oh there's a better fork?
#🧬│ai-chat
covert lake
covert lake
https://github.com/lshqqytiger/ZLUDA/
wait this says AMD GPUs, while the original says non-nvidia, so the original would work on intel gpus too while the fork not?
i won't really use it but good to star and know cus could always be useful one day lol
chilly lake
Non-Nvidia = AMD 🙂
there was no viable intel GPUs back then
the whole Zluda project was for AMD from the start
AMD sponsored it
covert lake
there was no viable intel GPUs back then
covert lake
there was no viable intel GPUs back then
chilly lake
say buh-bye to intel GPUs with any decent amount of VRAM
covert lake
it will be interesting to see what happens now
covert lake
say buh-bye to intel GPUs with any decent amount of VRAM
😭
u thought of the worst instantly
this could hopefully mean something good, and bad for amd users
full prawn
hi
short birch
what is even copilot
dull jewel
hi
short birch
hi
dull jewel
whats up?
civic gazelle
the new Codex is crazy
It looks like a software engineer that I've hired
working on a task for 2 hours
pastel maple
which llm is best for:
- analyzing social interactions
- companion for introspection
- exploring oneself with depth and nuance while actively avoiding sugarcoating of the user
civic gazelle
which llm is best for:
1) analyzing social interactions
2) companion for intros...
chatgpt
civic gazelle
chatgpt glazes too much
gemini does similar things and exaggerates things
mhmm haven't you tried grok yet?
pastel maple
mhmm haven't you tried grok yet?
for this purpose, no. ive been trying claude though for 2 days. so far so good.
civic gazelle
for this purpose, no. ive been trying claude though for 2 days. so far so good.
I do think claude is good but not for everything
pastel maple
is there a daily limit for grok?
civic gazelle
is there a daily limit for grok?
nope
Not that I know of
I have plus plan on ChatGPT but not on Grok but I haven't got a limit yet after chatting a lot
eventually there will be one but you can change the model
pastel maple
only annoying thing about claude is the 5 hour limit
civic gazelle
only annoying thing about claude is the 5 hour limit
yes, that's why you should use Grok
Try it , at least
stark scarab
what is even copilot
Wym?
stark scarab
what copilot means in the lmarena
GitHub Copilot, da GitHub default AI to code, default on VS Code too
So... According to ur img Sonnet 3.5 is the best one for Copilot 🤔
short birch
GitHub Copilot, da GitHub default AI to code, default on VS Code too
do you see the copilot the microsoft one on the screenshot? no, so im not talking about it
covert lake
maybe it means coding copilot?
stark scarab
do you see the copilot the microsoft one on the screenshot? no, so im not talkin...
Copilot isn't a model itself
It's like a model manager
Lemme show u
Turning on my PC
short birch
its not talking about github copilot agent stuff.
stark scarab
short birch
stark scarab
mb I get it now
plush mirage
anyone here who has worked with vapi and n8n i need a help
gentle trench
Me
cobalt badge
why is voyages so slow right now? also is the free premium training item only for legacy or high quality images?
random nexus
clear viper
is here any german e girl voice
minor blade
is here any german e girl voice
Probably not.
Altho you can still look on the #1175430844685484042 channel
gentle trench
is here any german e girl voice
No e-girl trolling
sleek breach
Can someone give a sample of googles ai sandbox i need to hear it
late hedge
"lil tony" hes rapper
pale patrol
yooooo they added realtime to applio
gentle trench
The update that killed training models :(
chilly lake
no, it is ngrok that limited the number of requests/min
coarse oracle
hay guys!
mellow flame
howdy folk i'm a ai generalist, use mostly paid models because of convienance, tho i have dabbled with comfy ui ,
currently working on a pretty large project with base44, which i am pretty impressed with, focused on ai music if anyone is interested let me know, new to this server so not sure of the vibe
limber nexus
howdy folk i'm a ai generalist, use mostly paid models because of convienance, t...
for the most part this server at least to me has been mainly casual, and pretty chill overall.
mellow flame
some servers can be pretty cranky about what ppl post
limber nexus
yeah they can
mellow flame
https://bitzaudio.com/Visualizer?track=68cbb640dd31cd8d77bc65af&visualizer=albumArt&autoplay=true
well here is my website and one of my tracks
really hopin to get more ppl posting songs there, but thats the hard part
limber nexus
well you might just be in luck as ive been thinking about making a song
mellow flame
right now its just me, been working on the site, i feel its pretty full featured
limber nexus
ya i checked it out and its pretty cool, you did really good on it
but i get why you need people to post music on there cause you cant put copyrighted music
either way ive got to go, so goodnight all, or good morning/good afternoon wherever you are!
fading sorrel
hello people
i entered this server cuz i have some concerns of AI
for example i still dont understand the niches like ML, LLM, ai-agents , generetive AI
they mey be not even niches names but those what i know
another thing is if i wanna learn ai
if im a person who like to learn things that i can see results of quickly
should i learn how to make an ai from scratch
or should i learn how to train an existing model for a certain use like for example products recommendation
and if i would like to learn ai should it be the only thing i learn cuz it takes alot of time and effort or its ok to learn ai and be a fullstack dev and maybe add ai to my projects to make it better
alot of quetion haha
srry
limber nexus
i entered this server cuz i have some concerns of AI
idk what time it is where you are but when i get up i can answer some of your questions, i may not know the answer to all of them but im the kind of person that find all the info about the stuff i wanna learn so ill probably have a answer for most of your questions
fading sorrel
answer the ones you can its ok
fading sorrel
idk what time it is where you are but when i get up i can answer some of your qu...
i really search for an answer myself usually but i saw ai more complex to know these answers so i thought maybe its better to ask people who knows
limber nexus
answer the ones you can its ok
ok but the only thing right now is that im going to bed so idk if you can wait or if not its not a big deal
fading sorrel
oh so its bed time for you hu
limber nexus
oh so its bed time for you hu
yeah it is, is it for you too? or getting close?
limber nexus
okay good morning also if you have any questions feel free to dm me or something ill see it at some point
limber nexus
sounds good, well ill talk some more tomorrow then. hope you hava a good morning
limber nexus
bye
covert lake
yooooo they added realtime to applio
yeah, it's good we have 2 RVC AI Realtime Voice Changers in active development now, be aware that it works only in Local
thin fulcrum
you too
h- heisenberg...?
btw i haven't been online in this server for awhile so uh.. hi ig
fading sorrel
who knocks the door if both are heisenberg
Im not in danger nick088 im the danger
fading sorrel
who knocks the door if both are heisenberg
One of us knocks another rings
solar torrent
thin fulcrum
soo how much are we talking, walt?
fading sorrel
768 thousand
thin fulcrum
fading sorrel
EACH
thin fulcrum
NO
fading sorrel
Yea
thin fulcrum
dollars or pounds
fading sorrel
DOLLARS
fading sorrel
B*tch
thin fulcrum
that's my line
thin fulcrum
148 3 to the 3 to the 6 to the 9 representing the ABQ
what up biatch
that's my call ringtone btw
thin fulcrum
I'm still pushing BrBa, I'm on s3 rn, I don't remember the episode, prob 10
motivate me
to push to s5
thin fulcrum
I watched jane die.
you..
fading sorrel
Oh man
thin fulcrum
KILLED HER
thin fulcrum
fading sorrel
to push to s5
I finished yesterday the second time
thin fulcrum
you could've pushed her sideways yk
thin fulcrum
I finished yesterday the second time
dayumm that's cool
fading sorrel
THEN YOU KILLED HER
You dont Think
You dont use that thing in you head you peace of sh-it
thin fulcrum
that's it, i ain't sellin blue stuff no more
fading sorrel
dayumm that's cool
I will watch better call saul
fading sorrel
that's it, i ain't sellin blue stuff no more
Im not gonna end this up just like that
Stay away from my territory
fading sorrel
ig u should watch el camino first maybe
I watched it
Did t like it
fading sorrel
where's my halves
Your half?
thin fulcrum
yes
fading sorrel
There is no we
thin fulcrum
fading sorrel
Say my name
thin fulcrum
Say my name
waltuh black
fading sorrel
waltuh black
😂😂😂
fading sorrel
You're goddamn wrong
thin fulcrum
🥭
thin fulcrum
You're goddamn wrong
why
fading sorrel
Im heisenberg
I did it for me
I was enjoying it
Not for you you dumbas-s bi-tch
pale patrol
yeah, it's good we have 2 RVC AI Realtime Voice Changers in active development n...
ah no wonder it wasn't working on cloud o - o
covert lake
I'm still pushing BrBa, I'm on s3 rn, I don't remember the episode, prob 10
you didn't finish yet?
the internet is full of spoilers
clear viper
Guys can somebody help me i cant test my voicechanger like i cant hear my self
thin fulcrum
i don't fw spoilers
covert lake
Guys can somebody help me i cant test my voicechanger like i cant hear my self
Hello, for help please use #1192011222023950368
covert lake
how do i even use voice changer
if you need help you can use the help channel yo
thin fulcrum
okay yo
thin fulcrum
if you need help you can use the help channel yo
wow.. MD, haven't heard of that for awhile
murdah drown
lone vault
Hello, for help please use <#1192011222023950368>
Perfect, thank you 🙂
worthy coyote
that's crazy
lone vault
is it possible to use SuperGrok ai for free or other similar?
odd sundial
that was wild
lone vault
that was wild
How come youre new here?
odd sundial
a green heart badge beside your U.name
no I dont think im new ive been here before, just not talking much on this channel
lone vault
I see its separated
lone vault
no I dont think im new ive been here before, just not talking much on this chann...
