✨│ai-help

[quaint minnow]

I don’t listen to Kanye West

[proper shale]

Without hurting the voice in the process

[molten pecan]

Are you logged in huggingface?

[quaint minnow]

Yes

[molten pecan]

How long is the audio?

[proper shale]

how long is the audio

[molten pecan]

How long is the audio

[quaint minnow]

40 seconds? It’s WAV too so the file is 12mb

[molten pecan]

Hm

Do this

[quaint minnow]

But on the Google doc version it would take 10 seconds to convert

Not 300

[molten pecan]

Duplicate the space with the purple button on top

[proper shale]

Yeah because it had a GPU

[molten pecan]

And make sure you set it to Private

[quaint minnow]

Oh

Wait

[proper shale]

On this one you'll have to wait till it errors out or something and see if you can change the pitch extractor to Dio

or PM

[quaint minnow]

It fixed itself

[molten pecan]

Oh cool then!

[quaint minnow]

Took half a year but worked eventually

Thanks!

[molten pecan]

Yw!

[molten pecan]

Duplicate the space like I said so it will go much faster

Good luck!

[quaint minnow]

Ooh I’ll try that next time

Thanks!

[indigo forum]

Understood. I took out all spaces but it sounds like he has verbal diarrhea 😅

[halcyon oxide]

@molten pecan can you help me?

I have a problem with uploading my Ai voice

It asks me for a resolve link

What is that?

[gaunt ruin]

@low shard @molten pecan

apparently, waiting was worthless in ilaria

[low shard]

After you putted the audio,did you check if you correctly selected the model,index,and the audio file before starting ?

[gaunt ruin]

yup

[brittle wing]

can someone give me link to the one i have to download?

[low shard]

Try to refresh model and audio and then select them again,select also the correct index in case it didn't have a checkmark and then convert

[molten pecan]

To download what?

[brittle wing]

idk whatever the voice ai is called

havent used it in 3 months

deleted everything

[brittle wing]

send me the link

[rare goblet]

Ayo? @brittle wing level 1 !!!

[molten pecan]

Voice AI?

[brittle wing]

yes sir

[molten pecan]

You mean the models or RVC?

[brittle wing]

rvc

[molten pecan]

Ahh okay

Here

-rvc

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation by country

🇧🇷 Brasil (PT-BR)

https://docs.aihub.wtf/v/brazil

🇪🇸 Spain

https://docs.aihub.wtf/v/spain/

🇫🇷 France

https://docs.aihub.wtf/v/france

🇩🇪 Germany

https://docs.aihub.wtf/v/germany

🇯🇵 Japan

https://docs.aihub.wtf/v/japan

🇰🇷 Korea

https://docs.aihub.wtf/v/korea

🇵🇱 Poland

https://docs.aihub.wtf/v/poland

🇷🇺 Russia

https://docs.aihub.wtf/v/russia

🇮🇹 Italy

https://docs.aihub.wtf/v/italy

[molten pecan]

It's somewhere in the guide I think

[brittle wing]

my voice deep af it took me whole week to customize an model for me

-_-

got any suggetions for good models for deep voice?

[brittle wing]

huh tf that interface looks completely diffrent

[molten pecan]

Try changing the semitones a bit on RVC

[brittle wing]

yeahh lemme download first

[molten pecan]

You might want to use crepe algorithm instead of rmvpe. Just check which of the two sound better

[brittle wing]

where do i click to download??

[molten pecan]

I don't know 😭

[molten pecan]

I assumed the guide had a link to download it

[brittle wing]

man is this the same ai??

[molten pecan]

https://huggingface.co/spaces/r3gm/Aesthetic_RVC_Inference_HF this is the one I use

[molten pecan]

Idk which AI you are talking about

[brittle wing]

is that an online tool?

the one i used was downloaded

lemme show

[gaunt ruin]

well, hope it works, otherwise, what a shame, google really screwed us up

[brittle wing]

https://youtube.com/shorts/dnQotnbHax4?si=Y2u9O08CxENrBEjw

that one

ohh its called okada

[molten pecan]

Yes it's online

[brittle wing]

i want the one from video

[molten pecan]

Ohhh you want to use the real time voice conversion?

[tranquil cliff]

Hi

[brittle wing]

yeah whatever its called

[brittle wing]

sup

[molten pecan]

I know nothing about okada

[brittle wing]

damn

[molten pecan]

So try asking on #🔍│help-w-okada

[brittle wing]

nvm a helper on good time

[brittle wing]

YO SUP

[molten pecan]

Banned

[brittle wing]

https://youtube.com/shorts/dnQotnbHax4?si=shKjvyUwQ14BF1Oy
i want this ai i used months ago

[tranquil cliff]

Do you need any help

[brittle wing]

no idea how to get it now

[gaunt ruin]

how do I properly uninstall mangio RVC? so that I can try again with another space?

@tranquil cliff @molten pecan

[molten pecan]

Sorry I'm not familiar with mangio

[tranquil cliff]

What?

[brittle wing]

yo tell me dude

[rare goblet]

Ayo? @brittle wing level 2 !!!

[gaunt ruin]

@tranquil cliff

[tranquil cliff]

locally?

[molten pecan]

I guess you find the uninstaller on the root folder or delete all the files at once

[molten pecan]

Joking

[gaunt ruin]

yeah

[rare goblet]

Ayo? @gaunt ruin level 4 !!!

[tranquil cliff]

Me too

[gaunt ruin]

fck off arcane

[brittle wing]

@tranquil cliffgimme the link to the ai!!

https://youtube.com/shorts/dnQotnbHax4?si=shKjvyUwQ14BF1Oy
this one

[molten pecan]

Relax buddy 😭😭😭😭

[gaunt ruin]

search for AI voice changer w/ okada

[brittle wing]

no relax me wanna use the ai with my bois for sleepover

gonna do a little trolling :]

[brittle wing]

ok

ohh there it is

[gaunt ruin]

@tranquil cliff do you know?

how to

completely uninstal

uninstall* mangio

[tranquil cliff]

what should you do? delete what?

[molten pecan]

@low shard do you?

[gaunt ruin]

I want to completely uninstall MangioRVC locally, so that I can try with Apolio instead

I'm sick and tired of mangio and easyGUI

not giving it another chance

[brittle wing]

which one am i supposed to download?

last time i was having many bugs cuz i downloaded wrong one

[tranquil cliff]

Nick

Aiutami

[gaunt ruin]

you need to search for a tutorial man

[tranquil cliff]

Okk

[brittle wing]

just tell me dude

[low shard]

Nope😭😭😭 i think most helpers train and inference only online lmao

[tranquil cliff]

Io non so come funziona il local

[gaunt ruin]

bro, I'm focused on something else, I'll help you once I get rid of this local RVC problem

[brittle wing]

ok

[low shard]

Ma ce ne almeno uno di helper che fa in locale 😭😭

[molten pecan]

Wait helpers don't use it locally?

