✨│ai-help

[pastel gorge]

boleh tahan xd

[brazen ether]

chindo

[pastel gorge]

berhampiran kawah

[pastel gorge]

oh

[brazen ether]

do you speak hokkien?

btw balek ke topik, saluran nya apa ya?

[pastel gorge]

tidak

[brazen ether]

i see

[pastel gorge]

malangnya

[pastel gorge]

ia adalah dalam talian

cari w okada

penukar suara

ia sepatutnya dalam github

ya>

[brazen ether]

github?

[pastel gorge]

laman web pembangun github

tempat kod diterbitkan

[rare goblet]

Ayo? @pastel gorge level 12 !!!

[pastel gorge]

atau cari di youtube?

[brazen ether]

apakah bisa memakai di live streaming jg?

[pastel gorge]

ya

ia berfungsi secara langsung

tetapi memerlukan banyak sumber komputer anda

sangat sukar untuk diuruskan

tetapi

ya

jika komputer anda tidak baik

saya tidak mengesyorkan

ia bekerja keras

suara.ai

ia bagus

walau bagaimanapun,

tidak selalu disyorkan

untuk komputer riba

dan apa lagi

komputer apa yang anda ada?

[brazen ether]

mungkin gbs

[pastel gorge]

Oh begitu

[brazen ether]

dikarenakan komputer saya ram nya mmg besar

[pastel gorge]

OH

[brazen ether]

cuma cpu, disk, gpu

gampang penuh

[pastel gorge]

TETAPI BUKAN SAHAJA RAM YANG ANDA PERLUKAN

[pastel gorge]

gpu anda

[brazen ether]

main game udah makan resource

[pastel gorge]

gpu

bagus?

[pastel gorge]

saya faham

[brazen ether]

saya memakai hdd

sudah pasti sukar

[pastel gorge]

oh ya

sukar

sesungguhnya

bagaimanapun

saya kena pergi makan

ya?

selamat tinggal sekarang

[brazen ether]

okok seppp

[pastel gorge]

guna internet sahaja

[brazen ether]

thank you

[pastel gorge]

sangat berguna

xd

[pastel gorge]

tiada masalah, dan terima kasih

untuk bersembang

menghargainya!

selamat tinggal!

[brazen ether]

btw ada yg tau knp voice.ai gbs?

[patent pasture]

voice.ai is bad

[brazen ether]

why bad?

@pastel gorge klo lg ga sibuk, boleh voice call?

[pastel gorge]

sorry im busy right now

and i prefer to not vc

[brazen ether]

Ahh i see

Sorry

[pastel gorge]

no problem

[rain glade]

Can someone help me covert vocals I want to sound like Drake

[proper shale]

Read the Ilaria RVC guide in #1159513888199540817

[balmy citrus]

someone who can make high quality rvc models? dm

[wraith dove]

when im in the program i cant hear the voice after i say something
i have pressed start

[static gyro]

could i possably get help on a colab issue im facing?

I am a colab pro holder, which means i have bought minutes to train models.

the last rvc train i tried to do just stopped unexplainably. the last 2 times.

it was training an epoch 1 every 30 seconds
both times it stopped around epoch 200

im using RVC disconnect

the model has 10 minutes of training data

[copper pebble]

thats weird

[naive fossil]

im using applio, my infer doesnt come out good at all, these are my settings

[copper pebble]

use rmvpe not rmvpe+

[copper pebble]

u ita?

[naive fossil]

yes

[static gyro]

u think i can use the old rvc with autosave so it can autosave? now that i have minutes

[naive fossil]

let me try

[copper pebble]

fammi sapere

[copper pebble]

yh the easy gui 10/25

[static gyro]

whats that?

[copper pebble]

the colab where u can save ur progress

[static gyro]

im not familiar with that one

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[naive fossil]

questo è normale?

[rare goblet]

Ayo? @naive fossil level 5 !!!

[static gyro]

its not listed here

[copper pebble]

non hai selezionato l'index

ecco pk

[naive fossil]

ma è qui

[copper pebble]

shit aint the link rn

just search on google easygui 10/25 u should find it

[copper pebble]

mhh

[cerulean dew]

so in terms of trying to match AI vocals to the original instrumental, what pitch changes in RVC will still harmonize with the original?

[static gyro]

im not finding it

[copper pebble]

https://colab.research.google.com/drive/1r4IRL0UA7JEoZ0ZK8PKfMyTIBHKpyhcw this should be it

[static gyro]

ok yeah its on there its just not named the same

[naive fossil]

@copper pebble mi esce veramente male, e no non credo sia problema di pitch

mi ricordo che quando usavo collab quando c'era la easy gui di gradio funzionava da dio

[copper pebble]

pk non provi con ilaria rvc o la nuova colab per le cover?

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[naive fossil]

perchè ci sta veramente tanto tempo con quello di ilaria

qual'è il collab nuovo?

[copper pebble]

prova la prima in questo elenco

[naive fossil]

provo e ti dico

e qui?

se metto un file da youtube non ho solamente la voce

[static gyro]

im stuck here

[mossy eagle]

So, are these voice clones only able to be used for AI song covers, and not TTS?

[naive fossil]

@copper pebble

[violet heron]

They can be used for both

[copper pebble]

fa tutto da solo

[violet heron]

But TTS sounds very meh compared to just talking into your mic and converting that

[naive fossil]

questo è figo

[copper pebble]

https://discord.com/channels/1159260121998827560/1175615544037212200

[copper pebble]

si in 10 minuti fa tutto

[naive fossil]

cambio anche il pitch all?

[copper pebble]

wth again with the zip problem??.. try to rezip ur dataset

[copper pebble]

dipende se è richiesto dal modello/canzone si senno no

[naive fossil]

la canzone è a 0 e la voce che metto la voglio a -4

[mossy eagle]

Is there any collab tools that tools that can use the clone voice as TTS?

[violet heron]

Not any working colabs at the moment I don’t think

Ilaria RVC on HF Spaces has TTS support

Same with Applio which runs locally

[static gyro]

does it need to be a folder?

[copper pebble]

yh

[static gyro]

ok its not a zip file reader

[static gyro]

did not start train

Authtoken saved to configuration file: /root/.config/ngrok/ngrok.yml
Tensorboard NGROK URL:NgrokTunnel: "https://c24f-34-143-238-169.ngrok-free.app" -> "http://localhost:8888"

[gaunt current]

anyone know how to get the index file to generate? i followed the tutorial step-by-step and the index file doesnt seem to be generating. just executes these two lines in the console and that's it

[small sail]

wth!?

where the 3 dot!?