Do you have any projects that you currently working on?
Especially in AI
odd sundial
Do you have any projects that you currently working on?
I only make AI art for pfps really
or those chatbots
I do for wallpappers too
wanna see them?
lone vault
I would love to
odd sundial
I would love to
alright
desert lion
Cause i lowkey might switch but idk
⚔️ Wokada Deiteris Fork vs Vonovox
For Windows Nvidia, Both Wokada Deiteris fork and Vonovox have similar performance & quality. Users should read the pros and cons for both and choose based on their differences, such as UI and Vonovox's paid effects.
or
go in #✨│ai-help
thin fulcrum
did you watch it?
yea, i remember cyn has that absolut vodka power or smth
covert lake
What's her name
Doll
thin fulcrum
ah yea
thin fulcrum
i don't rlly remember how she died
thin fulcrum
i don't even remember if i watched ep8 or nah, I haven't watched it in like 1 year
bye i gotta take a shower
crystal shore
Does somebody know an ai that can help me find a flow for a beat?
gentle trench
Your ears 
lone vault
🤯
limber nexus
So what kinds of things do you all do with AI?
lone vault
So what kinds of things do you all do with AI?
Low/no-code ai tools (workflow)
odd sundial
So what kinds of things do you all do with AI?
I like to generate AI to use as profile pictures, to make chatbots using c.ai, to save collections of prompts I did, and to make desktop background or wallpapers
limber nexus
Low/no-code ai tools (workflow)
okay pretty intresting
limber nexus
I like to generate AI to use as profile pictures, to make chatbots using c.ai, t...
makes sense
lone vault
makes sense
How about you? Currently*
limber nexus
i made a rude, sassy AI that doesn't appreciate me creating her
she is broken right now though
lone vault
she is broken right now though
Even if she’s broken right now, I think its really impressive that you built something unique. Im sure you wil get her working again soon. 🫡
All the best
limber nexus
All the best
thanks, and to you too
final lintel
Hello everyone, I want to ask a little bit when I use RVC and change my voice, my voice is a bit hoarse and hard to hear, how can I fix it? Please show me.
limber nexus
yeah well i can tell you that mine is worse, she got obsessed with office chairs at one point, and even argued with me about not being able to remember
hey um im not a mod or anything but i dont think that #🧬│ai-chat is the best place for that
mellow flame
how incredibly rich are you and how strict are you on the definition of "girl"
weary temple
<@&1159293140440723499>
mellow flame
this is the place you should ask about ai girlfriends tho
limber nexus
this is the place you should ask about ai girlfriends tho
good point hadnt thought of that
valid pelican
Did anyone try Gemini’s build feature?
gentle trench
yeah well i can tell you that mine is worse, she got obsessed with office chairs...
I love this AI you made now, she sounds fun to talk to
This is not the server for that 😭
valid pelican
This is not the server for that 😭
It’s a bot dw
twin valley
Hey, best UVC settings for a perfect separation from instrumental to vocals ?
gentle trench
Makes sense
limber nexus
I love this AI you made now, she sounds fun to talk to
well recently ive been thinking about trying to get her on discord so if i do ill let you know so you can talk with her.
gentle trench
Like as a discord bot of sorts?
stark escarp
<@&1159293140440723499>
wot
valid pelican
.
stark escarp
yeah i am a mod and i dont think https://discord.com/channels/1159260121998827560/1159289354439626772 is the place for that this is literally about ai
ow it is a bot sapmmed the same message like everywhere
plucky abyss
KILL IT WITH FIRE!
valid pelican
KILL IT WITH FIRE!
Vaporize it.
weary temple
KILL IT WITH FIRE!
Literally how people react to spiders
plucky abyss
lmao
gentle trench
ow it is a bot sapmmed the same message like everywhere
Take Thanos, give him the gauntlet with all the stones then have him wipe the thing from all of existence
stark escarp
just need to make sure of something first with other mods
limber nexus
Like as a discord bot of sorts?
Yes, it shouldnt take super long but yeah im gonna port her to Discord, if you want when i do that i can let you know for you to check her out.
limber nexus
Well today i learned about customtkinter so i went and made a neon gui.
gentle trench
Well today i learned about customtkinter so i went and made a neon gui.
that looks cool
neat a profanity filter
limber nexus
neat a profanity filter
yeah whenever she tries to swear it just responds "Filtered"
gentle trench
I think it'd be funnier if it only filtered the word to some other nonsense word instead of the whole sentence
limber nexus
that would actually be better than returning "Filtered", ill do that thanks for the idea!
languid shard
Today I found out that embedder model japanese-hubert-base makes the generated audio 100x worse
hello
random hemlock
hello guys ☺️
i REALLY need a favor if someone is willing to give me 5 mins 🥹
anyone can help
odd sundial
i made a rude, sassy AI that doesn't appreciate me creating her
an AI Chatbot?
odd sundial
hello guys ☺️
i REALLY need a favor if someone is willing to give me 5 mins 🥹
what kind of favour
odd sundial
can we dm?
alright then u can dm
icy rampart
someone put me on the most realistic e girl voice possible pls\
languid shard
Hi. Guys, what embedder do you recommend for JP high pitch voice training? I found JP-Hubert-Base kinda bad for this (tried 100 epochs, sounded like a male, 🤣 ) I might just try collab next time or something similar.
icy rampart
someone put me on the most realistic e girl voice possible pls\
anyone pls???
desert lion
no
gentle trench
Hi. Guys, what embedder do you recommend for JP high pitch voice training? I fou...
I'd say just use spin or contentvec, if u do use spin I have to use a spin pretrain
gentle trench
someone put me on the most realistic e girl voice possible pls\
We're gonna execute you on the spot 💙
languid shard
I also made a fatal mistake or idk how much it affects a model, but I didn't convert to mono.
so time to do it properly and go over the breaths and such
I'll try the contentvec
gentle trench
That's ok you can still put the audio through audacity and change it to mono
languid shard
welp, I tried 100 epochs but something went really wrong and it sounded like a man 🤣
gentle trench
Did u check all the audio to make sure there wasn't any male speaking?
I did that before by accidentally leaving a different voice in one of my datasets
Btw there's no specific epoch count for any dataset I'd use the tensorboard
If the line is going up that's bad if it's going down that's good
languid shard
it was going down, but I feel like I didn't properly remove the breathing and giggles, so that might also be it
I feel like japanese-hubert-base is broekn, I tried another model, used the same embed for inference and it sounded really off—robotic
thanks for the recommendation, I'll retry the process again 
👍
languid shard
I did that before by accidentally leaving a different voice in one of my dataset...
