#✨│ai-help

1 messages · Page 339 of 1

latent kraken
#

i mainly do Ai stuff with like voice lings so that could fall under music right?

grave light
#

you guys know any other ai voice websites that isnt weights.GG

#

His exact wrods

#

words*

proven hill
grave light
#

Thats what I sent

latent kraken
#

alright i did it, now what? it told me to make an account and i did

proven hill
grave light
latent kraken
proven hill
#

youre the best bet on kaggle stuff

viral mason
latent kraken
#

i am trying to use Kaggle, Because ever since Weights.GG got discontinued and I can;t do the AI voice things

#

any other one i go to either doesnt work, just has very basic ai voices or is just paid to use

#

and I am broke misc_sob🙏

proven hill
viral mason
proven hill
viral mason
#

o

proven hill
latent kraken
viral mason
latent kraken
viral mason
silent zinc
#

guys i need arabic female voice

viral mason
proven hill
viral mason
#

<@&1159293140440723499> Hacked account

proven hill
viral mason
grand hinge
#

what is the current best pretrain to use? been out of the loop for the past 2 years for the most part due to losing interest in rvc etc a lil (dont really see a reason for me to list allat other info since the question is pretty basic)

heavy mountain
#

what would i use to train a model? im used to using ones made by others but i wanna make some of my own

void flume
void flume
stark shadow
#

why when I talk works with delay

void flume
#

Because the current AI Voice Changers aren't optimized to be instantaniously fast. There is always some delay, because your voice needs to be recognized, what you're trying to say needs to be recognized, etc.

#

If you pick a tool or model that is lower in quality, you'll get a faster output though, since there are less options for the AI to pick from.

viral mason
#

Legacy core 1.5 is more of a general pretrain while Legacy core 1.6 is best for singing models, 1.6 is not recommended for a voice model such as Kratos from God of War for example but 1.5 and OG is

analog obsidian
#

legacy core 1.5 is not more of a general pretrain, it uses the same exact dataset as 1.6, minus two speakers

#

both are singing pretrains
and just because your kratos model didn't worked well for you, doesn't mean the pretrained is not going to work with any deep voice, it's mostly rng based
stop spreading misinfo
there are times where og works better as a pretrained model, sometimes another one

viral mason
#

that's mostly how he sounds so it's a good example

#

I'm just going off what I know and have experienced from using both 1.5 and 1.6

analog obsidian
viral mason
#

ah

delicate cradle
viral mason
#

could try it again tho definitely

hoary musk
#

guys
how to use codex on phone

#

guys
how to use codex on phone

fringe heron
spark mica
#

is there any videos on how to actually use the thing? the models always get crazy vocal fry

#

lemme be omni man i beg 😔

fringe heron
spark mica
#

idk what its called exactly

fringe heron
#

Pretrained?

spark mica
#

its like a voice changer

fringe heron
#

Oh

spark mica
#

where u put in the model

fringe heron
#

Oh okay okay

spark mica
#

yeye

fringe heron
spark mica
fringe heron
spark mica
#

windows

spark mica
hallow thistle
fringe heron
#

You will also need a virtual audio cable

spark mica
spark mica
fringe heron
#

Damn

fringe heron
spark mica
fringe heron
#

For what purpose thought

spark mica
hallow thistle
spark mica
fringe heron
#

Bad

spark mica
#

whats bad about trolling my friends tho

#

😔

fringe heron
#

What tupe of tr9lling

#

What type of trolling

#

Like anime characters and so or

spark mica
#

like characters from shows

#

e.g. omni man

#

invincible

hallow thistle
#

Anyway, check out Tg Develop's voice changer.

spark mica
#

raiden

hallow thistle
#

-realtime

patent trellisBOT
# hallow thistle -realtime
🔊 Realtime RVC

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

• Vonovox

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

• Wokada Tg-Develop Fork

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.