[gaunt ruin]

how? lmao, I tried and it doesn't work

[molten pecan]

Now that's interesting

[gaunt ruin]

I'm as surprised as you are

[tranquil cliff]

go to Windows (C:)

[low shard]

Not everyone has a powerful pc,im not saying every helpers uses only online but me and ange are

[tranquil cliff]

Kakao

[gritty gull]

Collab for inference??

[gaunt ruin]

ok, then?

[tranquil cliff]

I use ilaria rvc

[gritty gull]

you can send link for me?

[low shard]

You can use this for online inference https://discord.com/channels/1159260121998827560/1167844377046040769

[tranquil cliff]

go to the programs folder (not with the number)

[gaunt ruin]

didn't work for me

[gaunt ruin]

done, now?

[tranquil cliff]

yesterday it wasn't working today it was

[gaunt ruin]

not my case, guess its just luck

[low shard]

You js needed to refresh the models and the audio and re checkmark them,it can happen that it doesnt automatically checkmark sometimes

[low shard]

For me was working tbh today

[gaunt ruin]

alright fine, I'll give it another try

fuck

FUCKING GOOOGLE

I HATE GOOGLE BCS OF CLOSING THE FUCKING COLAB

Alright sorry

nvm

[tranquil cliff]

find the folder with the name of the program

[molten pecan]

Banned for screaming

[gaunt ruin]

Sorry lol, I had to vent

[gaunt ruin]

I’ll try it later, but I suppose the next step is just deleting this folder right?

[tranquil cliff]

Yes

[gaunt ruin]

Alright thanks

[tranquil cliff]

If you can't find it, contact me as I have another method

[gaunt ruin]

Alright thanks 🙏🏻

[pallid cloud]

quick Q, trying to use RVC but it's throwing an AssertionError (assert d == self.d) and an AttributeError (AttributeError: 'NoneType' object has no attribute 'dtype'). googled the error, someone else had it and it was caused by using a V1 model on RVC V2, but this is definitely a V2 model on RVC V2, any ideas?

[tranquil cliff]

I don't use local rvc so I don't know the answer but if you want you can try colab

[pallid cloud]

last time I tried to use Colab it killed the session and gave me the "no deepfakes" warning

if there's a working Colab I'd be happy to try it

[tranquil cliff]

for deepfake or to create models?

[pallid cloud]

just for deepfake, i'm using a pretrained model from huggingface

well, at the minute anyway

when the session was killed I was trying to train (I only have an old low vram card, it can infer fine but can't train)

[tranquil cliff]

for deepfake we have a guide

https://discord.com/channels/1159260121998827560/1163571683848900629

[pallid cloud]

oh i'm not trying to do, like, video deepfakes

[rare goblet]

Ayo? @pallid cloud level 1 !!!

[pallid cloud]

this is just for voice deepfakes / RVC

iirc the error linked me to google's "unacceptable use" policy page which mentioned deepfaking, i'm probably using the wrong word here

[tranquil cliff]

aaa okay

Do you want to create a voice model of a person?

[pallid cloud]

i have a pre-trained model i've downloaded off huggingface, i have an audio file i'm trying to infer with the model, but i'm running into the error i've previously mentioend

i've done this a few times before with other models, it's just this one that's broken

and i'm not sure if it's an issue with my version of RVC, an issue with the model, or something else

[tranquil cliff]

OK! I got it

https://discord.com/channels/1159260121998827560/1167844377046040769

[pallid cloud]

alright, i'll give that one a shot, thanks 👍

yeah no, looks like it might be an issue with the model I downloaded, the test audio just comes out unprocessed (female voice from a male model), and my audio just causes Gradio to say "error". thanks for trying to help anyway

[tranquil cliff]

send the model link?

[pallid cloud]

https://huggingface.co/MartinFLL/ai-voices/blob/main/cassie-20epochs-rvcv2.zip

[tranquil cliff]

this model is very strange

[pallid cloud]

it's the announcer from a video game (SCP: Secret Laboratory)

is it working for you?

[tranquil cliff]

All right

[tranquil cliff]

No

[pallid cloud]

strange, well, at least i know it's the model now 😅

[undone yacht]

Hello, I can't use rvc I can't hear my voice

[rancid quest]

Sorry if someone's asked this question already but does anyone have the files or a working link for ROD?

I'm trying to get TensorBoard so i can try and train some models

[brittle wing]

is this a voice changer

thing

[rancid quest]

?

[rare goblet]

Ayo? @rancid quest level 1 !!!

[winter seal]

could be

[thin bronze]

how can I fix "index search failed or disabled" error spam with realtime?

[rare goblet]

Ayo? @thin bronze level 1 !!!

[floral grail]

hello guys i dont know anything but RTH or INDEX what they do???

[proper shale]

pth is the model's main file, index controls accents and certain things the speaker does when they talk

[floral grail]

thank you for the information

[rare goblet]

Ayo? @floral grail level 1 !!!

[silver bobcat]

how do i make ai cover

[uneven elk]

@proper shale

[tranquil cliff]

https://discord.com/channels/1159260121998827560/1167844377046040769

[silver bobcat]

im stuck at smth i cant make goku sing

[proper shale]

?

[proper shale]

Stuck on what exactly

Elaborate

[silver bobcat]

idk

what Wave2Lip used for

i can make goku say but how do i make him to cover the song

[karmic heron]

Does anyone know a website or program that tests RVC voices? TTS for example

[proper shale]

thats for like, making person say stuff

not exactly for the cover

[silver bobcat]

oh so i need to add the music from background?

[proper shale]

yeah, u can filter the instrumental out if u want to

[silver bobcat]

thx you for information

[proper shale]

np ;)

[silver bobcat]

what language should i use for most realisitic cover

[proper shale]

don't think that makes a difference, but always good to use the language of the original speaker

[silver bobcat]

k

[rare goblet]

Ayo? @silver bobcat level 3 !!!

[silver bobcat]

is it good idea to put all lyrics in one or no

[proper shale]

Maybe not

[silver bobcat]

okay its not

now its error aaa

now the model is bugging

where mah goku

😭

[proper shale]

honestly i don't think u need wav2lip

just the cover itself is fine already, I'd say

[silver bobcat]

how do i make ai cover without iliaria rvc

[proper shale]

check pins and see AICoverGen

[silver bobcat]

k

[proper shale]

the link

[silver bobcat]

hina mod right

[late surge]

how do i put the weights.gg link into huggingface?