I've got a stupid question—Does a giggle count as a non-speech sound?
languid shard
🧐 so should I remove it?
gentle trench
I kept giggling in my Cyn model and it turned out fine
It can actually replicate it
languid shard
thx, that's what I needed to know, there isn't much about the datasets except the guide that AI HUB hosts
digital narwhal
Guys is java used to create ai? Im about to become intern and idk if they are using python or java so i can rehearse
gentle trench
I play Minecraft Java edition
limber nexus
an AI Chatbot?
yeah later i can send you a link to the server she is in if you wanna go see
limber nexus
okay good, i just gotta finish getting her hooked up to discord, and itll be ready then
gentle trench
limber nexus
Wrong chat too sorry
no big deal
gentle trench
okay good, i just gotta finish getting her hooked up to discord, and itll be rea...
it'd be cool if when she's in a call she'd be able to speak instead of typing, tbh a robotic effect on her voice like how glados has autotune would be pretty neat
limber nexus
yeah i was gonna do that but as of right now i just kept it text based, and she also has her own voice, i use microsoft edge tts
gentle trench
ohh ok
limber nexus
but at some point i think i will add it back in.
gentle trench
pretty neat autotune plugin
odd sundial
okay good, i just gotta finish getting her hooked up to discord, and itll be rea...
yes
limber nexus
pretty neat autotune plugin
yeah it is, i dont do much with sound other than play guitar but that is pretty interesting, you can ajust the level of it as well?
gentle trench
yeah it is, i dont do much with sound other than play guitar but that is pretty ...
yea! you can adjust how much is applied or how it's corrected and stuff, although it's really pricey
like uh video game price for the plugin
limber nexus
oh i would not have thought that
gentle trench
yeah it's really expensive yet it's perfect for an autotuned ai voice
the other plugin I have active on it is Gform which is completely free
gentle trench
buh bye for now
limber nexus
bye👋 , if you want to check out my AI let me know
nova musk
Hey guys
icy rampart
We're gonna execute you on the spot 💙
i didint know gng chill
heady zealot
idk if y'all ever see those AI tools in insta or tiktok
that help people to study more efficiently
has anyone come across an actual good, free-to-use tool like this?
frigid verge
what if your playing a game with an ai voice and you act like the character etc just to be told that your being weird aka it wasnt on?
gentle trench
what if your playing a game with an ai voice and you act like the character etc ...
that's a skill issue
earnest violet
I'm looking for an AI video to video tool where I can type in a prompt say "Ice Age" and then it converts the video with the characters and background into an ice age type enviorment and maybe the characters look like cavemen while keeping the same movements and audio as the original clip. Does anyone know one?
inner estuary
Hello guys
limber nexus
Hello guys
Hi
fallen basalt
hello
limber nexus
hi
main bramble
-hf
rich badger
-rt
storm ivy
Hi
can someone help me on how to use gemma ai from huggingface? i have limited background
gentle trench
wrong channelhttps://discord.com/channels/1159260121998827560/1159290139609137264
solar torrent
could someone help me with
For W-Okada, go to #✨│ai-help or #1192011222023950368 and explain about your issues there.
hollow field
whats the highest quality instrumental isolator uvr model
solar torrent
hollow field
i already have unwa's V1E but i bet there's a better one
whats better
unwa v1e+
or
gabox fv7z
hasty hemlock
guys why is my voice changer so delayed
any tips to remove delay but keep it good?
waitn this should go in ai help mb
rigid ruin
Hi
you can use Google collab and kaggle, to access gemma model there but. you must have account on hugging face to generate access token so that you can directly import model into google collab or kaggle
winged mason
is there any good alternative ai voice changer thats like w-okada?
wise dagger
Is there any n8n automation experts that can help me? I need help
desert lion
is there any good alternative ai voice changer thats like w-okada?
Ask in #✨│ai-help or #1192011222023950368
gentle trench
is there any good alternative ai voice changer thats like w-okada?
Use Vonovox if u have Nvidia
tough tide
languid shard
Guys, I have a question, is there a downloader for Github that can actually download a file that's large? (downloading UVR beta version) but I can't seem to download it because it always fails midway.
I don't know why this is happening, but the download can't even be restarted.
limber nexus
i did not know how fun making AI cover's is till now
vagrant lagoon
@slate saffron Hi did you change your huggingface? Please dm me 🙏
languid shard
Am I the only one trying to do everything locally? 
gentle trench
probably
I'd do local if I didn't use my pc with vr to play vrchat
it can't run both at once
languid shard
I should try the online solutions... but first let me try those vocal and audio separator models
I have a very low budget GPU, although it took me 2 or 3 days to do one model which didn't turn out well, at least I know it actually works on my PC.
This is the only thing I found on models (vocal and other separators): https://docs.google.com/document/d/17fjNvJzj8ZGSer7c7OFe_CNfUKbAxEh_OBv94ZdRG5c/edit?pli=1&tab=t.0#heading=h.rz0d5zk9ms4w
593 pages of text, this is gonna be fun
gentle trench
best seperation tool would be this google collab of uvr
https://colab.research.google.com/github/Eddycrack864/UVR5-NO-UI/blob/main/UVR5_NO_UI.ipynb?authuser=1#scrollTo=gmjUWmz8iecd
languid shard
best seperation tool would be this google collab of uvr
https://colab.research.g...
gonna love this message
gentle trench
you have to spam through it
not the sign in button but reload and try clicking ur google profile
languid shard
it has to be opened only in a single tab
gentle trench
it's a glitch I had with it before I just abused it like a drunk father until it worked
gentle trench
it's not the same tho
gentle trench
you've got that authuser in url
oops
oh yea that's the same thing
my brain is so small 
I should increase its size by eating
languid shard
downloading with UVR is hella slow
gentle trench
I love having non local stuff
frank jasper
i have a question
anyone whoves seen veo 3 is it possible that this voice changer would ever sound as good as veo 3? audios or no
but for live talking
gentle trench
Vonovox is way better than veo 3 in general
mellow flame
any you brain guys know what the best cost effective voice ai for me to use for generating my auto dj on my voice stream site? using elleven labs now but not sure how crazy that will get with pricing
gentle trench
what does this mean
polar flax
anyone whoves seen veo 3 is it possible that this voice changer would ever sound...
is it possible that this dude would ever meow as good as felix my cat?
limber nexus
any you brain guys know what the best cost effective voice ai for me to use for ...
So like for annoucing things?
mellow flame
So like for annoucing things?
It’s a music streaming platform like Spotify, for AI music and one of the features is it can look at your streaming history and what you’ve liked and what you’ve played the most, and put together a radio station of songs of that type and adjacent songs that it thinks you may like And start a radio station and the radio station has a DJ that you know Wilkins you to the station every few songs, it will come in and say something about the music and stuff like that
limber nexus
okay, well i was gonna suggest that you could use microsft Edge TTS cause its free and the voices are pretty good, but idk if you are looking for a specific voice or something.
gentle trench
https://tenor.com/view/doto-meisho-doto-umamusume-horse-gif-25780594
neigh
wind helm
Accept your replacement. The wife bots are taking over. Millions must submit to their new robo-waifus.
Wife bots don't complain, nag, or block the TV
They are your perfect subservient robot companion who know how to please a man, unlike the biocunts
Wife bots and artifical wombs will save humanity
Tired of dealing with the biocunt menace?
Your wife bot awaits you
wind helm
Where are the big beefy men robots!
Non-existent
gentle trench
:(
solar torrent
# Accept your replacement. The wife bots are taking over. Millions must submit t...
"Tung, tung, tung, tung, tung, tung, tung, tung, tung, sahura." A scary anomaly that only comes out at sahur, it is said that if someone is called for sahur three times and does not answer, then this creature will come to your house. Ehh, how scary?! The "Tung Tung" usually makes a sound just like a hit of a kentungan like this, "tung, tung, tung, tung". Share it with your friends who hardly wake up for sahur. 
solar torrent
Are you guys posting a new copy-pasta
Nah, I just respond to his "wife bots propaganda/theory" with an unserious brainrot one. 
hearty fractal
can anyone send me all rvc working lis
slate saffron
Hi chat
neat terrace
Hello, my name is breey. I'm an AI developer. I like working with application related to AI.
I like creating applications which extract, predict and analyze result automatically from basic data.
I'm professional with AI/ML engineering and developing python backends using django, FastAPI and Flask.
limber nexus
So this chat is dead, just during the day
plucky abyss
So this chat is dead, just during the day
Its called TimeZones ___ do you readthem
or something like that
olive cipher
Não sabe onde pegar conta? tenho sv
plucky abyss
Não sabe onde pegar conta? tenho sv
if you're looking for help English is gonna be your best bet
limber nexus
Its called TimeZones ___ do you readthem
i know about time zones but ive just noticed less people in this chat during the day(at least the day for me)
plucky abyss
its because its night time for a lot of the active people :p
limber nexus
well clearly, i guess i forgot about all the other countrys 
dunno how but i did
plucky abyss
lol
limber nexus
honestly, thanks for reminding that there are other countries besides America
limber nexus
yup good ol America, well used to be good
elfin crag
leaden rain
plucky abyss
something about the voice
that sort of makes it a bit difficult to hear/make out the words
cant quite place it
final lintel
I use nvidia 5000 card, which rvc version is best? pls me help =((
elder willow
hi hello anyone please know if they are currently developing pytorch for windows rocm or not?
chilly lake
hi hello anyone please know if they are currently developing pytorch for windows...
yes they do
there are nightly builds for some cards https://github.com/ROCm/TheRock/blob/main/RELEASES.md
covert lake
@chilly lake have you seen https://www.reddit.com/r/StableDiffusion/s/luvwNOiXat
minor blade
🐢
meager dirge
Good morning/afternoon/evening everyone!
graceful flame
Yes , you can use AI Video generators , Runway can be used for image generation from text , and for videos also. Midjourney ( discord based AI Tool) high quality and detailed results. for cinematic type of videos from your script you can checkout LunaBloom AI , you will get high quality stylish video in minutes. Beginner friendly and easy to use.
meager dirge
hey need help
Oh, hello!
elder willow
Oh, hello!