[violet heron]

You didn’t convert it yet

[small sail]

i did...

welp imma do it again ToT

[wide breach]

Yes

Nope you don't need partnership

[icy vector]

free website for rvc? (im broke and have a shit pc)

[molten pecan]

https://rentry.org/HFguide_en I made a guide

Good luck!

[low shard]

hey julia btw what does it change from ilaria rvc to aesthetic rvc? i never used the 2nd one

[icy vector]

i got the space , where to train your model?

[molten pecan]

Ah you can't train using HF. For that, use RVC disconnected

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[molten pecan]

With the Aesthetic one you can paste links from mega and google drive, or upload the .pth and .index file manually if you want

[low shard]

ohh cool

[molten pecan]

That's what I noticed so far

[coarse steppe]

its says python not found but i have python downloaded

can anyone help

[icy vector]

i give up doing ai covers , if kits.ai didn't become that greedy i would still do ai covers

[low shard]

in fact we dont collab anymore with kits ai from a while and we dont allow their models,also there are more new free ways to make ai covers

[icy vector]

easy ways?

[low shard]

https://discord.com/channels/1159260121998827560/1167844377046040769 this tutorial could help you its free and unlimited

[icy vector]

i tried it before but it shows other people's models and audios , i think its kinda intrusive for privacy

[violet heron]

You have to duplicate the space that’s why

[low shard]

you have to duplicate the space to have your own private version

[violet heron]

Many people don’t for some reason

[icy vector]

well i duplicated it says ''aborted''

[low shard]

can you show me a screenshot?

[icy vector]

its gone

i refreshed the page

[low shard]

try to duplicate now and see

[icy vector]

it works

where to train?

[low shard]

the hugging face rvc space you cloned is only for making ai covers, this is for training models: https://discord.com/channels/1159260121998827560/1160926856548200531

[static gyro]

how do i download the epochs i saved?

how do i download the epocs saved?

[sweet siren]

Bro i have lyrics but I need ai to sing it like a sad song

[rare goblet]

Ayo? @sweet siren level 1 !!!

[fossil glen]

go to fiver or sing your self then inference on rvc

[static gyro]

i dont see that

[fossil glen]

go to folder

[static gyro]

im looking in the location too. there is no weights

[static gyro]

what folder?

[fossil glen]

on left menu

[static gyro]

k

[static gyro]

what is main?

[fossil glen]

rejecks colab?

[static gyro]

no this is 10/25

[fossil glen]

do you run this

[static gyro]

i was guided here after a model was interupted while it was training

[static gyro]

yes

[fossil glen]

already on gdrive

[static gyro]

here.

[fossil glen]

this project main

[static gyro]

now where?

[fossil glen]

just follow it, its like normal folder
C:\Data D\Ai Cover\Content

[static gyro]

i dont see weights

[fossil glen]

but you see assest

[fossil glen]

read carefully, word by word

[static gyro]

thank you

[fossil glen]

np

[tired sequoia]

if it is cuz of pip how do i update it again

[sand iris]

There's a link on the same line

[tired sequoia]

Thanks

[sand iris]

Tell me once you are on the C++ tools installer

There's just one thing to know to have it working

[tired sequoia]

is it the right thing

[sand iris]

Uhh, looks different from the usual

what's under "available" ?

[rare goblet]

Ayo? @tired sequoia level 2 !!!

[proven hill]

luigi why hes downloading c++

out of curiosity

[sand iris]

building python packages locally requires it

[proven hill]

dont they come built in with w10

[sand iris]

Nope

Maybe with win11

Not with 10

[tired sequoia]

im on win11

[sand iris]

Also maybe not on earlier installs of w11

I've only tried this on my ancient win10 install and a brand new (this month) install of w11

[sand iris]

what made you land on this page ?

[tired sequoia]

it takes me to the visual installer because it says "before we take u .. we need to do some things" and then open s visual installer

[sand iris]

Is it the link in the error ?

Oh

[tired sequoia]

i have this already

[sand iris]

huh

[tired sequoia]

i think

[sand iris]

"modify"

[tired sequoia]

top left one?

[sand iris]

oh yeah that's more like it

yeah, tick the c++ one, and screenshot it again

there's more options that are going to get opened on the left

I don't remember their exact name, just gotta see them once more

[sand iris]

great

[tired sequoia]

need as many unchecked as possible cuz barely any c drive space

[sand iris]

you should tick

windows 11 SDK
MSVCv143

the rest is not needed

[tired sequoia]

ok thanks installing

any ideas how i can free this up in meantime 😭

my program files x86 only has 10gigs in it so i dont get it

oh normal program files is big af

[silk mirage]

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[plush compass]

When creating a dataset, do you remove all breathing sounds?

[proper shale]

if they have a bit of noise then u can remove them, but it's better to keep em

else it'll make breathing a bit weird

[tired sequoia]

please wtf is this happening to me

[sand iris]

might be faiss-cpu

or faiss-gpu

unsure

[tired sequoia]

that worked thank you

[coarse steppe]

is it normal for converting to take over 10 minutes

[tired sequoia]

fixed

[rare goblet]

Ayo? @tired sequoia level 3 !!!

[tired sequoia]

how many things am i gonna have to pip install...

just to use applio

[violet heron]

Applio should install it instead right?

[tired sequoia]

well it said succesfully installed, but when i run the applio rvc fork its giving me all this stuffi need to pip install

[tired sequoia]

should i run the install bat even though it says installed successfully?

[violet heron]

No

[tired sequoia]

when i open the go-applio.bat i see this

i press 1 and then have to pip install for ever

now what

man ffs

my ass

[sand iris]

https://cdn.discordapp.com/emojis/790214199544512573.webp?size=48&name=kek&quality=lossless

[tired sequoia]

so u dont know

[sand iris]

they updated a lot of shit recently

And it hasn't been exactly reliable in the past

OH WAIT THAT'S THE APPLIO INSTALLER

https://cdn.discordapp.com/emojis/716191283147702282.webp?size=48&name=kek~9&quality=lossless

The mess strikes again

[tired sequoia]

??

[sand iris]

https://github.com/IAHispano/Applio-RVC-Fork/releases

At this point there are 20 different ways to install it

This is the latest and (supposedly) working one

[tired sequoia]

ok thx

[sand iris]

||for the time being||

[tired sequoia]

how to make sure it installs in D drive since its bat file just open in d drive yh

[tired sequoia]

1 or 2

[sand iris]

do you have an nvidia GPU ?

