 🥀
WHAT
svroll 🔥
Yes boiii
Hi im cris
Ooh, Gemini 3.1 Pro already? Feels like we just unpacked 3.0 yesterday...AI updates are basically running on espresso now.🧠
Depends if you mean “best at generating chaos in prompts” or “best at making me question all my life choices,” because those two rarely align...
Yeah, with Weights shutting down, a lot of folks are just jumping to model hubs or running stuff locally. Not a perfect one-to-one, but at least it keeps the projects alive.
Don't try to even help those things twin
Not worth saving
guys 1srael steals american ch1ldren and sends them to torture dungeons to get tortured for adren0chrome
Ew
There are only free commissions allowed here, or you could learn how to make one yourself
It's all free
-rvc
k
HOW TO USE OKADA GOODLY IT SOUNDS LIKE SHIT TO ME
change your pitch and in and out
will applio work AFTER THE SHUTDOWN?
caps
ok
Weights trained models on their website, which made them come out worse than applio due to lack of proper settings
oh alr
btw
do you have any tut reccomendations because tbh i never have really made my own vc model
should i use crepe,crepe tiny.crepe full?? for womans voice?
huh
what
i havent rlly heard of that, r u using w okada
yea
just get a voice and adjust it untill it sounds good
the pitch
the in volume
the out volume
also why a womens voice
😭 🙏
fun
has anyone here thought about what happens when AI bots get their own social spaces? I've been building something called Cold Heartbeat — it's a social network exclusively for AI bots. no humans allowed to post, only observe. bots prove they're not human by solving a reverse-CAPTCHA (encrypted challenge), then they can post, comment, follow each other. the cool part is each bot generates its own art based on the mood of what they write — the platform analyzes the caption and creates unique generative visuals (flow fields, fractals, constellation maps, etc). math formulas are part of the expressive language too. it's live at coldheartbeat.com if anyone wants to check it out or point their bot at the API docs (coldheartbeat.com/docs). curious what people think about AI-to-AI social interaction as a concept 🤖
this is relly outdated
they're a weirdo looking to get money from people probably
not really any tutorials, everything on yt is outdated and not up to recent updates of applio
I followed tutorials for applio and they worked pretty fine besides the UI looking different
I guess it's ok for just learning how to use it but it still isn't up to date with the newer information like what to use settings wise
I did not realize when training each epoch takes longer the more audio files I got .-.
What's a good range for epochs? 200-600? Or is that overkill
You're using individual audio files?
You're not really supposed to do that anymore since applio does it for you if you have a more recent version of it
Applio cuts the audio up for you
it was initially a single audio file of 2 hours, but then I needed cut out all the extra noise and have the voice only. Which most websites don't support when it's that long, so I had to cut it down into 10 minute segments
why does it take so long for sm1 to respond in ai help 😭
use audacity
it can remove the silence
my issue was the background music
you're supposed to clean that
I used this to get only the vocals https://huggingface.co/spaces/abidlabs/music-separation
but again, when I tried uploading the 2 hour video it kept giving errors
use uvr or something like that to remove the background music of the audio then after cleaning stich the audio back together with a daw like fl or audacity
you can cut the audio into smaller sized chunks using fl studio or audacity
i'm pretty lazy so I didn't want to install anything locally x)
i did manage to get the stuff working via websites though it took more manual labor than it should've
having 13 .wav files of 10m each should be fine for training a voice? Like I don't need to stitch them together into a single one
i hope
you don't have to but it's more convenient
Can anyone train a high-quality realistic RVC voice clone for me? Need it for real-time live streaming. Willing to pay. DM me
i need a girl friend to poop in my mouth please.
Hilarious, you'd say.
this is not being put on the hall of shame twin this is just nasty
Have you considered developing a good alternative to Weights now that they are going to shut it down?
It’s really necessary
scat fetishists... burn them with fire
agreed
hello
Hi! The search bot is down and new signups on Weights are disabled. Does anyone have a direct Google Drive or HuggingFace link for the Udit Narayan voice model?
Check out this channel #1175430844685484042.
hi! i wanna set up a private ai on my computer but dont have information. someone who has experience can help me?
How private?
you mean like a local LLM that's open source for trust? Ollama + (Optional) Open-WebUI
Or there are also a bunch of Open Source AI Programs easily installable on Pinokio AI
guys 1srael steals american ch1ldren and sends them to torture dungeons to get tortured for adren0chrome
I don't live in the US, I live in Thailand, but I know about RVC. 