• Applio Realtime

A Realtime Voice Changer with similar performance to Vonovox & Wokada Tg-Develop Fork, with extra features.

• Wokada Deiteris Fork

Deiteris' fork (modified version) of wokada that doesn't get updates anymore.

⛔ Outdated/Discouraged

These options are not recommended for use.

• Original Wokada

Not suggested, older versions in youtube tuts are even way worse. GUIDE

• RVC GUI Mainline Realtime

The program is worse compared to the ones above, and much less updated. GUIDE

spark mica
#

or tried using

fringe heron
spark mica
hallow thistle
#

If the W-Okada version is b2397, it's the Tg Develop one, mate. BocchiPlushStare

spark mica
#

(mb, im really bad with this whole voicechanger thing, never dont it really lol)

fringe heron
fringe heron
fringe heron
spark mica
# fringe heron Wydm by this

like, it works perfectly, sometimes, if i say 5 sentences, 1 of those will be perfect, the other 4 will sound like _||_X|Z_XQW_D|W_D|

fringe heron
#

But I am not sure

fringe heron
spark mica
#

like the pitch is way off

fringe heron
#

Like unintelligible

spark mica
#

yea

fringe heron
#

Probably is the model fault

spark mica
#

hmm, ill try some other models

#

any good ones i can use to test?

fringe heron
#

They all bad on me anyway

spark mica
#

lol

#

alr ill be back, gimme a sec, ima try some other models

#

also, my gpu usage on it is really low

#

like 7% when it could be using way more

fringe heron
#

It shouldnt be like high

fringe heron
#

Or also on some odd speech sounds

spark mica
fringe heron
fringe heron
#

If they all sound perceptually natural to be honest i would be impressed

hallow thistle
#

This is Tg Develop's voice changer.

image.png
spark mica
spark mica
fringe heron
spark mica
#

but without the bet

hallow thistle
fringe heron
fringe heron
spark mica
spark mica
fringe heron
#

I barely speak english right 😭

spark mica
#

to be fair, aii is just shortened aight

#

and aight is shortened alright

hallow thistle
# spark mica mines at default , 512 and 0.5s

On Tg Develop voice changer, if you mean to make your audio to sound low quality, you can leave extra time to default, but setting it to 2.7 s might be better because sometimes model has to process more. You can set chunk size to around 110 ms for less audio delay.

spark mica
#

also i changed the pitch extraction algorithm to RMVPE and its like 10000x better

hallow thistle
#

More extra time means higher quality but it also causes more delay as well. RMVPE is better than any F0 model; rmvpe_onnx one is made for non-NVIDIA GPUs.

fringe heron
spark mica
#

thanks guys

#

ima go try it out now

#

in game

fringe heron
spark mica
fringe heron
spark mica
# fringe heron Cool, on all type of sounds?

Yea lool. I mean it struggles with lil laugvhs like, "heh" cause it just goes pshh, but normal laughs like hahhaha it works perfectly, super high pitched noises dont work to well, but thats understandable, regular speech is absolutely peak

#

I tried a female voice after the omni man one and it didnt work, AT ALL

#

My voice doesnt work well with female voices lolol

#

Male voices work so well though

#

It was so goated tho, like everyone in the lobby dressed up as invincible characters lmao

#

Got a few goated clips too

fringe heron
#

Lucky you, my voice doenst work well on everything

spark mica
#

What have you tried?

fringe heron
spark mica
#

Ahh

#

Have u tried the omniman one?

fringe heron
#

Not yet

spark mica
#

Want me to send u the link to the post?

fringe heron
#

I tried many funny ones thought

fringe heron
#

But I doubt it will work (well)

spark mica
#

What do u usually use it on btw?