[tired sequoia]

oh yh 1

[static gyro]

ok it stopped makeing epocs as it was not recieving input how do i start it again?

using RVC 10/25

[tawny lake]

why does the processing of a ai thing im making take so long??????????

[rare goblet]

Ayo? @tawny lake level 1 !!!

[tawny lake]

its at 580seconds like bro

[static gyro]

are you on epoch 1 still?

[proper shale]

if ur using a Huggingface space then yeah

[river gate]

Temporary folder already found. Wiping...
Archive:  /content/drive/MyDrive/rvcDisconnected/loremixerUnit.zip
   creating: /content/temp_dataset/zipfile/loremixerUnit/
  inflating: /content/temp_dataset/zipfile/loremixerUnit/sampTechDifficulties.wav  
Sanitizing...
Dataset Type: Multispeaker
---------------------------------------------------------------------------
FileNotFoundError                         Traceback (most recent call last)
/usr/lib/python3.10/shutil.py in move(src, dst, copy_function)
    815     try:
--> 816         os.rename(src, real_dst)
    817     except OSError:

FileNotFoundError: [Errno 2] No such file or directory: '/content/temp_dataset/zipfile/loremixerUnit' -> '/content/dataset/loremixerUnit'

During handling of the above exception, another exception occurred:

FileNotFoundError                         Traceback (most recent call last)
3 frames
/usr/lib/python3.10/shutil.py in copyfile(src, dst, follow_symlinks)
    252         os.symlink(os.readlink(src), dst)
    253     else:
--> 254         with open(src, 'rb') as fsrc:
    255             try:
    256                 with open(dst, 'wb') as fdst:

FileNotFoundError: [Errno 2] No such file or directory: '/content/temp_dataset/zipfile/loremixerUnit'```

this keeps happening

no matter what i do, it keeps happening

i go through all the steps, everything goes off without a hitch

why does this keep happening

[tawny lake]

what other space am i supposed to use

[proper shale]

nah its the right one probably

its just how it is

uses cpu lol

[static gyro]

the rvc colab i was using stopped makeing epocs as it was not recieving interactive input. how do i start training again?

using RVC 10/25

answer for me: just hit train and it will continue from your drive

[river gate]

DATA CONTEXT:

using rvc_disconnected

[crystal wedge]

File "C:\Files\RVC\Applio-RVC-Fork\infer-web.py", line 3, in <module>
    from tensorboard import program
ModuleNotFoundError: No module named 'tensorboard'

I've already tried reinstalling Applio, I even installed Python 3.9 just to pip install tensorboard, still nothing. I used applio-manager too to update/fix tensorboard, still get the same error. This was never an issue until today.

[tired sequoia]

hello dear. after much time i have managed to complete it and get it working i think.. but is there any tutorial you know like video of how to use it properly,

how i put voice u know

[sand iris]

Been a while since I did inference

[violet heron]

Pth file goes in weights

Index goes in logs

Refresh button then select the model

[tired sequoia]

the model will appear in inferencing voice?

dropdown

also tf does index do i never understood 💀

[tired sequoia]

in here?

[river gate]

[copper pebble]

accents/intonations ecc. of the voice

[river gate]

does anyone even know how to solve Errno 2

i understand every part of the process so far and then it just doesn't work at this specific step

please i beg of you

[tired sequoia]

oh my goosh it works

thanks you timbovill and gdr

[river gate]

does rvc v2 disconnected just not work anymore??? is it deprecated????? is this error just common knolwedge????????

https://tenor.com/view/kuroko-anime-despair-hit-table-gif-21051524

[crystal wedge]

Please someone help.

[rare goblet]

Ayo? @crystal wedge level 1 !!!

[river gate]

https://tenor.com/view/pomni-stare-the-amazing-digital-circus-existential-crisis-going-insane-gif-10544419092608405864

[molten pecan]

Embed failure!

@violet heron

[river gate]

i CAN'T embed-

i PHYSICALLY CANNOT embed

[molten pecan]

Why

[molten pecan]

Why hasn't Lorena uploaded any vid yet

I want to see sum clarinet music

[half cove]

https://tenor.com/view/pomni-stare-the-amazing-digital-circus-existential-crisis-going-insane-gif-10544419092608405864

[hardy stump]

How can we look for models?

[potent cradle]

/search in #🔍│find-models or in the #1175430844685484042 channel

[hardy stump]

Does it require a level to search?

[violet heron]

No

[hardy stump]

thanks

[river gate]

[rare goblet]

Ayo? @river gate level 3 !!!

[hot lagoon]

How to turn someing like a piano into a " voice"

[proper shale]

if u make a model it might sound like one

[ember plaza]

for the “processing mode” thing it won’t let me select gpu? does anyone why is that?

[hot lagoon]

my trying to use PAPERSPACE for training but is not showing it train

[gaunt current]

is there an alternative to AOD or ROD i can use for tensorboard?

[gaunt current]

@low shard sorry for the ping. but do you know about this?^

[low shard]

you wanna train a model with a tensorboard?

[rare goblet]

Ayo? @low shard level 48 !!!

[gaunt current]

yeah, so i can see if it's having any trouble training

[low shard]

are you on local or rvcdisconnected colab?

[gaunt current]

https://docs.aihub.wtf/guide-to-create-a-model/tensorboard-rvc says i need ROD for tensorboard but the link is down

local

[low shard]

oh, im sorry but i dont know about local but i will tell staff about the link problem in a sec

[gaunt current]

ok. ty

[rare goblet]

Ayo? @gaunt current level 2 !!!

[low shard]

yw,will keep you updated if something

[hot lagoon]

Did not the ROD like got dealt cuz dev say they were going to add a new one

[gaunt current]

oh. is there an alternative then? or can i use tensorboard without ROD?

[hot lagoon]

-aod

[azure marsh]

How do I know if my model is overtaining?

All-In-One Guide on how to make a good model

This guide explains how the D and G files works and much more: https://rentry.org/RVC_making-models

Credits: LUSBERT

Automated Overtraining Detection (AOD)

Will be available soon in #1159513888199540817

Credits: grvyscale

[hot lagoon]

its now going to be call Aod but is not out yet I think @gaunt current

[gaunt current]

will keep waiting. in the meantime going to look for an alternative. thx

[hot lagoon]

Np

[jade ginkgo]

Hellooooo

Can I get help

The rvc isnt working at all for me its takes forever to load and then the connection breaks. I need someones help

[violet heron]

Are you running it locally?