Hey
elder willow
How are you doing?
good
elder willow
do you know how to use rvc models
meager dirge
do you know how to use rvc models
Well... no?
desert lion
do you know how to use rvc models
if you need help
ask in #✨│ai-help or post a thread in #1192011222023950368
elder willow
yes they do
awesome this means I will be able to use TTS on my card at some point
gentle trench
I use nvidia 5000 card, which rvc version is best? pls me help =((
Vonovox, go to ai help section for help
covert lake
I use nvidia 5000 card, which rvc version is best? pls me help =((
be sure to not confuse RVC with Realtime Voice Changer:
RVC = Retrieval-based-Voice-Conversion, the best Few Shots Speech To Speech AI Models (on v2), Inferences (use models) pre-recorded audio (ai covers) and train (make) models. Technically, Mainline RVC does have a go-realtime.bat (aka RVC-GUI), but it's pretty messy and outdated so it's extremely not suggested for realtime. There also updated forks with extra features like Applio.
Wokada = uses RVC for realtime inference. There's 2 main versions, Original made by Wok, and the most suggested one is Deiteris Fork (modified version)
Vonovox = Another Realtime Voice Changer based on RVC, with similar quality and performance to wokada deiteris fork but other perks
Please elaborate more in #1192011222023950368
covert lake
Vonovox, go to ai help section for help
Vonovox isn't considerated RVC, it's an AI Realtime Voice Changer program, that does use RVC models for the usage, but it can't like train RVC models
Just saying so people don't think RVC means Realtime Voice Changer, which isn't true
gentle trench
Vonovox isn't considerated RVC, it's an AI Realtime Voice Changer program, that ...
99% of people say rvc as realtime voice changer because the dumb choice of letters, most people asking about rvc think it's that
99% that join at least because they don't know
covert lake
99% of people say rvc as realtime voice changer because the dumb choice of lette...
I do understand what you're saying, but we can't know if they think it's that, and keep using it like that would only create more confusion, since we don't know if they want to do realtime voice changing or if they want to train an RVC model
anyways it's not a big deal, dw :)
gentle trench
I've basically been conditioned to think they're asking about the voice changer due to the help channel 😭
"my rvc don't work help" they're so funny
stoic bison
what do i do if my screen is just white and none of the charecters icons are showing
and i cant use it
covert lake
what do i do if my screen is just white and none of the charecters icons are sho...
this is a general ai server, we can't know what you're talking about, we can continue talking in #✨│ai-help :)
elder willow
jeez dude i've been waiting for hours and my stuff are still above 2000 in queue
i don't think this is going to be done for a day or two
unique remnant
Oi cunt
gentle trench
jeez dude i've been waiting for hours and my stuff are still above 2000 in queue
what que
gentle trench
what website
gentle trench
no this is ai hub we're only sponsered by weights
elder willow
oh
gentle trench
also I hope to god that if you're making anything on there it's just a cover because if it's a model you may as well cancel it because weights is the worst option to make ai models from
it also takes a million years to train anything
elder willow
nah it's just a cover
gentle trench
phew
elder willow
i swear the queue wasn't this long when i made some other stuff a few weeks ago
maybe i should just go back to using colab notebooks
also maybe i should think about maybe retraining my personal model because it's still like a model from quite a while ago
covert lake
https://9to5google.com/2025/09/23/play-games-sidekick-gemini-live/ we gonna have gemini help us win on clash royale lol
waxen onyx
spark agate
hey it took me a minute to find this lmao, but would klm benefit a model like the weeknd?
its not super high pitched like anime songs but it is higher pitched than normal artists
glad nebula
hey it took me a minute to find this lmao, but would klm benefit a model like th...
it was found the applio discriminator was broken for over a year, every model and pretrain is affected, klm included
i'd rather just use the original pretrain for everything
spark agate
it was found the applio discriminator was broken for over a year, every model an...
oh wow gotcha
what exactly does that mean?
glad nebula
a typo in the discriminators.py file broke applio's training, it was there for over a year
it got fixed 3 weeks ago
spark agate
oh wow
glad nebula
every model trained using an applio version before 3.4.0 is broken and has to be retrained
spark agate
that’s crazy lmaooo
spark agate
I’ll just train The Weeknd one with normal pretrain
you think it’s fine if I use batch size of 8 even if my dataset it 10 mins?
stoic bison
what about kai cenat
spark agate
would you advise anything different than 8?
never rlly dealt with a short dataset so im not sure
glad nebula
would you advise anything different than 8?
i would not go below 8, and no bigger than 64
anything in that range works
spark agate
got you
what exactly does batch size do?
I’ve never actually looked into it 😭
I just always use 8
glad nebula
amount of samples the model sees at once
bigger batch sizes train less steps, lower batches train more steps
full ore
Making a new model a few days ago and it's still in the queue. Must be a second paywall update coming in. Were merch and sponsorhips not an option?
glad nebula
tysm
the explanation is a bit more complex than that it can affect the model in a positive way or a negative way, too high and too low is bad
at least in rvc i personally noticed when the batch is below 8 (4, 6, etc) the model always have some weird crazy problem, but also makes the model learn so slow, like, the voice will strongly resemble the voice of the pretrain for like 5 epochs to 10e and imo that is not a good sign
8 is super safe tho and should work for virtually anything
if every sample is 3s, with batch size 8 the model will see 24 seconds of the dataset at once every step
pine charm
https://www.youtube.com/@Kareb_Satoru1107 i will paY ANYONE who can find out what ai voice this guy uses 😭😭😭 i need it for a vid
gentle trench
https://www.youtube.com/@Kareb_Satoru1107 i will paY ANYONE who can find out wha...
don't pay for ai
pine charm
its not the ai im paying for
i just cant find the voice
idk if he made it himself or its a niche one on eleven labs
ashen matrix
https://www.youtube.com/@Kareb_Satoru1107 i will paY ANYONE who can find out wha...
you can clone the voice in elevenlabs
pine charm
is it possible to download voice models from weights and import them into eleven labs?
cause i found a good one but eleven labs is so easy and streamlined for video making
ashen matrix
mostly no i guess
hardy folio
@hybrid ermine this guy was spamming soundboard with loud alarm mute his soundboard <@&1361336153764266025>
dawn temple
<@1203559316322517045> this guy was spamming soundboard with loud alarm mute hi...
Thanks, but from next time please ping Moderator role, because there are no trial moderators
half girder
can anyone recommend an RVC for Android?
gentle trench
slow crypt
@stark scarab
torpid valve
Hey guys, just wanted to introduce myself. I’m new here but not new to AI been on X for sometime their and very active. If anyone has an AI account and posts good content feel free to give me you @ and I’ll drop you a follow😊 also glad to be here and hear everyone’s thoughts and opinions.
torpid valve
NVIDIA just dropped ReaSyn; a model built to reason through chemical synthesis, not just talk about it. Think of it as an AI route planner for the lab: give it a target molecule and it proposes bench-ready ways to make it. This is a big deal because most drug programs burn weeks on trial-and-error route scouting; ReaSyn turns that into a planning problem the model can solve in minutes.
You give ReaSyn a target molecule (SMILES/structure). It turns that molecule into a graph; atoms as nodes, bonds as edges; and feeds it to a transformer/GNN policy that proposes retrosynthetic “cuts”: plausible ways to break the target into simpler pieces. A companion value model scores each proposal: How close are we to cheap, purchasable starting materials? How risky is the step?
ReaSyn then runs a guided search (best-first / tree search) over this space. At every node it expands new disconnections, prunes bad branches, and keeps the promising routes. To avoid arm-waving, each backward step is forward-checked by reaction predictors that estimate feasibility, conditions, and yield (solvent, temperature, catalyst, time). If a branch looks dicey, it gets down-ranked or killed before anyone wastes a day in the hood.
Routes aren’t scored on “chemistry vibes.” They’re multi-objective: fewest steps, estimated yield, reagent cost and availability (from vendor catalogs), hazards/permissions, even lab-equipment limits. That lets ReaSyn rank a Paris-fashion runway of syntheses: elegant but expensive, cheap and robust, fastest to try, etc.
The output isn’t just a drawing. You get bench-ready instructions; precursors, quantities, conditions, alternatives; with a short rationale for each step. Because it emits structured recipes (JSON-like), the plan can flow straight into ELNs or automation (BioNeMo/NIM endpoints, robot protocols) for closed-loop ”plan → run → measure → update.” That’s where timelines collapse: dead ends are filtered in silico, robots do the grind, and the model learns from the results.