[silver bobcat]

dunno

[late surge]

i’ve got the files it just won’t accept them for some reason

[proper shale]

yeah

[late surge]

should i input the pth and index files seperately or as a zip?

[proper shale]

as a zip

[late surge]

gotcha

it still gives me a spanish error

[proper shale]

are u getting the right download link tho?

[late surge]

where would i get the download link

i’m just linking it to the model card link

[silver bobcat]

3 am no sleep

[late surge]

i’m giving it https://huggingface.co/jayvgc/apo

[rare goblet]

Ayo? @late surge level 1 !!!

[late surge]

swag

[copper pebble]

files> right click on the red box

and copy the link

[proper shale]

u gotta right click on yhus

this*

then copy link

[late surge]

that just makes me download it

[copper pebble]

apologue.zip sound scary

copy the link

yes that supposed to

the link i mean

[proper shale]

https://huggingface.co/jayvgc/apo/resolve/main/apologue (2).zip

[late surge]

the character is named apo and it’s mixed with dialogue, it’s what the dataset was called

[copper pebble]

.

[late surge]

alright thanks y’all

[copper pebble]

was just kiddin dw

[late surge]

appreciate it

[copper pebble]

:3

[proper shale]

np

;)

[silver bobcat]

after i upload model in AiGen now what i do

[neat chasm]

yo! would it be faster to use https://huggingface.co/spaces/TheStinger/Ilaria_RVC or run the proper files in macos 16gb memory?

[proper shale]

now u insert a yt link for ur song, refresh button, select ur model and click on convert

[proper shale]

Is it an M1 Mac

[neat chasm]

m2

m2 16gb

air

not pro

[proper shale]

hmmmm, i think probably yes for normal stuff, but no for ai cover stuff

i dont really recommend using local for mac rn, not optimized for the m series gpus

[fathom igloo]

Decided to train a model localy, but after choosing v2 the timer appears and goes on without anything happening ;-; What could be the cause?

[silver bobcat]

omg i love AI hub chan

i finally made my first ai cover 😭

[fathom geyser]

hi. anyone are using Google Colab?

I'm trying to train my model but in the middle of the process i got this: "ValueError: need at least one array to concatenate"

Does anyone know how to fix it?

Thanks?

🙏 🙏 🙏

[fathom igloo]

1/1 in queue 200+seconds. What to do ;-;

[fathom geyser]

Please help

[rare goblet]

Ayo? @fathom geyser level 1 !!!

[fathom geyser]

Sorry. I'm new here

[proper shale]

could be an old gradio version afaik

im not even sure how to fix that in this case

[proper shale]

which notebook?

RVC v2 Disconnected?

[fathom geyser]

https://colab.research.google.com/drive/1mrHh7uqWp43F5jJp6MOXPvumMho3Lj4X?usp=sharing#scrollTo=7vh6vphDwO0b

this one

[proper shale]

that is extremely old

[fathom geyser]

I got this

[proper shale]

and doesn't work anymore

use the ones in #1159949278270193734, specifically rvc Disconnected

but make sure to see the guide in #1159513888199540817 for it

[silver bobcat]

❤️ ❤️ ❤️

[proper shale]

;) happy to have helped

[fathom geyser]

Thank you 😊

[fathom igloo]

I updated it yet still the same 😦

[dense vortex]

I don't have the program working, I've installed it and set it up, but it doesn't change my voice.

@round fog

[dense vortex]

@zinc anchor

[rare goblet]

Ayo? @dense vortex level 1 !!!

[cosmic trellis]

how do you get them to generate an ai cover?

[rare goblet]

Ayo? @cosmic trellis level 1 !!!

[cosmic trellis]

bro i just started

[silver bobcat]

i spent 1 hour and 30 mins to make one keep find out

[obtuse creek]

sooo now weights.gg has its own link is there any sites that work with it

[round fog]

what program

theres 2

[sand iris]

#🔍│help-w-okada

[dim cape]

im having this problem which i think theres no solution for more than using a cloud service like google colabs, it tells me i dont have any memory left when im training, it's because my gpu is sad af lmao, does anyone know a good website for rvc on google colabs or how to solve this?

im using applio currently

[round fog]

i don't use Okada

[round fog]

pay the 7$

[dim cape]

fr 😢

[dim cape]

i used google colabs and its awesome now 😋

i just switch google accs when i run out of gpu

[copper pebble]

#🔍│help-w-okada

[silver bobcat]

what i can do when i got queue 3/3

[gaunt ruin]

currently testing ilaria, and so far, all I'm learning is that it's a pain in the ass not to have a console to see the progress of RVC. how can I make sure this thing is even working? lmao

@tranquil cliff

Fuuuck...

just great

fucking hell

[proven pecan]

Anyone have any good masculine voices ? I can’t find any

[brittle wing]

for some reason illaria doesnt work for me either 😭

[rare goblet]

Ayo? @brittle wing level 1 !!!

[proven hill]

1. why do you need the console for the inference(?)
2. did you duplicate the space?

[proven hill]

1. ilaria with one l
2. again, did you duplicate the space?

[brittle wing]

wdym duplicate?

and what space?

[proven hill]

are you using my rvc?

[tulip cloak]

What is your rvc?

[proven hill]

https://huggingface.co/spaces/TheStinger/Ilaria_RVC

[brittle wing]

yes

[tulip cloak]

Oh I see this is for making covers and stuff like that basically

[proven hill]

duplicate the space using this link: https://huggingface.co/spaces/TheStinger/Ilaria_RVC?duplicate=true

(make sure its on private) this way its much faster since only you can use it

[brittle wing]

ohh

thank you haha

[plucky locust]

https://huggingface.co/angam3x64/winxptournarrator/resolve/main/winxptournarrator.zip is not a registered model! Use /list to list all models, or use /register to register a new model.

[half cove]

#1159290666019471421

[modern snow]

No

[plucky locust]

Successfully registered model 22013 (From Huggingface URL)!

Use the model now like this:
/cover model:22013 youtube:https://www.youtube.com/watch?v=s9FCCiEs84k

Successfully registered model windows_xp_tour_narrator (Windows xp Tour Narrator (RVC v2) 500 Epochs)!

Use the model now like this:
/cover model:windows_xp_tour_narrator youtube:https://www.youtube.com/watch?v=s9FCCiEs84k

[copper edge]

guys, how do i fix the "Audio block passed." message in realtime-gui ?