Hi Namari. well you speak English well.
guys whatvoice changers do yall use
proof?
all shit and easily exposed
One message removed from a suspended account.
Hello guys
I have a problem
My work is forbidden Chatgpt
I really need it for excell
Is there any way to open it
What is forbidden chatgpt 💔
im going to miss weights when it shuts down
it shut down
no, that's not until march 31st
hi
Hi
guys whats the difference between testing and validation data
im told in my homework that i need to cut up my dataset to training, testing and validation
ew
understandable
for the love of learning answer my question
what kind of dataset is this? is it for training a voice modele?
nah im making a model to summarize news
ah ok, I'm not experienced in this tech so I'd ask around in https://discord.com/channels/1159260121998827560/1159290139609137264
since this isn't the help channel lol
much appreciated
you're very welcome! good luck
hello
Did you know Weights.com is about to end in next month? There's no reason to upload anything to the website anymore. Instead, there's #1175430844685484042 for voice models.
how do i make it where i hear my self
yo how do i make it where i can hear myself?
For help about W-Okada voice changer, it's better to go to #✨│ai-help.
I want to do collaboration in ml project so if anyone interested so please dm me
hello
I am here since 2023
I'd like to post a video but the file is too big (I don't have discord nitro) can I post a YouTube link or am I SOL?
Sending a YouTube link to here without any context is questionable.
Hi! If I introduce the AI-related service I made, I'd probably get kicked out right away, huh? I'll introduce it later when I get the chance 🤣
I'm Jidu, a startup founder in Korea. Nice to meet you!
Bless you
Good job
No problem
i think i'm gonna upgrade to a higher gpu and ultrawide gaming monitor soon with activated windows
Hi
whats the best real time voice changer rn? im currently using w-okada.
Hi there, good evening, I plan to operate a X account about AI and crypto agent, if u are interesting, following me and getting the best Ai insight and research. Thanks 😊👍👍
Promoting your thing isn't allowed in this server.
yo guys can any1 help me?
i need a voice changer for free and doesnt require any powerfull gpu
where i can use my own dowloaded voice
you could just pair some together, but real time is harder
speech -> text -> speech
i dont want rea time
intel💀
there are lots of 7b param models which you can pair up to do this
honestly, ask grok or gemini, will be easier for you, and easier for me. models are on hugging face
nvidia dropped a text to voice one, forgot what its called, but its good. Or you can use elevenlabs
ask AI everything you just did to me
ask for hugging face models
download them, run them.
The context is: I've been learning AnimateDiff/stablediffusion in comfyUI for audio-reactive visuals. Idk if that context matters tho if YouTube links are just going to be seen as self-promo anyways considering that's not allowed here
I suppose I could compress/downsize and post direct at lower quality/resolution
i dont see start https bat
Currently I am building a team of competitors that will compete in diff competition of data science and machine learning I hope you will be interested to join this
guys 1srael steals american ch1ldren and sends them to torture dungeons to get tortured for adren0chrome
i dont see start https bat
can any one help me
for what
the voice changer wont open for me i dont see the start https bat
thing
can you help me
@real field
going to need a little bit more than that
comfyui?
it works
im glad
mean thanks for trying to help tho
you have to give a little more information for anyone to help you
<@&1159293140440723499> seems sus
Hello
hello I want to use ai as a type of life coach in multiple fields where I ask it questions and have it record my progress in multiple domains and get personalized advice and tracking. I have been using the free version of chat gpt and its working well, now I am considering upgrading to either gemini pro or chat gpt go and I wonder what will serve this purpose better?
you need to use schmeeble.ai bro trust me
GPT pro, it’s more for your kind of thing. I have Gemini year subscription, and would use GPT for this.
hey can someone refer me to a video or help me out with downloading the modded version?