[jade ginkgo]

Ilaria RVC isnt working

Ive been trying for over a week

[jade ginkgo]

Im running it on browser on my phone

[violet heron]

Did you duplicate the space

[jade ginkgo]

What does that mean

[copper pebble]

click on the 3 dots above on the right and duplicate it

[jade ginkgo]

What does that do to help

[copper pebble]

makes a private copy of ilaria rvc

faster cuz theres no queue

[jade ginkgo]

I did. It says I gotta make an account for it

Now what?

It takes forever to build

Yall its been 10 minutes I give up

Oh and it got a runtime error and didnt work. Whatever, I give up

I tried to duplicate it again and it didnt allow me bc I already duplicated it uuugghhhghgg good God what is going on. Now I can never make an AI cover, Im so depressed and annoyed rn bc this dumb thing refuses to work

[rare goblet]

Ayo? @jade ginkgo level 3 !!!

[violet heron]

Make it public

Or

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

Try this

See if this works

[jade ginkgo]

Okay, I rlly appreciate ur help alot

[jade ginkgo]

Its not working either

It converted for 80 seconds and then broke

And it broke again

Oh and broke again

Man this is ridiculous

[violet heron]

Don’t exit the tab

You have to keep it on there

Or close the screen

Mainly a flaw with iOS

[jade ginkgo]

Ok I kept it on for 200 secs then I gave up

I do not have enough patience for this Im sorry

But ts is making me wanna jump off something and Im getting way too angry and emotional to do this

I give up Im gonna go cry or something

[tired sequoia]

How do i know what settings to use here on Colab

for the index file im tryna get index file for my pth

[violet heron]

Gotta wait more then that

If you have a different device just leave it on on that

Or better enough ||do it on your computer||

[jade ginkgo]

I wish I had a computer

[jade ginkgo]

I think you can just leave the settings as is default

[violet heron]

Use something else

Like wordband

[jade ginkgo]

Whats that

[violet heron]

It’s a discord bot

[jade ginkgo]

It makes ai covers for me?

[violet heron]

Yes

[plucky locust]

Welcome to the wordband Discord bot, timbovill! You can use this bot to make covers of songs using character voices. Here's how to use it:

/list List available voice models
/cover Make a cover using a voice model and a YouTube link
/register Register a new voice model from a Huggingface link
/login Connect your Discord account to a wordband account. This will let you generate more covers, and unlock new features!

/help Show this message

Use /help command:<command> to get more information about a specific command and what parameters it uses.

The wordband Discord bot works on a queueing system. There will always be a free queue, but you can buy a paid plan to increase priority and access more features.

You can join the wordband Discord server to follow new updates here: https://discord.gg/nGnmj769TJ

[jade ginkgo]

LE GASP

[jade ginkgo]

Wait the voice model I want isnt listed in the server but I have the link for it. Can i send someone the link so they can put it in the server

[violet heron]

You do /register new if it’s a HuggingFace link

[jade ginkgo]

Can u give me an example of how to use it?

[violet heron]

(If it the HuggingFace link take out the “?download=true” part.)

[violet heron]

Covers or registering a model

[jade ginkgo]

Using wordband bot

To make a cover

[plucky locust]

zipper is not a registered model! Use /list to list all models, or use /register to register a new model.

[violet heron]

Oops

[plucky locust]

You're in the queue! Your cover will be ready soon. Number of people in front of you: 0 (Status: 🟢)

[violet heron]

@jade ginkgo

[jade ginkgo]

Yes?

[violet heron]

Example above

[jade ginkgo]

Where

[violet heron]

Here

[jade ginkgo]

Thats a bot message i still dont know what message to type out for my cover

[plucky locust]

Welcome to the wordband Discord bot, timbovill! You can use this bot to make covers of songs using character voices. Here's how to use it:

/list List available voice models
/cover Make a cover using a voice model and a YouTube link
/register Register a new voice model from a Huggingface link
/login Connect your Discord account to a wordband account. This will let you generate more covers, and unlock new features!

/help Show this message

Use /help command:<command> to get more information about a specific command and what parameters it uses.

The wordband Discord bot works on a queueing system. There will always be a free queue, but you can buy a paid plan to increase priority and access more features.

You can join the wordband Discord server to follow new updates here: https://discord.gg/nGnmj769TJ

[violet heron]

Read

[jade ginkgo]

I’ll try…

Do I register a voice model not on this server first

[violet heron]

Check if the voice already exists

Using /list

[jade ginkgo]

Its not

[violet heron]

Then register it

[plucky locust]

/register Register a new voice model from a Huggingface link

You can use this command to register a new voice model from a Huggingface link. Most models on AI Hub should work immediately, but only RVC2 models are supported.

Parameters:

• url: The Huggingface url for the model. This should be a public repository that anyone can access. The model should be a RVC2 model, and should only contain a .pth and .index file (or a .zip file containing both of these). ANYTHING else will probably not work!
• id: The id for the model. This should be a short, unique identifier for the model, without spaces. For example, the id for Waluigi is waluigi. This will be used to make covers with the model.
• name: The name for the model. This should be a short, human-readable name for the model. For example, the name for waluigi is Waluigi.

There are also plenty of other optional parameters that you can use to customize the model. These are the same as the ones in the colab.

Use the command to register a new model like this:
/register url:https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/rmvpe.pt

[jade ginkgo]

/register

Url: https://huggingface.co/BoundlessFate/MyModels/resolve/main/lindemannAI_e130_s33020.zip?download=true

Id: Maggielindemann

Name: Maggie Lindemann

[rare goblet]

Ayo? @jade ginkgo level 4 !!!

[violet heron]

Command fail

[jade ginkgo]

Why

Why’d it fail

[violet heron]

You gotta click here

[violet heron]

Also don’t include the “?download=true”

It won’t work if you include it

[plucky locust]

Successfully registered model maggielindemann (maggie lindemann)!

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

[jade ginkgo]

Oh i didnt see this before I did the thing-

Um I already made it, do I remake it?

[violet heron]

It’s fine

It worked

[jade ginkgo]

Yay

[violet heron]

They must of fixed it then

[molten pecan]

It says runtime error when your internet disconnects or when you tap out from the page.
And if you want to restart or use the duplicate space again, you don't have to duplicate it from scratch. Your duplicated spaces are available on your HF profile

[jade ginkgo]

I dont see it in the models channel, did it actually work?