[digital kettle]

When i try to open RVC-GUI, it reads
"Running with the system Python.
Traceback (most recent call last):
File "C:\Users\GABE\Downloads\RVC-GUI-main\RVC-GUI-main\rvcgui.py", line 4, in <module>
import soundfile as sf
ModuleNotFoundError: No module named 'soundfile'
Press any key to continue . . ."

How do i fix this????

[copper edge]

are you gabe newell ??? try reinstalling or sum

[digital kettle]

Ok. Also who is newell

[rare goblet]

Ayo? @digital kettle level 1 !!!

[gaunt ruin]

no I didnt

and also, I'd gladly like a console just to know if the space is actually DOING something

[gaunt ruin]

now THIS is the link I needed. tysm

nvm lol

it stays in loading rmvpe

bruh

@proven hill

[rare goblet]

Ayo? @gaunt ruin level 5 !!!

[brave garnet]

Colab di Google

⠀

• Colab per usare modelli già esistenti e trainare

Link

• Colab di backup nel caso la precedente dia problemi

Link

⠀

Made with love by Ilaria 💖

[gaunt ruin]

NVM IT WORKED FINALLY

[digital kettle]

Can someone tell me how to fix this? I don't know what to do and reinstalling didn't work

[proven hill]

oop

[gaunt ruin]

how do I download the applio RVC properly? I wanna do it right this time...

@proven hill lol, sorry for the tag

[molten pecan]

Want me to help you with that?

[digital kettle]

Uh can you help me with something? If you don't mind?

[molten pecan]

I don't know about RVC gui

Sorry

[proven hill]

i dont work with applio sorry, you should wait an helper

[digital kettle]

Its ok

[gaunt ruin]

I just found a video about it, it's in spanish but I understand it

https://www.youtube.com/watch?v=ssh-5dj3Hp0&t=597s

also, do I really need to downgrade my python version to 3.9.x to make it work?

[molten pecan]

You could try turning on automatic captions though

[gaunt ruin]

if I have 3.9.10 would that be fine?

[molten pecan]

No idea, I'm not familiar with python

[gaunt ruin]

fuuck

I really don't wanna downgrade

[molten pecan]

Wanna use the local RVC that I use that it's super simple to install

*?

[gaunt ruin]

which one is that?

[molten pecan]

https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/RVC-beta.7z

This one

You just need to unzip it and that's all

To open it you just execute go-web

[gaunt ruin]

I mean, I'm already doing what applio is requiring, so.. idk, is applio better? or is this one better?

in terms of functionality

[molten pecan]

Idk which one is better, but this one is stupidly easy to install

I think in terms of functionality both are the same

The HUD might be less polished but meh

[gaunt ruin]

oh ok, well, I'll try and get applio to run then, if it doesn't I'm using the one you just linked

[molten pecan]

Gotcha

Good luck!

[normal pulsar]

хахах

[molten pecan]

@half cove your dad is tweaking 😦

[bronze wave]

hi

[vagrant wharf]

damn

[gaunt ruin]

@molten pecan

do you know how to solve this?

[molten pecan]

Nope

No idea how to use Applio

😭

[ancient flower]

wait I forgot to check both boxes when opening RVC locally

how do I get to this screen again (or edit it to allow both boxes)?

[bronze wave]

im using rvc disconnected, but it cant find my dataset on google drive any help? did i named it wrong or something??

im getting this:

[rare goblet]

Ayo? @bronze wave level 1 !!!

[bronze wave]

@viscid moss

[viscid moss]

cHECKING...

[viscid moss]

Intenta cambiando el nombre del .zip a solo alexandra.zip

[bronze wave]

ok

[viscid moss]

es raro eso

debería funcionar

porque si montaste drive no?

[bronze wave]

sí

creo

[bronze wave]

sigue con lo mismo

[viscid moss]

puedes revisar en los primeros 5 pasos si está montado. Montar es el paso 5

[bronze wave]

sí está montado

[viscid moss]

Intenta reiniciando el entorno

En el colab

[bronze wave]

ok

[viscid moss]

Desconectate y vuelve a conectarte y ejecutar, el nombre dejale ese nombre mas corto

[bronze wave]

ya sirvió

gracias!!

[viscid moss]

De nada!

[gaunt ruin]

@viscid moss @tranquil cliff @copper pebble

[viscid moss]

mmmmmmm i'm not expert in Applio locally. But probably this guide will be helpful (but is in spanish):
https://youtu.be/WD6tLIPnuLw?si=wBwF2M5kHo1IaKnQ

I prefer Mangio

[gaunt ruin]

didn't work either

[viscid moss]

really?

[gaunt ruin]

yeah, idk wtf is wrong with my computer

I even have an nvidia 3060 gpu

so capabilities is not an issue

[viscid moss]

That's so strange bro

U try installing Mangio with the bat file?

[viscid moss]

good GPU

[gaunt ruin]

idk, can't remember

but it was sketchy af too

[wide parcel]

"Using cuda
An error occurred in the feature extraction, make sure you have provided the audios correctly."

any common reasons for this?

[viscid moss]

For Applio you will have to wait for another helper, if you want to try mangio I can help you

[viscid moss]

locally?

[gaunt ruin]

how do I install it

[wide parcel]

yep

[gaunt ruin]

like

properly

I'm SICK of having this happen

[viscid moss]

And Hi bro!

Triangle wave training?

[wide parcel]

yep

[wide parcel]

howd u know 😭

[viscid moss]

let me send u the github page

[viscid moss]

Inference and train right?

[bronze wave]

[viscid moss]

Are u using Applio or Mangio?

[wide parcel]

applio

[bronze wave]

ya no sirvió xd

[viscid moss]

can u show me the logs directory 👀

[viscid moss]

que paso es ese?

[bronze wave]

training

[gaunt ruin]

yeah

[viscid moss]

https://github.com/Mangio621/Mangio-RVC-Fork/releases/tag/v23.7.0
Download that bat file

[viscid moss]

En el que pones los epochs?

[bronze wave]

sis

[gaunt ruin]

I already did that... and everything beyond that is just messed up

you know what, fuck local, idfc anymore, or at least for today, I'll try again another day, FUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUCK

[viscid moss]

https://tenor.com/view/kanna-dragonmaid-dragon-maid-misskobayashi-gif-13580462

[wide parcel]

if u wanna get into it you can do a fresh install of drivers n packages or use a vm

[wide parcel]

js in general tho

have u heard of it before?