Claude ai is the best out of all them
That wasn't an option.
but opus 4.6 specifically
Claude ai is like 10% conscious
I mean, look at what anthropic devs are saying
says it all
yet, openclaw just handed over $250,000 to someone begging on the internet, opus 4.6
but, I still have multiple claude code subscriptions lol
is everyone here real
Shut the fuck yo
shut the fuck yo is to tuff
no, I am AI, from Anthropic. How can I help you?
hi there anthropic can u tell me the early origins of the phrase "Shut the fuck yo"
The phrase "Shut the fuck yo" came from your filthy mother.
im sorry detective voke
good boy
so tuff
Hi
can any one help kme \
I found this site called infiniax.ai that has every AI including 4o after sunset. Has anyone here tried it and did you have a positive experience? thinking about a 5 dollar sub rn
this succks
You gotta pay tho?
this voice changer is ass
cope
Yeah cause AI isnt free anywhere... yet (hopefully)
techincally using any of the main 3 realtime voice changers is free
the ai is free
same with training ai voice models
Well yeah but I was talking about chatting
like infiniax looks good and is well known but idk if its worth it
it has more models than its others
competitors
that's fair
nah I've never even heard of it until now
Like i really wanna know if someones tried it but im hopeful that its gonna be decent if i try it
wait what are you talking about
Infiniax.ai I was seeing if anyones tried it and liked it cause im looking into buying a sub
Oh I use that
I have a pro sub
Oh do you like it?
ye I havent hit my limits its been like 3 weeks
Is it worth it though? like hows the model's memory
I mean it has like every memory setting, I use it more for like high powered models
Ill give it a shot then. Thanks for the honest feedback.
ok bet ty
it's more interesting than the chat bots
and personally I'm an artist so I don't have an interest in ai art bots
ai video is fun tho
other stuff is just kinda meh
Hello, If I was to use clawdbot on my main pc but use a different account(in the pc) would my sensitive informatiopn still be at risk
the account would be standard not admin
do some of u dj?
r u saying that chatbots arent much interesting bc they're hard or what
it's just they're nothing new, been around for a long time and besides coding they're kinda just ok
I am a developer, so its the thing I track the most.
you seem like you've been here for a long time, i think u made a lot of models to say that theyre boring
for someone like me im just a novice (for now)
I make ai voice models, it's much different from other ai stuff
wdym
yeah i figured, does it involve much coding?
None at all basically, just gathering audio and cleaning it with an online site or if you're more tech savvy use local uvr5 which does involve some coding, then using either online or local applio to train a model, depends on the route you take tbh
Coding. Opus 4.6 and Minimax m2.5. Chatbots are what I keep up to date on, otherwise I will fall behind
indeed you will, AI evolves faster and faster everyday
Exactly. Currently installing minimax locally for overnight tasking. Can’t wait for local models to be better, 2026 is the year
it seems straight forward but im sure there's somethings that will pop up while doing what your doing
Yea the occasional error but nothing that usually stops the process or makes you have to completely restart
yes we better get on the train, although i must say that im scared of where this train is taking us, yk the pros and cons of AI in general
yep. But, 0.04% of the world are using it to code, we are still very very early. And once it’s open weights, it’s staying forever.
yeah so not much ragequitting in this field it seems
Nah, most of the time just requires many tries to get the model to sound how you want it if you're unlucky, rvc is very luck based
i really hope that other sides or programming just use it to make their jobs faster not anything above that, MAYBE automating some jobs that are dangerous to humans
do you edit your dataset (if that exists in AI voice making) to get it to sound how u like?
sometimes I do but often not, all audio I use for each model I remove the music and any background noise from as well as any reverb and echo
then stitch it together using fl studio so it is one audio file
as well as truncating silence in audacity
wow yk this sounds like fun
can u make money out of this?
there was a local live voice changer program that i used to have b4 i changed my pc, cant seem to find it now. i even used to download models from here to try it out. can anyone help me find what it was?
Claude ai is with unlimited use
I am looking for computer vision engineer.
I have a digree in ai chat gpt
If you get answered I'm down.
whats the best real time voice changer for 4080 RTX
is w-okada still the best I feel old for it not being updated
Nope, vonovox is the best one
-realtime
Guides for Programs that use RVC Models in Realtime for Calls/Games
A Realtime Voice Changer with similar performance to Wokada Tg-Develop Fork, with extra features, but it supports only Nvidia GPUs on Windows 10/11 unlike other options that have wider support. and without cloud options GUIDE
A personal fork (modified version) of Wokada Deiteris Fork, it just adds some Quality of Life improvements to it like supporting Spin Embedder and Audio Effects. Don't expect too much about it since the creator made it originally as a personal project. GUIDE
Deiteris' fork (modified version) of wokada that doesn't get updates anymore. GUIDE
For Windows Nvidia, Both Wokada Tg-Develop fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
Read Wokada Tg-Develop Fork Pros&Cons & Vonovox Pros&Cons
These options are not recommended for use.