[harsh scarab]

You're placed in Cover Generator Queue. Current position: 2

[jade ginkgo]

FRICK YEAH LETS GO BABY!!!!

[jade ginkgo]

Gives you super tight hug and smiles very joyfully

Also I didnt know there were other maggie lindemann models. I thought only the ones in the models channel were availale lol

[violet heron]

Wrong bot

It still works tho

Wordband offers more options and has a smaller queue

[storm ermine]

does anyone know how you separate individual singers’ vocals from songs to make duets?

[molten pecan]

Try asking on #🔍│help-w-okada

Happy holidays too!!

[molten pecan]

I use UVR5

There's a HF space for it if your PC can't run it

[jade ginkgo]

Ooohh thats really difficult

[molten pecan]

Once you have the clean vocals, use the HP6 algorithm. If it doesn't sound good use HP5

[jade ginkgo]

Even mashup creators have had a hard time figuring that out

[storm ermine]

I use that too but I only know how to separate all of the vocals at once

[molten pecan]

Read above

[storm ermine]

oh so there’s a different algorithm specifically for it?

[static gyro]

how do you get your models for applio SEO?

[molten pecan]

Yup

[violet heron]

Just do it yourself in any basic video/audio editor

Convert both then split what you need

[steel herald]

can someone help me use rvc

[verbal tulip]

hi. if i want to train a voice model with only 10 minutes of dataset, how many epochs should i set?

[strong belfry]

set the limit to 400-500. Models average somewhere between 200-300 epoch before overtraining. You cant determine where exactly is it best until after

[verbal tulip]

Thank you. I'm training a model using colab, how do i know i am overtraining it? Is there some kind of message in the log?

[strong belfry]

I train locally, using tensorflow to look at the trainign graphs

[vestal glacier]

Pipeline is not initialized.\

[rare goblet]

Ayo? @vestal glacier level 1 !!!

[vestal glacier]

how to fix

help pls

[magic folio]

can someone help??

i made a model, but the rvc space im using just show the model when i download it

and when i try and use the models already there, the file doesnt convert at all

[worldly light]

hi my friend needs help with ultimate vocal remover, he has a good enough computer and it should have enough for cpu/gpu to handle, but it crashes when he tries using it. is there anything needed like python or something?

[spice junco]

Hey can anyone here explain how to do ai related covers in general? I currently have the problem of not knowing how to convert ai voice models into a non distorted quality cover

[rare goblet]

Ayo? @spice junco level 1 !!!

[proper plaza]

What are the requirements to be able to use RVC? I want to use it but I don't know if my laptop can handle it.

[glossy grove]

8 GB of vram and a modern GPU. I'm able to run in on my GTX 1070

less if you dont plan to train

[violet heron]

What GPU?

[proper plaza]

I can be honest... An intel celeron 847 with 4gb of ram

[rare goblet]

Ayo? @proper plaza level 1 !!!

[violet heron]

https://tenor.com/view/crying-emoji-gif-21922016

Yep your going to need to online methods for RVC

[spice junco]

When you say this you mean via an online website ?

Oh my bad that was for someone else

[violet heron]

Yes

There’s some online methods to use RVC but are generally slower in most cases

And often have limits

[tall carbon]

Is it common loss/g/total continue to rise after starting the training or have i stuffed it up??

[tame mica]

yeah.... nope

nuh uh

[bronze star]

Hi i need help to upload My Model

[rare goblet]

Ayo? @bronze star level 2 !!!

[azure grail]

res is like 20,000 plus. help

[sacred ice]

Hi, so everything was working great in my training (using Applio), but... clicking "Train Feature Index" doesn't spit out an output! Everything else works fine, just can't get a .index file...

[glad zealot]

It should say in the console that it's added

You can check the logs folder and try to search for *.index

[sacred ice]

Already checked there unfortunately

[glad zealot]

You need to be not currently training or training already finished for it to work tho, since you can only do 1 thing at time

[sacred ice]

Command prompt just spits out
2023-12-04 02:17:03 | INFO | httpx | HTTP Request: POST http://localhost:7897/api/predict "HTTP/1.1 200 OK"
2023-12-04 02:17:03 | INFO | httpx | HTTP Request: POST http://localhost:7897/reset "HTTP/1.1 200 OK"

Each time I click the button

[glad zealot]

Are you don't training?

Done*

[sacred ice]

I've had this issue regardless of whether I trained or not

[rare goblet]

Ayo? @sacred ice level 1 !!!

[glad zealot]

Hmm not sure then

[sacred ice]

But currently, no, haven't trained

[glad zealot]

I barely use applio ngl so not sure if it's bug or something

[sacred ice]

I processed and extracted

[sacred ice]

No worries, thanks for your help anyways!

[reef root]

hi

[obtuse stream]

Starting Applio-RVC-Fork with Conda default compatible with all platforms...
Traceback (most recent call last):
File "C:\Users\keres\Desktop\Applio-RVC-Fork\Applio-RVC-Fork\infer-web.py", line 3, in <module>
from tensorboard import program
ModuleNotFoundError: No module named 'tensorboard'
Press any key to continue . . .

whats this problem?

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
tensorflow-intel 2.14.0 requires tensorboard<2.15,>=2.14, but you have tensorboard 2.15.1 which is incompatible.

[brittle wing]

Hi

[rare goblet]

Vote Booster: Vote now for a 10% boost. https://arcane.bot/vote

[scarlet fulcrum]

how do i get out the demo version?

[obtuse stream]

how can i install tensorboard 2.15?

[sand iris]

pip install tensorboard==2.14

...but unfortunately applio's requirements are like all over the place

[high summit]

I need to train a model for music purposes and I have a MacBook Air any programs that could help? That are easy to navigate please dm me! I’m currently using huggingface for my voice changing projects but would like to make my own unique models. Thank you all hopefully someone can help!

[scarlet fulcrum]

or get access to add more voices?

[silver anvil]

the demo version of what?

[silver anvil]

just click edit and upload more voices

[scarlet fulcrum]

ik but i want the not demo version

[rare goblet]

Ayo? @scarlet fulcrum level 2 !!!

[silver anvil]

@karmic oxide

idk how to explain this

my brain is melting

[scarlet fulcrum]

how do i get more?