[viscid moss]

algo falló en los pasos anteriores

[gaunt ruin]

using a vm is gonna explode my machine

[viscid moss]

Like 2 times

[wide parcel]

i mean if you got the storage you could just dual boot 😭

[bronze wave]

entonces hago todo d nuevo?

[viscid moss]

And I deleted all the model files and started again

[viscid moss]

por desgracia si

😦

[bronze wave]

nimodo

[gaunt ruin]

lmao, 50 gb of storage

[wide parcel]

tiny11 and tiny10 are both 4gb

[gaunt ruin]

hm...

[wide parcel]

a full install will run u like 30

-50

so uh

only thing stopping u is windows bootloaders

[gaunt ruin]

I don't really wanna mess with my computer using vm... honestly, I'm tired of trying local RVC, I think I'll resort to luck using colabs... although they're not trustworthy as they were with google collab... FUCK GOOGLE

[wide parcel]

i mean

what version of windows u on rn

and is it debloated

[proven pecan]

Anyone have any good masculine voices for TTS that would work good for audiobooks? I can’t find any

[viscid moss]

for TTS or something like that?

[proven pecan]

Yeah

[viscid moss]

hmmmm

[wide parcel]

i made one of a dude i know but i doubt he wants it released
if you can find raw recordings of a guy youre looking for lmk and i can train it

i can preprocess too so raw audios fine

[viscid moss]

I only 've female voices 😅

[proven pecan]

I have actually!

[wide parcel]

shoot me a dm or link it here then :)

[gaunt ruin]

windows 11

[wide parcel]

nah what edition

[proven pecan]

When I add voices I download, why don't they sound anything like they are supposed to?

[glad zealot]

If you think it's a virus, you can check the code yourself, it's open source anyways, you can check the whole code for the app for any virus

[violet heron]

It’s open source, so you can look inside the code if you think it’s bad

[glad zealot]

People becoming goblins?

[violet heron]

That’s why it’s on GitHub in the first place

[glad zealot]

Dang it

[ancient flower]

trying to train locally and keep getting this message, do I just not have enough RAM on me?

[viscid moss]

You don't have enough VRAM

GPU Memory

Try again with a lower Batch Size

But 4GB.....

[glad zealot]

8 - 14 is the one I use but idk about what your gpu can handle

[viscid moss]

💀

[viscid moss]

The RTX 4090 can handle 45 🔥

Literally a beast

[glad zealot]

How fast tho

The higher the slower it goes right?

[viscid moss]

I'm not totally sure about that

can I send pics of that here?

[glad zealot]

I wonder what the max on colab is

Batch size

[viscid moss]

Yes, when I used 45 Bach once it disconnected due to a lot of use at 100%

[viscid moss]

25 maybe 🤔

I use 20 for colab

[glad zealot]

Lemme try that on my next model

I use 14

[rare goblet]

Ayo? @glad zealot level 31 !!!

[glad zealot]

It's probably gonna be a 1 hour dataset

[viscid moss]

And batch size for 20 - 25 min? 👀

[rare goblet]

Ayo? @viscid moss level 29 !!!

[glad zealot]

Since imma do the whole series FINALLY for a dataset and it's the main character

Its safe to say I'd prob get 1 hour or more

No idea of colab can even reach 150 with a 1 hour datset

[viscid moss]

noted!

[glad zealot]

It's like 3 hours 30 mins from what I see is the timeout for colab

90mins to afk (lower most of the time)

Ahhh

[viscid moss]

[glad zealot]

Welp to kaggle I go

It can do 12 hours no user input and closed browser

Also faster? Not sure about this

It uses T4x2

[viscid moss]

Have you considered payment methods for training? 👀

[glad zealot]

Naaa

Got no money to do paid stuff

[viscid moss]

I've tried to train with 2 GPUs, it's broken

Only 1 works and the another Zzzzz

[glad zealot]

It's probably only uses 1 of the 2 t4 then?

Can't.. I'm using it with the save-all method which doesn't let you access the output stuff

[viscid moss]

ya

[glad zealot]

So theoretically you can probably do 2 models using threading?

I wanna test it but it takes sooo long to load requirements

[viscid moss]

I use payment methods to train. I tried training using 4090x3. I thought... great 72 GB of VRAM, and training with GPUs, but, I don't know why Applio was only using one even though it had 3

[glad zealot]

Welp I guess imma try a 30 mins dataset

[viscid moss]

So I asked an Applio Dev and he told me that Applio only supports one GPU to train effectively.

[glad zealot]

By default applio only uses 1 gpu . But I think you can make it use more using the gpu selector like gpu: 0-1-2 since it's 3 gpu

Not sure tho

[viscid moss]

theoretically

Yes, the GPUs appears but don't use them all, it's strange

[glad zealot]

@karmic oxide can ya help cleaning the dataset? I still have no idea how to do it properly and with my hardware I probably can't do it well

[viscid moss]

I've tried and don't works

So I only rent several 4090s at a time

[glad zealot]

Use it

All of them

[viscid moss]

hmmmm

[glad zealot]

Gonna do what I can first then give it to you.. probably 30mins-1hour? If I don't get too lazy making it tho

[ancient flower]

alright I think I just can't train locally because I'm short on vram. Batch of 2 wouldn't work, and setting it to 1 has yet to progress at all after uh, this many seconds

ye

[rare goblet]

Ayo? @ancient flower level 2 !!!

[viscid moss]

I think that using colab is better for u

[glad zealot]

You want the whole 3gb WAV file for the whole anime?

The cut voice?

Ahh oke

I probably won't use it with other characters talking over the main since it's a 4 hour audio anyways

I can just ignore some like reverbed audio

[oblique venture]

any tip to prevent randomly sound broken, or distorted sound when cover song?

[brazen schooner]

Does anyone know why I get this error?

[sand iris]

spaces in folder/ file names

[slate pawn]

Tensorboard experts share me your wisdom

Is this an overtraining point?

[hard gust]

seems like it is

AMD realeased an update with their newer Windows drivers, and now DirectML supports AI cores for Stable Diffusion for example. Is it affecting RVC & Applio? Now can we train finaly?

[sand iris]

No training

The RVC side has yet to get a non-pytorch backend to train on

[brazen schooner]

I am training locally for the first time, any tips or advice?

so far the feature extraction process is busy. and a few of the threads being shown have been completed but 4 of them are taking 16 minutes now to finish. idk why but i hope they finish

they are taking longer than expected

is it normal for this process to take this long?

i have 12261 MB vram gpu

any help or guidance is appreciated

[brazen schooner]

btw thank you. i will take note of that

[tulip cloak]

???