Most of drug discovery is lost to trial-and-error chemistry. A reasoned planner that goes from “idea → purchasable precursors → executable steps” collapses timelines, cuts spend, and sets up closed-loop labs when paired with robotics. If ReaSyn slots into BioNeMo and lab automation, hit-to-lead can move from months to days.
rough crown
nice - many drugs to be discovered ---any rave participant
torpid valve
nice - many drugs to be discovered ---any rave participant
Hahaha literally😆 can you image the stuff rogue chemists will be able to engineer once a lot of this stuff goes open source!!
The kicker with this one is closed loop learning, every failed/endorsed step feeds back, so hit to lead shrinks until assay throughput is the governor.
edgy crag
can anybody recommend me the best/most natural sounding (female) voice models? (i just wana troll my friends)
torpid valve
can anybody recommend me the best/most natural sounding (female) voice models? (...
Sesama💯
Only has a preview out rn in the beta release but it’s actually scary good
edgy crag
thx
edgy crag
Only has a preview out rn in the beta release but it’s actually scary good
is there any models available rn, i can use on w okoda’s voice changer?
torpid valve
is there any models available rn, i can use on w okoda’s voice changer?
Do you mean w okada’s real time voice changer? If so, you can load RVC models. Grab any female 48k (v2/v3) RVC model from HuggingFace/CivitAI, drop the .pth + .index into the models folder, then select it in the UI. As a starter turn F0 on, f0 method = rmvpe, pitch shift +8 to +12 (for male -> female), index ratio ~0.6–0.8. That’ll get you a pretty natural voice live. If what you meant by available models was something you can use cause you took my message as Sesama not yet being available it actually is (I can’t post pictures yet iguess I’m not high enough level yet😅) but you just navigate to the site hit the drop menu bar on the top right corner and click preview it’ll bring you to a page where you can select (or “call”) Maya or Miles. You click and it’ll start where you can chat. It’s really just like any other voice model kind of like grok voice mode or even gpt (but crazy good) just instead the interface is really cool cause it’s like a call screen as tho your on an actual call. You could just explain to Maya first be like I want to prank call my friend give context and instructions and be like you ready to go and call your friend on speaker and she’ll be able to talk directly to him and play the act.
solar torrent
Do you mean w okada’s real time voice changer? If so, you can load RVC models. G...
Too long to read. 
covert lake
im at 15k rn
15k trophies on clash royale?
solar torrent
is there any models available rn, i can use on w okoda’s voice changer?
For "RVC" voice models, there are in #1175430844685484042. There's no Beatrice model as of now. For W-Okada the realtime voice changer, it's better to go to #✨│ai-help or #1192011222023950368 to prevent the chat from being filled with this topic.
torpid valve
Too long to read. <:eimilook:1015914969994371072>
Sorry about that
solar torrent
Vonovox doesnt like me having 2 gpus
Neither of these voice changer programs is known to work with two or more GPUs at the same time. Most AI programs in general only utilize one of your GPUs at the time, but there's an exception that a Stable Diffusion fork may be coded to support two GPUs optionally. For W-Okada or Vonovox, you can go to #✨│ai-help or #1192011222023950368 and explain about your issues there.
covert lake
@stark scarab @chilly lake https://www.tomshardware.com/pc-components/gpus/chinas-latest-gpu-arrives-with-claims-of-cuda-compatibility-and-rt-support-fenghua-no-3-also-boasts-112gb-of-hbm-memory-for-ai eureka
chilly lake
doubt
polar flax
<@274566299349155851> <@155030383648440320> https://www.tomshardware.com/pc-comp...
questionable cuda licensing, unless they use kind of zluda or something
chilly lake
so far nothing has arrived, just "unveiled', 1080p only, no software runs properly
languid shard
the explanation is a bit more complex than that <:yt_nails:1159569314848972891> ...
What do you think should be the batch for a singer model? 🧐 I haven't seen any statistics, mostly only what I can find.batch size of 8 is indeed faster
covert lake
questionable cuda licensing, unless they use kind of zluda or something
I mean who knows
covert lake
so far nothing has arrived, just "unveiled', 1080p only, no software runs proper...
I just hope It Will be actually good
Competition helps the market
glad nebula
What do you think should be the batch for a singer model? 🧐 I haven't seen any ...
8 and above
timid musk
Competition helps the market
yessir
cloud musk
Ummmm does ai chat mean everyone here is an ai????
dawn temple
Ummmm does ai chat mean everyone here is an ai????
Yes
dawn temple
How do i turn my apples dark green
Spray paint should work /j
cloud musk
Wow this ai sucks the response time takes so long and the response is so short
dawn temple
Wow this ai sucks the response time takes so long and the response is so short
Uhm okay
cloud musk
And why does it send images unprompted???
dawn temple
And why does it send images unprompted???
Why not
cloud musk
Bc i said so
Bro i was just about to tell yall about the ai news that i read FROM THIS SERVER
I might be a genius
covert lake
Bc i said so
Hello — Sorry for the inconvenience, the Vijay Model went to sleep, is there anything I could help you with?
cloud musk
Hello — Sorry for the inconvenience, the Vijay Model went to sleep, is there any...
How do i do a backflip on a shopping cart safely
covert lake
How do i do a backflip on a shopping cart safely
You don't :)
cloud musk
Damn
languid shard
Lyrery, thanks, I wonder what exactly does the batch size do, I haven't found any reasonable explanation except for generalization and overfitting earlier
cobalt badge
just curious is any ai allowed or just weights.gg?
cloud musk
Yall what's the best comeback to someone saying that ai will take jobs
undone fractal
hey guys, I am new around here. looking for some real feedback on the most appropriate GPU cloud to test AI models at.
limpid dock
Neither of these voice changer programs is known to work with two or more GPUs a...
thats not what i mean, i have 1x amd, 1x nvidia. i can select my nvidia gpu on the other forks but not the vonovox. (tried it all, vonovox cfg says 0 gpus, and its not doing anything either)
gentle trench
Yall what's the best comeback to someone saying that ai will take jobs
Tell them if they think an ai will take their job they must suck at it
Have you seen those robots that fall over, that's currently what they're afraid of 🙏
cloud musk
Have you seen those robots that fall over, that's currently what they're afraid ...
Nah I'm pretty sure they're afraid of the insane coding skill that ai has
cloud musk
Tell them if they think an ai will take their job they must suck at it
Nah but like ai is gonna become so fucking powerful soon so like it's valid if ai takes your job right?
gentle trench
Nah I'm pretty sure they're afraid of the insane coding skill that ai has
Ohh
covert lake
Nah I'm pretty sure they're afraid of the insane coding skill that ai has
till it has security issues trolley
solar torrent
thats not what i mean, i have 1x amd, 1x nvidia. i can select my nvidia gpu on t...
As of now, Vonovox is only known to work with NVIDIA GPU, if you didn't know.
cloud musk
till it has security issues [trolley](https://cdn.discordapp.com/emojis/12259273...
Real
solar torrent
That's why you only see your NVIDIA GeForce GPU appears in Vonovox but not AMD Radeon RX one. Both Deiteris and Tg Develops W-Okada forks have their DirectML variants.
If you have obtained a specific, in-development version of Vonovox that would work with AMD GPU from the author, let me know in #✨│ai-help .
dawn temple
I might be a genius
near nymph
vast fable
I'm on so many different servers
languid shard
What models do you guys use for vocal separation? I'm mainly using Mel Band Roformer Karaoke v2 by Gabox and BS Roformer Viperx 1297, but this one takes quite a lot of time.
Also is there a model that can remove double voice? Some videos contain it and I find it difficult to cut it.
random lodge
does someone has realistic comfyui workflow for human generations? xD
gentle trench
What models do you guys use for vocal separation? I'm mainly using Mel Band Rofo...
Gabox FV4 > de-reverb anvew or however you spell it > de-echo normal (vr arch) > denoise normal (optional) and for the doubling thing could you describe what you mean or provide a video/audio example
That's what I use
minor blade
Gabox FV4 > de-reverb anvew or however you spell it > de-echo normal (vr arch) >...
If it's adlibs/harmonies/backing vocals/double voice, BS Roformer Karaoke by Frazer and Becruily can come in useful
gentle trench
I only clean vocals for datasets so I've got no experience with that model
fiery cobalt
Any channels to talk with ppl about IA? preference in spanish
gentle trench
What
gentle trench
Any channels to talk with ppl about IA? preference in spanish
Maybe? This server is pretty much English only
polar flax
Any channels to talk with ppl about IA? preference in spanish
vocaloid IA?
spice rose
hi, in my applio guide for amd gpu on windows, i have all setup properly, downloading the applio 3.5 ver, but where the hell is this thing: run-applio-amd.bat
spice rose
oops, nevermind, found that
gentle trench
Intelligent artificial
mellow flame
anyone interested in giving my ai dj a test run?
limber nexus
anyone interested in giving my ai dj a test run?