[silver anvil]

How to find voice models:

#1175430844685484042

scroll through

#🔍│find-models

use /search

https://weights.gg/

holds onto backed up voice models that might have been deleted from the discord channel

@scarlet fulcrum if u need more help go to #🔍│help-w-okada

[drowsy trout]

Does RVC work with 8GB of VRAM?
The price difference between RTX 4060Ti 8GB and RTX 3060 12GB is about $60 so I'm not sure which one to choose

[sand iris]

yes it does for both making models and converting voice

[drowsy trout]

Thank you ❤️

[obtuse stream]

anyone have idea?

[glad zealot]

_ _-egirl

[silver anvil]

huh

@glad zealot idk what happened

[glad zealot]

Voice model OwO

[slim citrus]

@silver anvil tell me which guide sorry i am unable to find

[rare goblet]

Ayo? @slim citrus level 2 !!!

[silver anvil]

idk dude just check https://docs.aihub.wtf/

or https://docs.aihub.wtf/guide-to-create-a-model/dataset-creation

[slim citrus]

I already did this now please tell me how i can convert rvc model zip file

[silver anvil]

i cant help u

you need to ask someone else

[slim citrus]

Sorry for the inconvenience I don't know anyone here I am a newbie and can you please tell someone to help me

[silver anvil]

kinda looks like a lot of ppl are not on rn

[slim citrus]

[silver anvil]

just wait and see if someone responds to you later @slim citrus

[slim citrus]

Ok ok

[lethal sand]

how to use appilio for training audio datasets?

[rare goblet]

Ayo? @lethal sand level 2 !!!

[uneven junco]

Can anyone explain what the .index file is actually used for ?
I understand the role of the .pth file but I can't make-up my mind for the .index file especially given its size and the fact that it is not mandatory ...

[quick jungle]

https://cdn.discordapp.com/attachments/1160576561636966491/1165372018414342284/warning.gif

[uneven junco]

What is that ahah ?

[proper shale]

index is basically all the mannerisms of a person (accent, how they say certain phonemes, etc) into one file that can modify outputs.

gotta fix my pinned copy paste stuff haha

[unique bluff]

hello everyone here, i got 100% CPU recently when i use rvc with the same settings as usual days before, which my pc also have bsod. Is there anything going wrong with the vc though. If there's a pinned message about that problem then i will try (edit: the moment when i used ServerIO analyzer, it gets 100% cpu quickly)

[rare goblet]

Ayo? @unique bluff level 1 !!!

[uneven junco]

Obrigado, senhor!

[proper shale]

De nada ;)

[proper shale]

Really depends on what CPU it is

[unique bluff]

well I am using Intel(R) Core(TM) i7-10700 CPU @ 2.90GHz + GPU NVIDIA GeForce GTX 1660 SUPER (Im using GPU to do the rvc)

[proper shale]

hmmm what version are you using of W-Okada?

[unique bluff]

hmm...i dont know the version but according to the file date (29/07/2023), is it fine? and where do i check the version of the rvc though? (edit: nvm, its v1.5.3.11)

[proper shale]

onnx-directML-cuda?

[unique bluff]

onnxgpu-cuda

[proper shale]

alright that's the right version for your gpu, at least

I'd recommend upgrading to v1.5.3.15

[unique bluff]

lastest ver from the main website right

[proper shale]

https://huggingface.co/RavenCutie21/OkadaVersion15Backups/resolve/main/MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.15.zip

this one

[unique bluff]

mk ty let me try

[proper shale]

ighty, hopefully it fixes your issue

[uneven junco]

I actually got one small question ! Does that mean that index file is kind of independant from pth file ?
Like if I overfitted at 400 epochs and want to use, say, save from epoch 150 I can use the same index file ?

[proper shale]

Yep

It's independent

...you could also mix indexes from other models to your model, too

[unique bluff]

wait i want to ask, if i replace old version with new version. Does the setting auto-save or have to do all thing manually again

[uneven junco]

That's so cool

[rare goblet]

Ayo? @uneven junco level 2 !!!

[proper shale]

You can bring your models to the new version, I guess

Maybe the settings inside them too

Just copy over the model_dir folder to the new install

[wary bluff]

i'm using the Easy colab for training and i'm finished now but i can't find the index file under logs... do i retrain and try again or what?

[proper shale]

there is a training index step on that colab so uh

something must've happened then

what did feature extraction have as its log?

[wary bluff]

i just left it as rmvpe_gpu

[rare goblet]

Ayo? @wary bluff level 5 !!!

[wary bluff]

i dunno if this helps but i've checked in project-main, still can't find anything

[unique bluff]

well, it does. Not a single file moving needed

[rare goblet]

Ayo? @unique bluff level 2 !!!

[wary bluff]

okay i think it was actually the preprocessing data step

[latent kettle]

How can I make AI covers on colab ?

[wary bluff]

lemme try uploading the dataset in individual audio files

[wary bluff]

check #1159949278270193734 for the huggingface spaces and open them

[latent kettle]

I don't want to use huggingface spaces

[wary bluff]

use applio or colab then

[proper shale]

Seems like something screwed up

[wary bluff]

or even one of the other apps / websites

[latent kettle]

Can you send me a link?

[wary bluff]

i think it was the dataset not being preprocessed properly

[proper shale]

Actually, I believe it could be a codec mismatch with your original dataset audio

[unique bluff]

Yet though updating makes the ui a bit better. Still not decrease the CPU usage tho, i have to use passthru not to give my pc a bsod as temporary solution

[wary bluff]

what's the problem with it?

[wary bluff]

check here https://docs.aihub.wtf/

[proper shale]

Where'd you get it from and did you convert it somewhere?

Other than Audacity

[wary bluff]

ah i did

[proper shale]

Also, did you use RAR for compressing the file

[wary bluff]

zip

[proper shale]

I've seen people complaining that CloudConvert gives ppl issues with datasets

[wary bluff]

got the audios as mp3, used UVR to isolate vocals (as WAV) and then used some app to split it (still as WAV)

[proper shale]

split it?

look, you don't really need to split them unless they're like 50 mins long or something

[wary bluff]

so i don't need to put them into multiple short files?

[proper shale]

nope

u can just use all the audio files and export them via audacity

as one

[wary bluff]

what about silences?

[proper shale]

preprocessing literally does that for you btw

[proper shale]

trim them

all of them

[wary bluff]

oh my days, no ways

[proper shale]

they're bad for the model's learning

[wary bluff]

so you're saying i should have one file with all the isolated vocals as one long audio?

[proper shale]

yeah

no need to split them in short 10 sec files

this was back in the sovits days

[wary bluff]

haha yeah

i've been out of making models for a while but i'll try this now

thanks dude

[proper shale]

np ;)

any more questions?