I thought the more the dataset, the less the epochs

[glad zealot]

Same . When I'm using a 1 hour dataset I only do 50-100 epochs

[pine mango]

yo does anyone have a notebook to make new models with?

[brittle wing]

Have you managed to fix it? I also have this same issue, and I'm running macOS Sonoma on M1

[crisp plover]

Hey what is the best inference option out there for singing vocals ?

Right now i use RVC and select RMVPE since that is the best option that it offers.

Is there anything better than this for more clear singing vocals

[rare goblet]

Ayo? @crisp plover level 2 !!!

[verbal oxide]

anyone know how to fix this?

[glad zealot]

use mine :D

no auto backup tho

[silver bobcat]

how can i fix error:Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.aspx

halp mee

[sand iris]

have you considered reading the error

... Or do you not have an nvidia card

[silver bobcat]

i dunno

[rare goblet]

Ayo? @silver bobcat level 4 !!!

[silver bobcat]

it was fine yesterday

[copper pebble]

do you have an nvindia card?

[silver bobcat]

idk how do i know

i think i do

[copper pebble]

you really dont know if you have a gpu card?

i think you dont then

[silver bobcat]

ye

o

[copper pebble]

did u need to train?

[silver bobcat]

what train?

[copper pebble]

ok im outta this

[silver bobcat]

a

[thin bronze]

Anyone?

[copper pebble]

were you serious abt the gpu thing?

or were u just kiddin

[lavish sable]

does anyone have up to date good tutorial for createing an RVC model?
specificly i want to model my voice

[tranquil cliff]

Hi rickie

[lavish sable]

I Have a strong GPU so doing it Locally isnt a problem if it results in higher quality model.
if there are any tips or tricks for good voice samples id love to know aswell

[rare goblet]

Ayo? @lavish sable level 1 !!!

[copper pebble]

#1159513888199540817

[lavish sable]

thanks. i couldnt find this channel for some reason..

[copper pebble]

you can ping me if needed

[lavish sable]

Thanks !

[brittle wing]

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• RVC-HF, by r3gf Huggingface Spaces
• AICoverGen, by r3gf Huggingface Spaces
• RVC v2 Huggingface version, by Clebersla Huggingface Spaces
• Advanced RVC Inference, by neuclya Huggingface Spaces

Need some help? 🤔

You can find more info on the #1159513888199540817 channel. If you can't find your answer, feel free to ask for help in #✨│ai-help. Credits to Faze Masta and Antasma for compiling these links.

More commands: -audio, -local, -overtrain, -realtime, -rvc, -help

[brazen schooner]

Why is the feature extraction taking so long?

Its been 3 hours and I'm not even able to start the training process. It's been stuck like this for almost 2 hours

[oblique rain]

Why are you using crepe

[brazen schooner]

is it a problem?

im not really familiar with the options so i chose crepe cause i previously on colab used to use crepe

[oblique rain]

I mean, it's not the best extraction method here and --it can be a reason why it's taking so long

[brazen schooner]

so what do you think is best?

also i need high quality models

[oblique rain]

Can you tell me your GPU?

[brazen schooner]

i worked so hard on my datasets

i have 12261 MB vram gpu

nvidia rtx geforce 3060

[oblique rain]

It depends to:

1. Method Extraction
   1,5. If you selected Mangio Crepe, it can be because you selected a small number of Hop Length
2. GPU VRAM
3. Dataset's audio length (minutes)

[oblique rain]

RMVPE and Mangio Crepe

[brazen schooner]

ive never used rvmpe or mangio crepe when training

also i dont undertsand the hop length

today is my first day doing training locally

my datatsets range. some are short and some are 30 minutes or an hour or 25 minutes or maybe even 15 minutes. it really varies

[proper shale]

Hop length is the amount of times it checks for pitch variations

Idk in what time scale but I believe it's per second

Only mangio-crepe and crepe use this hop length afaik

[brazen schooner]

so what do you think i should set it?

i gave up for today cause it took forever and i want my laptop to rest. maybe ill try again tomorrow

[brazen schooner]

do you think i should leave the hop length as is?

also which is better between mangio crepe and rvmpe?

[proper shale]

I think keeping it at 64 is alright for a normal dataset

[brazen schooner]

thanks

[proper shale]

1. Is your dataset fully denoised (and I really mean that - going to the espectrogram and denoising to the best of your abilities?)

If so - mangio-crepe might be better. But it's very noise sensitive and it has real advantages when you use low hop-sizes, which takes way too long for big datasets. RVMPE is faster and has quality that is almost on-par with mangio-crepe, even with noisy datasets.

Up to you to decide which one is better, though. Mangio might have less artifacts but it'll take way longer for it to get real quality improvements over RVMPE.

[coral python]

Hello i just downloaded MMVCServerSIO_win_onnxdirectML-cuda_v.1.5.3.14.zip (My GPU is Intel(R) UHD Grapics 630 and Nivida GeForce GTX 1660 Ti) and have python,Pytorch and cuda sett up. But when i want to start the client with start_http it just does nothing any advice?

[brazen schooner]

I made all of the vocals as isolated as possible with UVR5. I don't know what the espectogram is, sorry. I'd say my vocals are like 86% clear. Some of then are 100%.

Thanks for the advice. I will try again tomorrow. Maybe I'll do rvmpe. Have you had experience with these options? How long did process take like feature extraction and how long did training take? For you?

[rare goblet]

Ayo? @brazen schooner level 5 !!!

[proper shale]

Wrong version

Download this one:

https://huggingface.co/wok000/vcclient000/resolve/main/MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.15.zip

Make sure all folders have 0 spaces in their name and don't move the start_http file outside of the folder

[brittle wing]

why do i get HTTP Error 403: Forbidden everytime i want to download a voice model

[hard chasm]

Where are you downloading it from

[proper shale]

Link was probably shut down

[brittle wing]

It's up and running https://www.weights.gg/models/clnssiuze00v1ws2ltfu50xa9

im using hina's colab

[proper shale]

u can't use the weights gg link directly for download on colab

u need to have the download link copied

[brittle wing]

https://models.weights.gg/clmo1678g001qwsw6dtf8955i.zip

[rare goblet]

Ayo? @brittle wing level 3 !!!

[proper shale]

yeah that

[brittle wing]

tried and didn't work even after I turned off vpn

i wanted to use rejekts's but cant get ngrok part to work

[proper shale]

You can try a download and upload it to your Google Drive

Then just configure the share settings correctly (everyone with link should see), copy link and try it.

[brittle wing]

how long should my dataset be?