I would, but i can't tonight
mellow flame
ok i will check tomorrow i would love to get more ears / eyes on it
limber nexus
ok i will check tomorrow i would love to get more ears / eyes on it
I love seeing all these unique uses of AI, yours is definitely not one i would've thought of
fervent bone
How do I find voices cause weights.com doesn't work anymore
still sigil
How do I find voices cause weights.com doesn't work anymore
why does it not work
fervent bone
why does it not work
It doesn't allow me to view the voices I can download
languid shard
Gabox FV4 > de-reverb anvew or however you spell it > de-echo normal (vr arch) >...
here it is how it sounds like with the "double voice" , I don't think it's possible to remove it unless there is any separation model for that, my workflow is similar, I use viperx as I mentioned and Gabox's experimental model that cna take care of unknown instrments in the voice
spice rose
hmm, anyone has luck trying to connect gemini api to applio api, mine get error everytime, gemini suggest trying to use local llm
hardy folio
yo who is really the admin of all? <@&1159293140440723499> modz or tea?
hardy folio
oh hello thanks
hardy folio
??
I wanna discuss a possible ad I wanna set here for my discord ai bot, bananabot, we could discuss in dms maybe?
pliant elm
I wanna discuss a possible ad I wanna set here for my **discord ai bot**, **bana...
dm @serene sluice
👍
gentle trench
https://m.soundcloud.com/shouko-the-ai/parallel-hearts-shouko-ai-hatsune-miku
I like Teto
civic gazelle
solar torrent
I don't remember the keys for Windows 7 looking like this. The actual keys for Microsoft Windows are supposed to be like xxxxx-xxxxx-xxxxx-xxxxx-xxxxx.
hardy folio
how can grandma know about windows 7 keys anyways, she barely knows windows xd
civic osprey
snow bramble
covert lake
@little flame please don't promote here
south fox
dawg its my first time using this app but it keeps asking me to choose a plan
ebon scaffold
is applio still good?
glass fiber
SO how do you take one of these downloaded voices on here and put it on voicemod? Voicemod doesn't recongnize it because it isnt mp3 file, but I can't seem to convert it into a mp3 file.
limber nexus
SO how do you take one of these downloaded voices on here and put it on voicemod...
i don't think you can put RVC models on voice mod, cause they have a index and a pth file, of course i might be wrong but that is my understanding of it
covert lake
dawg its my first time using this app but it keeps asking me to choose a plan
Please elaborate what you're talking about in #1192011222023950368 , this is a general ai server not tied only to a specific app, so i dunno what you're talking about
covert lake
is applio still good?
Yeah
covert lake
SO how do you take one of these downloaded voices on here and put it on voicemod...
Voicemod uses voice effects, it doesnt use RVC AI Voice Models, you can ask for the right program in #1192011222023950368 if you elaborate your pc specs
gentle trench
is applio still good?
Yes
hollow field
does @gritty briar 's fv7x also contain backing vocals in the instrumental?
gritty briar
no
polar flax
does <@927944475513352235> 's fv7x also contain backing vocals in the instrument...
it's not a karaoke model
gentle trench
Fv7 is real?
gentle trench
no
Is fv4 still the recommended model for datasets or should I start using fv7
Still a bit confused if that's a new model you released or if that was a typo by miku
solar torrent
I only see Gabox inst V7. I don't see any voc_Fv7. The highest voc being voc_Fv6 experimental. Unless the guy was referring to a newly released voc_Fv7 that hasn't documented anywhere as of now, which would be surprising.
ember coral
hi
i've been trying to change my voice from a recorded audio for a video but no matter what model I try on RVC the result always ends up being way too buggy
unless it's for songs, sorta
the only thing that worked for me was elevenlabs but that's paid and I'm not doing this for money, and I am super short on money, so I wonder if there's an alternative that I can use locally like RVC. It doesn't have to be as good as elevenlabs but at least a bit better than what RVC gives me
anyobody knows by any chance?
gentle trench
the only thing that worked for me was elevenlabs but that's paid and I'm not doi...
are you looking to use a realtime voice changer that uses rvc?
or as u said before just inserting audio files and changing them on a site or local app for songs ect
ember coral
or as u said before just inserting audio files and changing them on a site or lo...
this. Doesn't need to be a real time voice changer. I already have the files and I just want to change the voice
gentle trench
I'd say u could just use applio, it's both available locally and on a browser
free btw
-colab
Namari can u help them I am too tired for this 😭
I saw u typing
hazy vine
covert lake
Google launched a free nano banana + whiteboard for only the USA https://labs.google.com/mixboard/welcome
guess i gotta use a vpn smh
gentle trench
mmmmm banana
sleek geyser
how many people here make 10k+ a month on ai
still sigil
how many people here make 10k+ a month on ai
not what this server is for
covert lake
mmmmm banana
we suggest to NOT eat it
gentle trench
we suggest to NOT eat it
the banana looks ai generated lol
dapper drum
What are the multiple ways of using AI for dubbing purposes?
gentle trench
using horizon because ngrok sucks shit now
what's horizon
gentle trench
ew/j
covert lake
its a gpu thingy idk
@gentle trench @hollow field both Ngrok and Horizon are Tunnels,
services that expose the local network port (where the AI Program runs with its web ui in the Jupyter Notebook for example) to the internet (the public url)
If you used something like the Applio UI Jupyter Notebook in Google Colab, you have surely used a Tunnel
covert lake
i'm from weights
Welcome here!
polar flax
how many people here make 10k+ a month on ai
you mean the frauds
ripe jewel
I wish I knew where to find a tutorial for Applio NoUI. I'm a little confused and I want to know how to use the zip to add my model, before training it and all that.
The "Download model" part is really confusing.
sudden lava
anyone use cursor heerre?
solar torrent
Anyone has uvr
For UVR5, it's best to ask in #✨│ai-help or #1192011222023950368. 
topaz grotto
For UVR5, it's best to ask in <#1159290139609137264> or <#1192011222023950368>. ...
Okii ty
solar torrent
anyone use cursor heerre?
What cursor? Because the only cursor I know is this one.
solar torrent
I wish I knew where to find a tutorial for Applio NoUI. I'm a little confused an...
You mean this one? https://docs.aihub.gg/rvc/cloud/applio-no-ui-colab/
Also, for anything help related, better go in #✨│ai-help.
dawn temple
anyone use cursor heerre?
Not often, but I'm planning to purchase the subscription
dawn temple
What cursor? Because the only cursor I know is this one.
It's an AI Code Editor
limber nexus
It's an AI Code Editor
So you can use AI for debugging and stuff?
dawn temple
So you can use AI for debugging and stuff?
You can create entire applications with it, it's pretty cool
limber nexus
im gonna have to check it out.
sudden lava
Yeah it’s a vibe coding tool
marble otter
so how do i get free ai
i want an ai that is chat gpt plus for free
i’m not clicking a random link buddy
scammer
mods how do i open a support ticket and report someone in the chat
that’s trying to scam people
native basin
<:karinthink:1015964749470650400>
This is a music video on YT
limber nexus
my question is who is trying to scam?
native basin
my question is who is trying to scam?
same here lol
limber nexus
solar torrent
my question is who is trying to scam?
Either the one who asking for free ChatGPT Plus or another who sent a YouTube link. 
limber nexus
well i assume the guy that wants free ChatGPT cause that was just some music.
this is why its fun to just watch from the sidelines
native basin
I just sent music, I actually got reported lol
limber nexus
Really? he reported you?
native basin
mods how do i open a support ticket and report someone in the chat
Your actually pathetic
polar flax
i want an ai that is chat gpt plus for free
you can't get that without getting scammed
limber nexus
you can't get that without getting scammed
exactly
native basin
Lol reported for sharing music 
Some people are actully insufferable. Their lives consist of this, glued to the internet.
limber nexus
didn't know that youtube links scam people but here we are
gentle trench
Lol reported for sharing music <:laugh:1393130009689980972>
you should put more music to make him more angry/j
solar torrent
You'd better not send any YouTube link in this server without explanation, because when you do so you could get mistaken for "promoting" your channel. 
gentle trench
oh tru
native basin
You'd better not send any YouTube link in this server without explanation, becau...