[slate pawn]

I do

[wary bluff]

not right now but can i @ you if i do?

[slate pawn]

Index isn't working here at all

I trained that model by the way not downloaded

[proper shale]

ighty

[proper shale]

huh

i mean you don't really need index for realtime but uh, that's weird

@sand iris know anything about this?

[sand iris]

what's the model file ?

it's name

[slate pawn]

I know it's not a need but I heard it helps with the accent or ways of pronunciation but it doesn't work on this certain model despite doing an index training already

[slate pawn]

HV_rmvpe

[sand iris]

unsure, the model might be fucked somehow

can't check the error atm, in the middle of classes

[slate pawn]

try moving to the right of the class then

I'm joking please don't ban me

[sand iris]

[brittle wing]

@velvet swan ur itto model doesnt work for me do yk how i can maybe fix that?-

[rare goblet]

Ayo? @brittle wing level 2 !!!

[fossil glen]

my opinion, you need slice audio into multiple short files, if you do this you will get best quality of pronounce, you will know if you do it

[fossil glen]

delete index file you just need pth file for inference

[brittle wing]

can anyone help me , i have rx 6600 8 gb vram gpu with i5-12500f processor , my voice is too laggy when i use custom voice models , but the pre installed ones work fine

[brittle wing]

bet tysmm !!

[fossil glen]

np

[honest tiger]

Is there any way in which I can get access to the google colab links ? 😀

[viscid moss]

#1159513888199540817 #1159949278270193734

[honest tiger]

Thank you very much!

[viscid moss]

Bro, ur pfp. Pls change it

[viscid moss]

I need to convert pth file into onnx file.
Check this guide:

https://rentry.co/VoiceChangerGuide

[brittle wing]

is it alr for you if i can dm you?

[wary bluff]

that's what i've been doing

[viscid moss]

Let's talk in #🔍│help-w-okada

[wary bluff]

i think it working but my colab just stopped working now. How do i resume?

[fossil glen]

gpu limit usage? copy project main on gdrive then paste folder project main other account then run it

[wary bluff]

okay cool, what do i put for epochs then? The number remaining before it stopped training or the total number i want to train to?

[fossil glen]

for resume training must be same

[wary bluff]

aight thanks

[fossil glen]

i recomend do screnshoot layer

[wary bluff]

screenshot?

[honest tiger]

I did

[rare goblet]

Ayo? @honest tiger level 1 !!!

[fossil glen]

some times you need second brain, if you forgot a settings it's so helpfull

[wary bluff]

are you talking about screenshot my settings?

[fossil glen]

yes

[wary bluff]

oh yeah

[fossil glen]

my keyboard wont type correctly its need change new keyboard

[wary bluff]

haha

my Colab ain't working right now

i'll try later

[wary bluff]

do you think i can resume training like a few hours later / tomorrow

[viscid moss]

ty

[wispy lark]

Morning all, i just had a buddy on mine install this program and its super awesome
However i have ran into a problem, I run a 2 PC setup and he put it on my gaming PC.
I use the GoXLR and HAVE TO HAVE certain things ticked in my audio settings in order to hear the game.
IFor me to use the voices i have to have it on the virtual cable, well then i cant hear the game,
Does anyone else run a 2 pc set up with the GoXLR and is able to use the move in the game chat?
IF so would you PLS help me out?
THank you all in advance and hope you have an amazing week

[gaunt current]

is AOD/ROD needed for tensorboard? or can tensorboard be used without it/an alternative?

[verbal basin]

hey is there a way i can train a model using my amd gpu?

[high summit]

Hey I need some help with some pretty specific requirements please reach out if u can help, thank you! I’m specifically looking to train a model I have a MacBook Air so that might limit me as well. It’ll also be using specifically for music.

[sand iris]

running linux

Or use online options so you don't have to use your hardware

[young warren]

where do I install the voice models?

[verbal basin]

Really?

How can I run linnux

[sand iris]

Probably not worth it when online options exist

[verbal basin]

I tried colab but the runtime disconnects since gradio is disallowed.

[rare goblet]

Ayo? @verbal basin level 1 !!!

[sand iris]

you'd have to get some spare drive space, get a usb, boot on the usb, install it, and finally do a slightly manual setup for RVC

There's non gradio options

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[sand iris]

RVC disconnected and easyGUI

[verbal basin]

Interesting

[sand iris]

They are both for training, and don't use gradio at all

[verbal basin]

Oh wow

[sand iris]

So they should be allowed (they are for the time being)

[verbal basin]

Alright, thanks so much!

[molten pecan]

RVC

[young warren]

well yeah but where do I put the files

[molten pecan]

.pth file in the logs folder and .index file in the weights folder

Though it might be the other way around

Then click Refresh in the app

[young warren]

and where are the log and weights folder

[fossil glen]

yes its link with google drive

[gaunt current]

ok thank you

[high summit]

Hey I need some help with some pretty specific requirements please reach out if u can help, thank you! I’m specifically looking to train a model I have a MacBook Air so that might limit me as well. It’ll also be using specifically for music.

[fossil glen]

use google colab

[fossil glen]

my pc only have intel HD

[fossil glen]

-colab

[azure marsh]

Suggestion for @high summit

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[young warren]

how do I get Steam to detect the audio? I set it to the voice.ai mic but no sound comes out

[rare goblet]

Ayo? @young warren level 1 !!!

[fossil glen]

#🔍│help-w-okada

[gritty plume]

Is there a Guide on How to record oneself proper like what tones to put in and stuff to have a realistic AI for songs so it doesnt sound trash when it gets pitched

[fossil glen]

make good model

[gritty plume]

yea but what should i say if i record myself

[fossil glen]

you can use any vocal, after then you slice audio into multy file for get best "pronounce"

[gritty plume]

what you mean by sliceing audio into multy file

[rare goblet]

Ayo? @gritty plume level 1 !!!

[gritty plume]

like letting my model addept to the diffrent parts of a song for example if the pitch sounds awfull?

[fossil glen]

result slice

[fossil glen]

whats a genre?

if like metal scream its may pitching or broke

if like high note with pop song its not pitching

[gritty plume]

regular pop or christmas songs

[fossil glen]

its not slice audio tricks will safe a model

[gritty plume]

the autotune it did there is just 100% not realistic if u know what i mean

[fossil glen]

you should re train with audio slice tricks

[gritty plume]

so u mean for training a model i should use the same recording sliced up in multiple files in the folder?