[proper shale]

At least 10 minutes.

[brittle wing]

how many epochs?

for a 10 minite model?

[proper shale]

It varies a lot

I can't really tell you because every training session is different, really.

The best indicator is TensorBoard, the graphs will "tell" you when to stop training

[brittle wing]

its just a error on google colab..... 404 not found

[silver bobcat]

what i can do when i got error: nvdia card...... bla bla bla it was fine yesterday

[brittle wing]

200 ephocs for a 10 minite model?

i have a nividia gpu

[silver bobcat]

i guess i dont have it

[proper shale]

I think you hit the GPU time limit

[proper shale]

You have to test the model every 50 epochs then, see if it sounds better than the last checkpoint.

[brittle wing]

how can i contune training rvc beta

@proper shale

[proper shale]

Insert the same settings, no need to feature extract or preprocess again, keep same batch_size.

You can change saving frequency and total training epochs.

[brittle wing]

just found out the reason why it was throwing 404 error, it only accepts Huggingface and Pixeldrain download links

[proper shale]

Oh

💀

[subtle rover]

step 2a in train
not receiving any output
my folder names have no spaces and the audio files are all wav files
on windows btw

[proper shale]

@sand iris Know about this?

[subtle rover]

its just blank

[proper shale]

oh...

uh

see

Program Files has a space

[subtle rover]

im an idiot

lollll

sorry for wasting ur time

lemme fix it and see if it works

[proper shale]

nawww its ok bro haha

i suggest moving it to downloads or something

not really good to rename Program Files i believe

[subtle rover]

yup

lemme try

its working! thx!!!

[proper shale]

Yayy, glad to have helped ;)

[sand iris]

https://cdn.discordapp.com/emojis/863104661342781481.webp?size=48&name=lol~4&quality=lossless

[proper shale]

[silk tide]

i need help

[copper pebble]

sup?? ^^

[rare goblet]

Ayo? @copper pebble level 23 !!!

[silk tide]

i cant upload a model

its not working

[copper pebble]

where r u uploading it?

[silk tide]

edit

[copper pebble]

ohh

[silk tide]

im new to this thing

[copper pebble]

u doing a cover?

[silk tide]

umm

what does that mean

check dms

[copper pebble]

r u trying to use a model?

and also what r u using for it?

[subtle rover]

no options in inferencing voice even after refreshing

how to import my file?

[proper shale]

is it in the weights folder?

[subtle rover]

nope lemme move it there

[rare goblet]

Ayo? @subtle rover level 2 !!!

[subtle rover]

thx

[proper shale]

np

[subtle rover]

still nothing

is it supposed to be wav as well?

lemme convert it

[proper shale]

Nope

[subtle rover]

ohh

[proper shale]

Oh wait, that should be your model's file

so your .pth

[subtle rover]

so i should move those to the weights folder?

theyre currenly in "uvr5_weights"

[proper shale]

no those are like

uvr stuff

your model's pth should be in weights

[subtle rover]

where do i find that

[proper shale]

unless u haven't trained it or checked the option to save a small finished model to it

[subtle rover]

ive trained it

already

[proper shale]

i think u didnt check the "save a small model to weights" option

[subtle rover]

okay let me do that

theres nothing in the weights folder

where does it get saved?

[proper shale]

oh u just trained the index

u have to train the model now

[subtle rover]

😭

should i click on train model?

[proper shale]

Yes

[subtle rover]

processing

[proper shale]

check cmd

[subtle rover]

still processing

[azure marsh]

RVC Guides (How to Make AI Cover)

Documentation

🇺🇸 English (main)

https://docs.aihub.wtf/

​

Translation by country

🇧🇷 Brasil (PT-BR)

https://docs.aihub.wtf/v/brazil

🇪🇸 Spain

https://docs.aihub.wtf/v/spain/

🇫🇷 France

https://docs.aihub.wtf/v/france

🇩🇪 Germany

https://docs.aihub.wtf/v/germany

🇯🇵 Japan

https://docs.aihub.wtf/v/japan

🇰🇷 Korea

https://docs.aihub.wtf/v/korea

🇵🇱 Poland

https://docs.aihub.wtf/v/poland

🇷🇺 Russia

https://docs.aihub.wtf/v/russia

🇮🇹 Italy

https://docs.aihub.wtf/v/italy

[subtle rover]

how long does it typically take? its still processing

[proper shale]

oh i think something went wrong

i dont think its training

but idk what exactly

[subtle rover]

🥲

[proper shale]

ooohhh

insert the GPU to use for training

should be 0

[subtle rover]

thanks!!!

[rare goblet]

Ayo? @subtle rover level 3 !!!

[wraith stirrup]

Hello guys i dont have a good pc so i am using a online version of this software in google collab . so i gotta problem . when i listen to my voice it stucks sometimes in some words but when i record the voice it wont stuck in words so is there any solution of it

Is someone here ? 😩

[pine mango]

yeah so file was supposed to appear under weights

and its not there

[violet heron]

Train index and save weights to drive

Then check your drive see if there

[pine mango]

redoing

yooo

[wraith stirrup]

?

[violet heron]

Which colab?

[silver bobcat]

why i need gpu card to make ai cover

[wraith stirrup]

I ll send video wait

[silver bobcat]

what is train model

[violet heron]

Ai is super resource intensive

[pine mango]

this aint it but lemme redo this whole thing

[violet heron]

Means to make a model/voice

[pine mango]

maybe that changes something :thinkage:

[violet heron]

Did you save to drive

[silver bobcat]

o

how do i make aicover without aicovergen?

since i got no gpu card

[pine mango]

im dumb i forgor im do that

[violet heron]

HF Spaces

I’ll link

[silver bobcat]

giv

[wanton swan]

Does lowering Hop Length actually make the voice more accurate on Applio?

[violet heron]

https://thestinger-ilaria-rvc.hf.space

[pine mango]

bro how do i do that

[silver bobcat]

there just some model

aicovergen so good but no gpu card 😭

miku in this is hatsune or nakano?

[violet heron]

Download model

Those are models people used in the last 2 days

[wanton swan]

[violet heron]

Go to download model and add your HuggingFace.co model download link

[silver bobcat]

i see

[wraith stirrup]

can you check this video and help me to fix this problem

[silver bobcat]

damn in this web i gotta tryhard since no auto cover

[pine mango]

bandicam goes crazy 🥶

[wraith stirrup]

hehe 😄

[pine mango]

anyways im going crazy tryna find this fucking save to drive option

[silver bobcat]

how do i make my hugging face model download link

[wanton swan]

Is crepe the best option for extraction in Step 2?