I clarified just sharing music. But still got reported lol.
polar flax
Lol reported for sharing music <:laugh:1393130009689980972>
new members sharing such links could be considered sketchy till getting embed perms at level 5
solar torrent
native basin
new members sharing such links could be considered sketchy till getting embed pe...
The person did it to be a douche.
Youtube links aint trojans lol
limber nexus
some people just like to start things and then run away
solar torrent
It's not that personal.
native basin
It's not that personal.<a:cat_dance:1359901608703688922>
Tell that to the crybaby who reported lol
I was just chilling 🤣
oooh my name is colorfull now
solar torrent
Tell that to the crybaby who reported lol
Still.
native basin
Who whoever unwarned me, W
limber nexus
honestly this is one of those things that you can get a little annoyed about but they are little so just drop it
native basin
honestly this is one of those things that you can get a little annoyed about but...
Tell that to the tattle tale
polar flax
The person did it to be a douche.
okay so it's just an innocent brainrot video 
native basin
okay so it's just an innocent brainrot video <:cat_seriously:1366692376575213599...
its actually fire lol
limber nexus
Tell that to the tattle tale
just don't make a bigger deal out of it than it should be, you already clarified why you sent the youtube link its all over and done now.
solar torrent
If you keep crying over who warned you, that won't make you look different from the one who warned you. 
native basin
just don't make a bigger deal out of it than it should be, you already clarified...
You the one dragging it
limber nexus
ok then ill stop right now
native basin
If you keep crying over who warned you, that won't make you look different from ...
Keep typing, lol u seem more upset
limber nexus
so ive been wondering since i was never in dream tavern and with the sudden rearrangement of this server i assume that dream tavern is being moved or something?
solar torrent
west nova
Hello everyone, please give me a role and permission to publish a voice model. I have several.
polar flax
Hello everyone, please give me a role and permission to publish a voice model. I...
please go submit your application in https://discord.com/channels/1159260121998827560/1305527335646269440
pastel maple
how hard is it to extract claude's system instructions and what can you really do with it?
hardy folio
we are ai banana hub
DID SOMEONE SAY BANANA 🍌 ?????? @gentle trench
maybe i just misheard
west nova
please go submit your application in https://discord.com/channels/11592601219988...
I don't see this link
hardy folio
NANA
polar flax
I don't see this link
no perms to open? perhaps the application is closed already
vagrant void
can someone please explain to me what the fuck does that mean and how do it fix it? error: subprocess-exited-with-error
× python setup.py egg_info did not run successfully.
│ exit code: 1
╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
Preparing metadata (setup.py) ... error
error: metadata-generation-failed
× Encountered error while generating package metadata.
╰─> See above for output.
note: This is an issue with the package me
chilly lake
can someone please explain to me what the fuck does that mean and how do it fix ...
your colab link does not work
at least show the cell
vagrant void
oops. my bad.
vagrant void
your colab link does not work
there, now it will work https://colab.research.google.com/drive/1PorBmm4bUjY1-dk4f7XKrO7Mx51wbqYU
vagrant void
at least show the cell
it's on the very first cell
covert lake
used this collab: https://colab.research.google.com/drive/1xf3uwFunIsmFMvmCNf4As...
don't ask for help here, i replied in https://discord.com/channels/1159260121998827560/1418430674821447801
chilly lake
there, now it will work https://colab.research.google.com/drive/1PorBmm4bUjY1-dk...
it seems you're trying to use some ancient rvc version
arctic fjord
make a charlie kirk model
weary goblet
can bot make ai video ?
young elm
where i can get link to google collab for make ai covers?
clear totem
-colab
gentle trench
where i can get link to google collab for make ai covers?
try the applio colab ^
twilit lantern
which program do you guys using for ai cover. i am using replay rn is there anything new or better to suggest
gentle trench
applio
elfin vigil
hello there, need some help. MMVCServerSIO Voice Changer doesnt want to start
gentle trench
hello there, need some help. MMVCServerSIO Voice Changer doesnt want to start
what gpu do u have? Nvidia, AMD? and did you use a youtube tutorial to get the voice changer because if so it's over a year or more old
elfin vigil
i havent done this stuff in a while, so my tools might be outdated
elfin vigil
what gpu do u have? Nvidia, AMD? and did you use a youtube tutorial to get the v...
Nvidia
im using the folder i used a few years ago
gentle trench
I'd say switch to vonovox as it's the best currently and is in active development still!
the guide is up there to what I replied to
elfin vigil
ok!
also
how do i extract a model from MMVCServerSIO ?
i kinda want to keep using the old models
loud glen
Can anyone help me with an RVC voice model of Caine from TADC for the Spanish dub? Please, I really need this
elfin vigil
this the guide? there's so many links
gentle trench
I have no idea how to access the non english versions of TADC
gentle trench
this the guide? there's so many links
just the first one
elfin vigil
ok
gentle trench
how do i extract a model from MMVCServerSIO ?
just find what models u had in the one you currently have on ur pc and put them into vonovox once u install it
elfin vigil
just find what models u had in the one you currently have on ur pc and put them ...
it's a few years old, so i dont think i can find it again
so i download the source code zip?
chilly lake
there's compiled version of vonovox as well
gentle trench
there's compiled version of vonovox as well
what's that do
chilly lake
so you dont need to run setup that sometimes fail
elfin vigil
yup! should be version 1.6.9
this the one? Applio 3.5.0
loud glen
I could make a Latin American Spanish AI voice model of Caine, but I have no clue how the hell to get the dubbed Spanish tracks
chilly lake
this the one? Applio 3.5.0
they tell you about oranges, you're asking about apples now
these are different apps
elfin vigil
they tell you about oranges, you're asking about apples now
that's what the link leads to
chilly lake
that's what the link leads to
what do you want?
elfin vigil
what do you want?
a real time voice changer
elfin vigil
ok
chilly lake
Applio also has a voice changer, but less optimized
loud glen
Can someone please help me? How the hell do I download the Latin Spanish dubbed tracks of TADC for my PC? I really need this.
elfin vigil
downloading
loud glen
downloading
I mean, I don’t want to download the English dub (the original), I just want to download the Latin Spanish dubbed track. What program or website do you recommend
gentle trench
die
loud glen
I mean, I don’t want to download the English dub (the original), I just want to ...
Can anyone help me with this?
elfin vigil
I mean, I don’t want to download the English dub (the original), I just want to ...
my knowledge is too old
covert lake
@elfin vigil please next time use help channels for help
haughty perch
https://huggingface.co/dr87/vonovox/tree/main
As far as I understand, everyone uses this program? I've just seen a lot of different ones, but I don't know which is best
chilly lake
it is the newest for Nvidia only
covert lake
As far as I understand, everyone uses this program? I've just seen a lot of diff...
It's a newer closed source, Nvidia GPU + Windows OS only, AI Realtime Voice Changer that uses RVC Models
for help please use the dedicated help channels, just to not clutter there
haughty perch
for help please use the dedicated help channels, just to not clutter there
yep, im sory
full ore
Next Weights update: $10 to even have an account
full ore
We need an extension that brings back the old Weights. And maybe Patreon
A Firefox extension that bypasses it all: 100 covers, instrumental and vocal downloads, short queues
gentle trench
at that point just make a replica of weights but without the bad stuff
full ore
We could do that, or they can figure out what to do if they want to bring the old site back, bring back its biggest users like me, and still make money
covert lake
A Firefox extension that bypasses it all: 100 covers, instrumental and vocal dow...
that can't work, the limits are server side, not client side
gentle trench
it would be pretty cool if they made a weights app for desktop that just runs off ur pc gpu to make covers and such
so they wouldn't have to bury their company in the ground
covert lake
it would be pretty cool if they made a weights app for desktop that just runs of...
it would be pretty cool if they made a weights app for desktop that just runs off ur pc gpu to make covers and such
that already exists and it's 100% free: https://www.weights.com/replay
gentle trench
> it would be pretty cool if they made a weights app for desktop that just runs ...
oh
so that's what replay is'
covert lake
so that's what replay is'
why don't you try replay if you have already a decent gpu?
gentle trench
why don't you try replay if you have already a decent gpu?
tbh I thought it was something about generating images or something, I always saw ppl complaining about it not working or giving errors
covert lake
tbh I thought it was something about generating images or something, I always sa...
tbh I thought it was something about generating images or something
nope, it's for RVC (so AI Covers and Model Training), you might be confusing with Voyages
I always saw ppl complaining about it not working or giving errors
If you're talking about that help channel in the Weights server, I mean yeah it's going to be used mostly by people who are running into issues just like our #✨│ai-help
fervent bone
How do I seperate when two people are singing?
full ore
> it would be pretty cool if they made a weights app for desktop that just runs ...