#

The voice changer?

fringe heron
#

Vonovox

#

But I tried many

spark mica
#

I meant like do u use it on calls, games, recording software etc

fringe heron
#

Anyway its not fault of the software that runs the models, its the models fault

fringe heron
#

Recordings

#

Offline

#

I dont need realtime pretty much

#

I use applio for offline conv

spark mica
#

Ah, i was gonna suggest we hop in a call/game and compare lol

fringe heron
#

Sure we can

#

Thats not a problem

spark mica
#

Woo

fringe heron
#

I can later thought

spark mica
#

Yea, my headsets chsrging rn, and thats what i use for my mic so thats perfect lol

fringe heron
#

Just dont mind my english which is bad

spark mica
#

Im used to it fr lol

fringe heron
spark mica
#

Im not completely fluent at it like an american since i have 2 native languages

#

In a way

fringe heron
#

Sweet

spark mica
#

Yee

cerulean sable
#

have clients for an ai web desighn ageny but dont know how to build websites if anybody willing to help me let me know?

delicate cradle
fringe heron
nimble finch
#

hello i can even hear myself it works perfectly earlier but suddenly when im trying it on discord even though the input was line 1 it cannot detect any voice from the application what should i do?

image.png
#
  • Goal my goal is to get the thing detected again by discord and someone could hear it not just on hear myself function
  • Specific Issue: not outputting a sound from my microphone towards line one even though i can hear myself and i can hear the voice properly
  • Full GPU Name: RTX 3080 laptop GPU
  • Operating System: intel core i9
  • Tutorial Link used: its working for a long time
viral mason
#

But have you tried switching to server mode, client mode doesn't work at times

nimble finch
#

i was always in server mode i can hear myself but when outputting using line one it cannot detect any input thats so weird its suddenly happens

viral mason
#

You have the microphone settings setup correctly

#

My bad I'm looking at the picture on phone and it's tiny

nimble finch
# viral mason My bad I'm looking at the picture on phone and it's tiny

yes everything is what im always doing the same all this time its just this time its not working was this a server side issue? cuz as you mentioned now i tried client side instead of server side i can hear it its just the sound is too little but it was outputting from towards line one from my microphone

#

and i switch to server side again and suddenly it cannot detect it again for some reason but i can hear myself function is working fine

viral mason
#

Odd

#

Server mode works for me but client doesn't and it's opposite for you

nimble finch
#

misc_sob what should i do wait for tomorrow maybe maybe its just server side issue but the thing is i can hear myself fine its just not outputting the thing from my mic towards line one

viral mason
#

Since you do have a 3090 you could try switching to Vonovox, it's better than wokada tg fork

#

Maybe that'd be better than dealing with the weird issues it's giving

nimble finch
#

i have 3080 laptop GPU

#

is that a new launcher or app for it to make it work i mean the line one seems working fine

#

since if i swich to client side it works its just has weird and low sound

viral mason
nimble finch
viral mason
#

Quality is improved with the upscaler it has and it's more optimized 3ft

viral mason
#

And yes same voice models work

nimble finch
viral mason
#

The thing that is line 1

void flume
viral mason
#

-# heh larping

nimble finch
void flume
#

ok

nimble finch
# viral mason The thing that is line 1

yes i have it all this time whats weird is i have the hear myself thingy working perfectly fine but it wont output now when im try to use it on discord and any software i have as an input it says it cannot detect any sound even though it was outputting a sound on hear myself

nimble finch
viral mason
#

Alrighty

nimble finch
viral mason
#

For Vonovox you'll have to go into your settings in control panel and enable listen to this device on your line 1 in the recording section

#

It has no monitor setting for some reason

viral mason
nimble finch
patent trellisBOT
#
AI HUB | Technical Support Desk
📋 Required Help Template

To receive assistance, you must provide your system details. Copy and paste the block below into your reply and fill it out.

⚠️ NO INFO = NO HELP

👉 Fill this form:
- Goal (e.g., TTS, AI Covers, Roleplay):
- Specific Issue:
- Full GPU Name:
- Operating System:
- Tutorial Link used:
📍 Quick Checklist Before Asking

• Check Docs: Many fixes are in the AI Hub Docs.
• Be Specific: Say "RTX 3060 12GB", not just "NVIDIA".
• English Only: Keep all discussions in English.