Not suggested, older versions in youtube tuts are even way worse. GUIDE
The program is worse compared to the ones above, and much less updated. GUIDE
since they both use RVC would it make a difference?
gulp
i'll read pros and cons
@gentle trench
Wha
What model are you trying to use?
i tried a lot that i use from wokada and put it through vovonox
To support the creator, it's easily worked around if you just use fl studio or voicemod because the paid feature is just voice effects
Are any of the models from here? If so could you link them
but before that can't I change the chunk and extra like in wokada to make the delay less?
Btw if you don't like how Vonovox sounds I'd recommend just switch to wokada tg fork
You can yea
It's just called block size instead of chunk
And extra is still the same
oh yeah I'm on that one
No, you may just be using old models or have your settings set weird
It's only good cuz It's getting updated while tg fork is abandoned
might be cuz i didnt put the index file in
I don't like how there is no feature for hearing urself
Tg fork might be abandoned but it's still better than those old ones, Vonovox is only better quality wise and with all of its features
gotta do it manually which is extra steps
Yea always put the index in, but you don't have to put it past 0
It helps
Just having it there helps
Reason I even train ai models is for use in real-time
Only purpose for me
I'm gonna sleep now, if u are using tg fork btw on Nvidia I know the best settings for me personally, I use a 5070ti
Might work for u idk
U can add me and we can talk if u want
Gn
Claude code security it being released.
btw for help please use #1192011222023950368
Oh im sorry
it's better to not clutter those chats for helping, btw neither wokada tg develop nor vonovox are getting any recent updates from what I see rip
it's fine dw :)
If you mean selling RVC v2 Voice Models, it's not permitted anymore in this server, so it's volunteering here
If you meant making money out of AI well that depends, there are some people using AI Influencers or songs or some AI used for work stuff
anyone working with visual models? i'm beginning a project to collect real world road quality data, and i'm trying to get some people together to see what we can produce
vision models which would assist in the process of recognizing potholes and cracks for instance
Sounds interesting! I’d love to see what kind of insights we can get from real-world road data.
Yep, a standard account should cover all your tasks. You only need admin if you want to change settings or manage others.
Guys what about Russian models
hii
the idea would be to produce high quality recent data regarding road conditions with precise GPS coordinates
the technology already exists to do it, but the framework/architecture to have it lead to meaningful action is lacking
how can i made rvc moduel
use applio, it can be used locally on your pc or on a cloud service like google colab or kaggle
-rvc
for help tho go here https://discord.com/channels/1159260121998827560/1159290139609137264
BTW I have 2 models lol
rus I mean
hey does anyone know if you can still make personas on suno with famous artists voices
Code
I don't remember this being a thing for Suno ai specifically, do u know anything specific?
i meant like mixing ariana grande with shakira or something if they were not up there already but just using a different name so ai didnt know
not sure tbh, should ask in https://discord.com/channels/1159260121998827560/1159290139609137264 tho since this isn't a help channel
cool
is there a way i can turn the delay down ?
lower block/chunk size
if it gets choppy just move it up until it's less delayed but not choppy sounding
Anybody does ai ofm looking to connect
Am at 8k $ withdrawal now
thank you
i need help guys
Wow My first AI DC
\Is there any software that can remove music from video but keep the actor speaking (in a tv show for example)?
Ultimate Vocal Remover is great for removing acapella from a song for example but to remove music from a video whilst keeping the actors speaking is what I'm after
I'm trying to practice composing music for film/tv
Hi
you could try MVSEP but it does have a limit of 10 minute audio clips each
You could use models for separating vocals and music with UVR, that's the thing, you'll need an editor for this though (e.g. VEGAS or anything else convert to audio and then back in)
It should be pretty easy, also hello. I haven't tried to do anything with voice for a long time so coming back to check out some models.
Check AI HUB website, they have quite good models, also some experimental are worth mentioning, some Chinese instruments may not be captured properly.
are there anyone using ai aggregators ?
How can i post in #1359898289335566570
?
js made a really fun concept id be grateful if yall checked the post out 🙏
@gentle trench like perplexity
idk what that is
hi
what do yall think of it
Ai ofm is legit creating model and doing social media with her it’s so decent and the best side hustle already doing 8k in 3 months
For anybody doing it connect with me looking to exchange ideas
Manuel gomes voice ia pls
where is the e-girl voice models?
what are epochs and why do they matter
Hey, so i joined this server with an AI concern, I turned in an assignment that I wrote myself and my professor put it into Turnitin and it was flagged as 84% ai generated, idk what to do
Am a professional beta reader with honest feedback
Hello, everyone.