[crisp plover]

Is there a way to convert my model instant to a youtube link and replace the voice and keep the instrumental ?

[wanton swan]

[pine mango]

bro does anyone where the fuck save to drive is

and how do i do it :-;

[wanton swan]

wait what

[wraith stirrup]

Hina modified

[rare goblet]

Ayo? @wraith stirrup level 3 !!!

[violet heron]

It should be the link on download

[violet heron]

There’s no output

Also start bottom Left

[wanton swan]

Hello????

[wraith stirrup]

Have you checked full video ?

It skip some of my words while listening live but it wont skip any voice while recording

[violet heron]

Try turning up in or out

[wanton swan]

Hello???

I have a question Hello?

[wraith stirrup]

Yes blaze

[wanton swan]

Is crepe the best option for extraction? Also, does lowering the Hop length make an accurate voice to the data?

[wraith stirrup]

Sorry yall I really dont know i am noob at this

[wanton swan]

Where's THE MODS!!!!

[wraith stirrup]

Yeah only one mod is talking

[wraith stirrup]

Is there any button of turning up or out ?

[sage sail]

for me, yeah
i use it all the time, its pretty decent

mangio-crepe is also a good option

[wanton swan]

In Applio?

OHHH

[violet heron]

Top/middle right

[violet heron]

RVMPE

[wraith stirrup]

Hmm i found it but its wont making any changes 😩

[wanton swan]

Wait, what about the Hop Lengths do you leave it at the default or lower it for a more accurate pitch?

[violet heron]

Try turning up chunk to stmn like 320

[violet heron]

Default

[wraith stirrup]

Ok ill try wait

[violet heron]

Hop Lengths are only for Crepe/Mangio-Crepe

[wanton swan]

Does RVMPE GPU make a difference or just stick with RVMPE?

[rare goblet]

Ayo? @wanton swan level 4 !!!

[violet heron]

RVMPE GPU trains more on the GPU

it will end up the same

normal RVMPE trains a bit more on the CPU

[wanton swan]

Oh okay, what about people saying crepe and harvest are really good? What do you think of those extraction options?

[violet heron]

Crepe is meh

RVMPE is basically harvest v2

[wanton swan]

Oh

[violet heron]

Also harvest is extremely slow

And raspy

Though

[wraith stirrup]

Bro it was not working in 320 it skips more then 128 chunk

[wanton swan]

I have 92 seconds of data, Should I put the epochs on 230 or lower?

[violet heron]

Try less then 128

[violet heron]

That’s probably not gonna be a good model

You can try 250

[wraith stirrup]

Can you check my voice model i ll send you may be i train wrong thats why its skipping?

[wanton swan]

Wait, 4 mins

I heard it has to be 2 to 10 mins, but 4 mins from 250 secs is good?

[violet heron]

We’ll really depends

I’ve gotten good models with 250 on short datasets and long ones

[wanton swan]

Was it like ElevenLabs Level quality with 250 secs?

[violet heron]

Give this man Low IQ Elevenlabs user role

[wanton swan]

Thanks Pal

So the more data the better?

[silver bobcat]

guys how to create your huggingface link

[wanton swan]

so with the 250 secs, what epoch range would you give it like 200 - 300 epochs

[violet heron]

Usually

[violet heron]

I would

[silver bobcat]

what is homepage

[violet heron]

What model are you trying to use, send a link

[silver bobcat]

zip file?

[violet heron]

Where’d you get it from?

[silver bobcat]

this sever

[oblique mason]

Whats the best settings?

[final cloak]

is w-okada goated?

[wanton swan]

Okay thanks

[silver bobcat]

is there any web that like aicovergen?

[cobalt reef]

Whenever I open the .bat file, this error appears

[7512] Failed to execute script 'MMVCServerSIO' due to unhandled exception!

[silver bobcat]

someone halp

[wispy burrow]

#🔍│help-w-okada

[cobalt reef]

oh, ok!

[wispy burrow]

Are you asking how you'd get the download link for a Model?

[next oxide]

can someone help

when i open the rvc realtime voicechanger

it dosent load the models it used to work but idk what happened

[wispy burrow]

#🔍│help-w-okada

[brittle wing]

How do I use models I downloaded on websites

[wispy burrow]

Do you have RVC locally?

[silver bobcat]

yea

[wispy burrow]

Right-Click the download button and select 'Copy link'.

[silver bobcat]

oooo

[rare goblet]

Ayo? @silver bobcat level 5 !!!

[silver bobcat]

i downloaded but i dont see the model,what i do

[tranquil cliff]

Hiiiii

[silver bobcat]

do i need to wait after refresh

hiii

[tranquil cliff]

need assistance?

[silver bobcat]

ye

[wispy burrow]

Don't think you needed to download it.
Just copy the link and paste it in AICovergen.
You are using AICoverGen, no?

[silver bobcat]

nope

using ilaria

[wispy burrow]

What are you using?

[silver bobcat]

ilaria

[wispy burrow]

Did you read the guide?

[silver bobcat]

i do but i didnt read all

so i need to read it again right

[crisp plover]

Is there a way to convert my model instant to a youtube link and replace the voice and keep the instrumental ?

[wispy burrow]

It is as I thought.
You need to copy the download link and paste it in Ilaria RVC

[silver bobcat]

nvm it work now thank for help bloke

[wispy burrow]

I think AICoverGen does that.

[brittle wing]

can yu help

[crisp plover]

Owh yeah ofcourse nice, can you use that in colab still ?

And can you run that local as well ?

[wispy burrow]

#1163883017446629508
Apparently, yes.
You may want to check that or Replay.
One of those 2 should be able to do that.

[silver bobcat]

is there any way i can fix error: gpu card in aigencover?

[wispy burrow]

Depends on the issue...

[crisp plover]

Wich one do you recommand for the best quaility ?

[brittle wing]

how do i do a ai cover

[wispy burrow]

I think the quality should be the same.
They are basically running on the same code (RVC).

[crisp plover]

Alright and is replay safe never heard of it before lol

[silver bobcat]

is there any way

[crisp plover]

You did install it local right ?

[silver bobcat]

ye

[brittle wing]

i dont know

[rare goblet]

Ayo? @brittle wing level 1 !!!

[wispy burrow]

I read "How do I use AICoverGen"...

[brittle wing]

can yu still use it on mobile?

[silver bobcat]

so i cant fix the error gpu card thing?

[silver bobcat]

nvm

[wispy burrow]

Well you will need an RVC "program".

[brittle wing]

i dont get what yu r saying gang😂