⚖️ Community Guidelines

• No assistance for NSFW/Porn or ANY Illegal Activities.
• Read the [Full Guidelines](#1402790586028789830 message).

Helpers are volunteers. Be patient and don't ask to ask.
nimble finch
fringe heron
nimble finch
#

oh wow thank you @fringe heron

viral mason
#

this one

fringe heron
viral mason
fringe heron
#

really?

viral mason
#

yea

nimble finch
fringe heron
viral mason
#

did u pay?

fringe heron
viral mason
#

odd

fringe heron
#

but i just looked around and i found after 2 sec the lite link

#

mb

viral mason
#

lol

fringe heron
#

this here

Screenshot_2026-05-28_212036.png
nimble finch
#

i extracted it right now but can't see install button

image.png
#

this is the vonovox

fringe heron
viral mason
#

run start

fringe heron
viral mason
#

No they use a 3090

#

Even if it's the laptop version I've asked in the past multiple ppl have said it should work fine on the normal beta version

fringe heron
#

okay

fringe heron
#

but if it works already then :)

viral mason
#

Ohh

#

Ye nah it should be fine

fringe heron
nimble finch
#

i tried reinstalling the vac lite then it still has the same issue T^T

fringe heron
fringe heron
nimble finch
fringe heron
nimble finch
fringe heron
nimble finch
fringe heron
#

also reinstalling virtual audio cable will put the installed devices as default devices on windows so you might want to change that

nimble finch
fringe heron
nimble finch
#

oh i saw it let me try again

image.png
fringe heron
#

more under you have even more settings about chunk, pitch extraction algo and much more

nimble finch
#

oh yes the chunk thing is important where i can see that

image.png
fringe heron
nimble finch
fringe heron
#

ye

viral mason
#

Block size is chunk size in vonovox

fringe heron
nimble finch
#

alrighty guys uhm let me check it a little bit to learn it much more one sec

viral mason
#

take ur time

upbeat copper
#

Which AI is best to try, GLM, minimax, claude, etc. I tried cursor but I hit limit so fast. I want something that is budget friendly and last me a lot more than cursor pro plan. Any help is greatly appreciated

nimble finch
# viral mason <:anime_pray:1159685390156967936>

hello i think i get the hang of it and tried it was now outputting the voice from mic towards line 1 virtual cable towards the headset when trying to listen i think the main problem since i reinstall the vac lite was the server side on the app i was using last time. when i used the vonovox it works perfectly fine its just the listen to myself or hear myself needs to be listen to this device which is the vac lite to be able to hear the output thank you so much guys for your help i really do appreciate it keep up the good work and stay safe thank you so muchie !

viral mason
#

You're very welcome! Glad the issue is solved

viral mason
viral mason
sullen mantle
viral mason
#

try this guide I made

sullen mantle
#

I tried this all before

#

and it sounded staticky

viral mason
#

the guide I made is quite good

sullen mantle
#

to clean the audio

#

the exact same way

#

same steps

viral mason
#

wait fr?

sullen mantle
viral mason
#

it sounds good to me

sullen mantle
#

so why it sound staticky when I did it

#

ion understand

viral mason
sullen mantle
#

for me

#

and see if it's any different

#

@viral mason

viral mason
#

tbh I'm busy with models but I can try as a test

#

not training anything rn

sullen mantle
viral mason
sullen mantle
#

@viral mason

viral mason
#

oh

#

Is that all u have?

sullen mantle
#

I have more

#

but idk if it will be good

#

ill jus send it here and u can tell me to delete after

#

it might repeat some of them it's my old dataset

#

it's a bit bigger

viral mason
#

what songs should I use, I'll clean it myself

sullen mantle
#

file size too big

sullen mantle
viral mason
#

kk