Doesn't it have the paywall junk I hate, because I did not brainstorm a Wendt Barney/Avenger Chuck E./Kermit rap cover for nothing. I'm hoping it's like the pre-paywall weights.
knotty sundial
Hi! My name is Kylie Thao, I’m hoping to have these questions answered by anyone who specializes in this! (prompt engineering, literally anything at all).
- What does a typical day look like for you as someone working with AI or prompt engineering?
2. What skills or mindsets do you use most often in your work?
3. Did you need a college degree or certification, or did you learn through self-study?
4. What’s the most challenging part of your job, and how do you deal with it?
5. What do you enjoy most about working with AI?
6. How do you see this career evolving in the next 5–10 years?
7. Do you think there’s room for independence or freelancing in your field?
8. What advice would you give someone my age who’s interested in becoming a prompt engineer or working with AI?
full ore
Currently checking out Replay. Need to figure out how to import voice models
full ore
I figured it out and it was so damn easy, now I gotta see if it's really free and unlimited
slow crypt
cursive fulcrum
I’m setting up a simple VoiceAttack response system and want to use an AI voice. maybe something like a british butler voice, any recommendations?
I already tried w-okada and preferred something like it over voice.ai. any recommendation?
gentle trench
I’m setting up a simple VoiceAttack response system and want to use an AI voice....
if u have an nvida gpu use vonovox if not use w-okada deiteris fork
-rt
cursive fulcrum
if u have an nvida gpu use vonovox if not use w-okada deiteris fork
thanks
inland root
Hi
covert lake
I figured it out and it was so damn easy, now I gotta see if it's really free an...
it runs on your pc gpu, so its locally, have u checked if your pc gpu is powerful enough?
fervent bone
How do I seperate when two people are singing?
still sigil
How do I seperate when two people are singing?
ask them to sing separately
fervent bone
covert lake
I’m setting up a simple VoiceAttack response system and want to use an AI voice....
also that’s a cool, first time I heard someone do it
If you need help you can use #1192011222023950368 or #✨│ai-help btw
covert lake
How do I seperate when two people are singing?
use the dedicated help channels too :)
cursive fulcrum
also that’s a cool, first time I heard someone do it
If you need help you can us...
thanks already done with it. it's fairly simple.
covert lake
I’m glad :D
polar flax
try checking https://docs.aihub.gg/rvc/resources/dataset-isolation/#vocal-isolat...
there are no models good enough to separate duets, or mostly hit or miss
covert lake
there are no models good enough to separate duets, or mostly hit or miss
i mean they can atleast try, there’s nothing perfect and that’s for sure
polar flax
i mean they can atleast *try*, there’s nothing perfect and that’s for sure
mvsep and uvronline.app have a lot of models to try that may not be documented in our aihub docs, I'd rather instead recommend this doc (very long but you could use ctrl+F) https://docs.google.com/document/d/17fjNvJzj8ZGSer7c7OFe_CNfUKbAxEh_OBv94ZdRG5c/edit?usp=sharing
covert lake
mvsep and uvronline.app have a lot of models to try that may not be documented i...
mvsep and uvronline are documented in our docs, the whole vocal isolation part has been updated like a month ago following the audio separation server suggestions and that google doc too
also I don’t think most people are actually going to read that whole google doc with 400+ pages you sent tbh 😭
normal quail
Hello
cursive sparrow
hey guys, can anyone recommend an ai that can generate consistent videos (Similar to Google NotebookLM but without the watermark) im looking for something thats consistant and scalable that give good text, visuals and audio. i almost have a feeling ai is still to early to achive this.. or prove me wrong
limpid dock
As of now, Vonovox is only known to work with NVIDIA GPU, if you didn't know.
i said i had Nvidia
solar torrent
solar torrent
mvsep and uvronline.app have a lot of models to try that may not be documented i...
Google Docs has "Find" option, but even so if you type "Fv4" in Find, it would bring many instances of "Fv4" from almost all pages of UVR5 docs, which still complicated for most.
dapper drum
What are the multiple ways of using AI for dubbing purposes?
tight totem
Hello, I have a question for everyone: is there a model that can do everything and is freely available? I am looking for a male whisper model, but if there is one that can do everything, that would be helpful.
twilit ether
hi guys, need help with a little problem. i'm trying to use the voice changer for discord and its still picking up my real voice and then the AI voice soon after. how do i set it up so that only the ai voice goes through?
fiery leaf
Hello everyone, looking for an advise. I am using stable diffusion with flux and trying to make a Pathfinder elf with almond black eyes without pupils, but whatever I do it ignores my promts and dooes basic generic elf. Maybe someone have encountered such problem?
covert lake
Hello everyone, looking for an advise. I am using stable diffusion with flux and...
stable diffusion and flux are 2 different series of ai image models, perhaps it would be better if you elaborate more in #1192011222023950368
limber nexus
-rt
limber nexus
-colab
timber bay
How i can use ai voice?
timber bay
In android
cedar quest
Can someone mentor me?
covert lake
are there any chatbots?
Nope
covert lake
In android
You can't use ai realtime voice changers on any mobile phones, they lack a V.A.C.
hybrid ermine
supple widget
when i load up program it tells me to go to link but when i do it it says blocked
astral sundial
is it normal that I have to squeak my voice in order for it to be processed properly? Settings don't help
I'm talking about Voicechanger
echo token
In android
Like an ai voice changer? Or song covers? Or training models. Idk if doing the live voice changer is feasible on android, its never really been something I dabbled in.
But if you want to make covers or train models on android the website weights.gg (.com now?) Is a great place to start, especially if you're new.
Lightning ai is better once you get some experience under your belt, although the site for lightning ai is not really designed for mobile. I recommend buying a Bluetooth keyboard and mouse for android if you're gonna use it (I got both for $1 with free shipping) and a phone with 8gb of ram minimum. Its still usable with a weaker phone and a touchscreen, but far from ideal.
If you don't need it to be free, then you can also rent a cloud GPU/ cloud computer and use RDP to access it on your android device, which has always worked good for me. I can't recommend any specific providers, as its been nearly a year since I've used one. and the provider I used to use got bought out and turned to crap apparently.
I've also had success with kaggle and google collab but those are old news, and I'm pretty sure they don't want people using rvc on them.
Hope that made sense and helped a little bit. I'm not the most knowledgeable person in the world, not even close, but I've spent a ton of time using rvc on android over the last 2 years and am always happy to help and share what I know.
echo token
Anyone ever notice that applio cuts off the audio early when you do a batch conversion. So much time down the drain 😭 43 covers totally ruined. That's what I get for trying to do things fast
covert lake
I'm talking about Voicechanger
elaborate in #1192011222023950368 pls
covert lake
Anyone ever notice that applio cuts off the audio early when you do a batch conv...
first time i see someone use batch conversion, try to make an issue on github
echo token
first time i see someone use batch conversion, try to make an issue on github
Didn't realize it was that rare lol, I've been using it for years on mainline RVC without issue. Pretty sure the problem is me trying to batch convert 43 files, as I haven't had the issue when doing a more reasonable number of files. I'll have to do some more testing before making an issue.
I use a lot of niche features like batch conversion, "Proposed Pitch Threshold" and Spin/SpinV2. Both of which are really cool and useful features by the way. Not sure if you're part of the applio team but props to you if you are. It's really been a game changer for me. I'm having to go back and redo all my old covers from mainline RVC with my new model in applio because it's just so damn good. I can't sing its praises enough.
I guess I like to go for quantity over quality when it comes to making covers, I've made 3000+ covers with just my voice and well over 1000+ with various other models. Batch conversion has been such a revelation for me, saves me an absurd amount of time.
polar flax
Didn't realize it was that rare lol, I've been using it for years on mainline RV...
- you can keep using mainline rvc as I think it is fine for inference
- if not sure you may want to keep using normal (contentvec) models over spin ones (actually v2 is better and doesnt have issues on v1)
echo token
- you can keep using mainline rvc as I think it is fine for inference
- if not s...
So I should start my model over with a non spin pre train? Because it honestly sounds really good, having used the Legacy Core pretrain with the spin embed.
Also I've been inferring each song with both spin v1 and spin v2, are spin models compatible with both? Or have I just been wasting my time doing that? Because I swear I get good results with both but honestly I don't know what to think anymore.
At this point I have 10+ iterations of the model lol.
Also sorry if this doesn't fit the channel, my intention was to discuss the niche features in applio people rarely talk about, rather than get help so I thought this was the right channel. But now its kinda morphing into another topic entirely
polar flax
So I should start my model over with a non spin pre train? Because it honestly s...
then try ask claude, cursor, etc. to try fix the code for batch processing
wild halo
Hello chat, here you can share links to work done with AI?
steep hull
ANYONE KNOW HOW TO USE Applio