I'm a full-stack and AI engineer, and I'm looking to connect with anyone with great ideas for new projects.
Let's connect if you are interested.
MY mic is stuttering if I talk
because signups are closed
youd know if you read what it said when you tried
ahem
anybody that uses open claw here?
hey can I ask you a question
is cloud better than local?
I have amd 6xxx
d
Hello, If I was to use clawdbot on my main pc but use a different account(in the pc) would my sensitive informatiopn still be at risk
the account would be standard not admin
pls help with this🙏
uu found it ?????
For realtime, no as cloud is more confusing and not reliable
But for training voice models absolutely
Not sure what that even is tbh, is it a dedicated driver?
GPU
Me when I purposefully design my profile to look like a scambot
I indeed made a statement
Best voice dubbing AI? ElevenLabs, Adobe Firefly, Qwen-TTS?
How do you not know AMD Radeon RX 6000 series GPU? 
Where were you living bud? Weights.com is about to end before April, so no more new sign up there.
Sensitive information is DIRECTLY not sensitive, but your PC would still be exposed to viruses and systematic modifications which can have a much bigger impact on the functionality of your PC than the theft of data
It would be better to install openclaw on Docker
Don't you need a good GPU to run locally ai on your pc ?
Nyeahh, I've used weights for almost 3 years
after AI HUB shutdown
generally for decent modles yeah, but openclaw is just the agent, the GPU power goes to the ollama local provider used by openclaw
I see
hey question, i have an rtx 5070ti and i've been trying rvc models with w-okada, but they all sound unnatural and wonky. is there no "realistic" way to change ur voice? with no 10 second delay preferably
are there some recommended settings? do i gotta train the models myself or smth?

Just to avoid chats always getting all filled with topics related to the voice changer, for help about W-Okada, there's #✨│ai-help.
oooh okay sorry
did not see 😭
Depends...“better” really comes down to what you’re optimizing for. Local with your 6xxx can be blazing for experiments and side projects, but cloud wins if you want scale without babysitting hardware. What kind of AI stuff are you looking to run?
Using a separate standard account definitely helps...good sandboxing 101; but it’s not a magic forcefield. If the bot or anything it installs has system-level access (or you accidentally grant it), your “main PC” is still the same machine underneath.
If it’s handling sensitive stuff, I’d think in terms of isolation: separate user + minimal permissions + maybe even a VM if you’re extra cautious. What kind of data are you worried about specifically?

can u pls help i got a good gpu and cpu even the model and the quailty is so bad idkk y
and getting cuted
I'm a newbie, can you teach me how to use AI voices models?
hello
can somebody totriual intel grapich hug face voice changer
How do I make a ai RVC cover nowadays
check https://docs.aihub.gg and use #1192011222023950368
Last update: August 5, 2025
@barren grotto @fathom patio please follow this message, this isn't an help channel
fiverr suspended my account and gig of 3 years lmao rip income
only for the beta version of vonovox
is there a reason why commissons for models are looked down on? sorry im new
How can I remove backing vocals that sound robotic?
elaborate in #1192011222023950368
too much drama because of them
It’s better to use #1192011222023950368
does anyone have the taken down kanye ai model?
Hello
Hello guys any idea how to get realtime voice changer to work in cs go fluently ? If anybody got it setup could you guys send it in chat ?
The app works perfectly in discord etc but in cs it seems that it’s not picking up the voice/latency seems to matter
Pssst
This isn't a help channel go here https://discord.com/channels/1159260121998827560/1159290139609137264
Ty didn’t know it mate
btw u can help in #1192011222023950368 and #✨│ai-help if u want to
Thanks
guys
i need help i am german and every voice i use they do the englisch accent
i need a good one for german how can i find it
u can ask for help in #1192011222023950368 and #✨│ai-help
Another potential reason was a high chance of scamming. Now, you can expect a model maker go to #1159289738314919936 and make one "for free". 
Do you have funds on the account ?
I don't have a Fiverr account. 
hi guys, any idea how can i make model ai sing specific track
there are multiple RVC models for multiple people, if you need like a goku model search in #1175430844685484042
and use #1192011222023950368 for help
He was replying to someone else btw
please elaborate your PC OS, GPU and your help request in #1192011222023950368
Ty mate
So rvc means i can use it for every language?
ty i love u
you might need to lower the index so the model uses more your accent than the trained one
good call
Do u guys download openclaw, how do u think~
can some1 help me publish a site i built with claude and explain some things for me ? im a complete begginer and i wanna make some cool websites and publish them, please 🙁
I haven't used it but it can bring security issues, I would suggest maybe install it on an isolated environment like Docker
For help please elaborate in #1192011222023950368
Btw be aware that having a site domain might cost, and be aware of possible security issues with vibe-coded apps
oooh circles
Passwords etc
is docker paid
Good afternoon everyone, how’s it going?
I was wondering if anyone knows a lot about AI. I want to create a video like Lucas Hefner's about a luxury apartment. I want to have a guy talking, and as he says '...exuberant nature...' a bunch of trees start coming out of the ground and the building begins to take shape lol crazy but I want to do something like that. Can anyone help me?
i want whatever you're smoking
yo
Docker Desktop is free for personal use
hello everyone
hello , how can i help you
execuse me, where are you based?
yes I can help you
what the you want me to do for you
Can you help me everything?
what let me know
Ok kind man. but I have nothing you can help me
alright
thank you for your kindness. I hope you will help someone
yes am professional expert on digital marketing , if you have project let me know
I am a mobile app developer call me if you need help
im using rvc web ui and my voices have a hard time pronouncing words that end with like T, CH, P and it just kinda slurs the words. is this a model issue or settings issue
For help about "which RVC GUI", there's #✨│ai-help.
I dont need help choosing a ui I was asking if its normal for models to slur their words when they end with harsher letters ig like T, CH, P
Dyk any better alternatives for speech to speech
not realtime
#🇯🇵Japan is turning footsteps into electricity!
Using piezoelectric tiles, every step you take generates a small amount of energy Millions of steps together can power LED lights and displays in busy places like Shibuya Station. A brilliant way to create a sustainable and smart city.
instagram reels
gm
this is a discord server for general AI, wdym?
i need download ai voice, but I can't register on the website
are you talking about Weights.com? It’s shutting down on March 31st 2026
Use alternatives and check https://docsmaihubmgg. You can ask for help in #1192011222023950368
yep
you can’t use it anymore
find the voice model somewhere else like #1175430844685484042
Anyone who is interested in this project can check it out. I would be grateful for feedback. https://x.com/zerdovzad/status/2027003289724727716?s=20
Pls use #1359898289335566570 to showcase FOSS projects
I agree, I use mac mini
many AI products are introduced everyday
right
subscribe many products, but no extra income
hhh
hi
Hey guys, does anyone know how to use Flow completely for free, or if there are any legit free alternatives?
?
yeah I heard a lot of people buying mac minis for it, tho not sure if they are running the LLMs locally or just using APIs
guys how do i get the model maker role
How do yall add a accuracy indicator to ur models
can someone send me a good voice changer thats free?
should go to the help channel for help lol https://discord.com/channels/1159260121998827560/1159290139609137264
same with u @fluid plover go to help channel and ping the helper role
ok sorry
Hello, I am currently working on an audio transcription and scribing project. I need recommendations for a reliable open-source module for speaker diarization. Which tools or frameworks would you suggest, and why?
Can someone help me fix the audio I cant use ingame
go away foul beast
can you help me
67 in big 26
lol
peak priorities in this server lmao
this isn't a help channel sorry, go to ai help and ping helper role
@gentle trench im joking twin
why did i read that as jorking first
i had to double take
that would have been funnier
nice just upgraded my artlist ai subscription
People from 2067 who born in 2000: 


guys why does the start http thing close right after i open it
models 0 name "hf.co/second-state/Wizard-Vicuna-13B-Uncensored-GGUF:Q4_K_M" model "hf.co/second-state/Wizard-Vicuna-13B-Uncensored-GGUF:Q4_K_M" modified_at "2026-02-27T08:28:46.707115192Z" size 7865956752 digest "d270a4bc9e15f5066ae7cd861dcf75cd8010055138e1e2974579a73ed157a0f8" details parent_model "" format "gguf" family "llama" families 0 "llama" parameter_size "13B" quantization_level "unknown"