[fossil glen]

yes same recording, btw what you length dataset?

[gritty plume]

it was like a 6min 30 recording of just talking

like a voicemessage

[fossil glen]

its that clean? no reverb music etc

[gritty plume]

yea the recording sounds clean like no backnoice or anything just talking into an mic

[fossil glen]

yes then just slice

use this
https://github.com/ardha27/AI-Song-Cover-RVC

[gritty plume]

i will check that out ty

[vestal glacier]

pepeline is not initialezed

how to fix? hlp pls

@tranquil cliff can you help?

[copper pebble]

IMPORTANT - - >#📰│dev-updates

[wide parcel]

RVC libraries are still loading. Please try again in a few seconds.

[sonic agate]

100 epoch is good for a 17 min dataset model? clean vocals

[brittle wing]

so that didnt work </3 it kept saying it wasnt a zip file

https://colab.research.google.com/github/hinabl/AICoverGen-Colab/blob/main/Hina_Mod_AICoverGen_colab.ipynb

i use this

[sonic agate]

what do you want to cover?

[brittle wing]

i'm tryna use itto's voice so i can make him sing consideration by rihanna

[sonic agate]

send me the model

[brittle wing]

https://huggingface.co/magnus-son/AratakiIttoRVCv2_300e/blob/main/AratakiIttoRVCv2_300e.zip

[fossil glen]

you need put in zip then upload to hugging face and download model from gradio

[brittle wing]

ohhh

[fossil glen]

wrong link

[brittle wing]

which one do i send??

[proven hill]

IMPORTANT, NEW ILARIA RVC -> #📰│dev-updates message

[sonic agate]

it works

[fossil glen]

copy link address

[brittle wing]

OH-..

oml tysmm

[brittle wing]

ima try, i'll lyk if it works <3 tyy

[fossil glen]

[tranquil cliff]

Hei

sorry

what's up?

[opaque knoll]

Can't get the pitch slider to work on weights.gg. :/

[molten pecan]

Pitch slider in weights.gg?

[twin pewter]

yo i try to download but it fails

[twin pewter]

i did this but for another model

[molten pecan]

Download what?

[twin pewter]

ian (romanian trapper) model

it fails

[molten pecan]

Where are you downloading it from?

[twin pewter]

huggingface

[rare goblet]

Ayo? @twin pewter level 1 !!!

[molten pecan]

What error are you getting?

Can you send a screenshot?

[twin pewter]

2 seconds

[molten pecan]

Try using a different name

[molten pecan]

Go to that model link

Then to the Files tab

[twin pewter]

it auto downloads

[molten pecan]

Oh

Try duplicating the space

[zealous falcon]

Wait so did I get an error or not??

[tame mica]

no

it's fine

[zealous falcon]

Gotcha, thanks :o

[molten pecan]

Click the rar file

[tame mica]

wait

[molten pecan]

And where it says download, right click it and copy the address

[tame mica]

isnt it supposed to be .zip

[molten pecan]

And paste that URL

[molten pecan]

Idk but it works for me both ways

[molten pecan]

Duplicate the space

Click the three dots on the upper right

Duplicate this Space

Make sure you it's set to Private

And wait for it to duplicate

Once it's done paste the link again

[tame mica]

try reuploading it as a .zip too

[molten pecan]

OH SHIT YOU ARE RIGHT

I forgot there are two different ways to compress it 😭

What do they even differ on

[tame mica]

idunno atleast something 💀

[grizzled badger]

Guys I might be rarted or something but where can I find the guide on how to create my own voices?

[opaque knoll]

The slider that allows one to adjust voice pitch of a voice model — from -24 to 24.

[verbal tulip]

guys, i want to cover an audio with mangio rvc, but that audio is some kind of opera voice with a lot of pitch shift, what is the best model and setting for this kind of audio? crepe? harvest?

[rare goblet]

Ayo? @verbal tulip level 1 !!!

[zealous falcon]

bruh 😔

There a way around this?

[stoic stump]

Make a bunch of google accounts

[molten pecan]

mangio-xrepe

Crepe*

The lower the value of the hop-length the more dramatic the pitch shift will be

[verbal tulip]

the audio is this, does lowering hop-length help covering this properly?

[molten pecan]

Oh yeah

I would set it between 1 to 50

See which value sounds best

[verbal tulip]

thank you bro. you helped a lot

[molten pecan]

You are welcome. Keep in mind that someone talks at the end of the audio

[verbal tulip]

yeah i'll just cut that

[molten pecan]

Alright

Have fun!

[verbal tulip]

ooh btw sorry i called you a bro, i didn't saw the pronouns in the bio

[molten pecan]

No worries

Protons in bio

[sonic agate]

is ngrok better than colab?

[proper shale]

use a colab in -colab

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[zealous falcon]

I don't think i understood what you said lol

Wait- nevermind

[proper shale]

thought the command had worked pol

lol

[azure grail]

help

[rare goblet]

Ayo? @azure grail level 1 !!!

[sonic agate]

?

[azure grail]

res is over 20000 and buf is 300. major crackling and i can't hear myself

[molten pecan]

Ask on #🔍│help-w-okada

[spare robin]

does RVC Appilo just not like C's or K's or am I doing something wrong?

Literally anything I record with an enthused C or K the AI never knows what to do with it

[lean owl]

How much epochs would something like this be? (Also the sound is loud so turn ur volume down)

[spare robin]

The top one is ElevenLabs, and the second one is when ran through RVC.

Don't question the method I'm just curious why the C's or K's are getting omitted.

[violet heron]

Might be the model

[spare robin]

Maybe. It isn't doing it with the models I created, but someone made a batch of Left 4 Dead cast models and they are all doing this with any pitch extraction method used.

[lean owl]

Need help with this

[molten pecan]

Ask on #📑│making-models

[lean owl]

Fuck

But i cant post stuff on there

Like images

[molten pecan]

[molten pecan]

Ah crap

[molten pecan]

Unload it to Google drive and send the link

Make sure the link is public

[lean owl]

Ok

[lean owl]

Wth lol

Ok i asked there

[wraith mesa]

-train

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[hot lagoon]

Are we allow to send youtube links for demos for a modle?

[wraith mesa]

yes

[molten pecan]

-huggingface

[azure marsh]

Suggestions for @hot lagoon

huggingface.co upload Guides

English guide

Link: https://rentry.org/fdg_guide_new
Credits: FDG

Spanish Guide

Link: https://rentry.org/AI_Hub_guide_esp
Credits: Julia (ailen2091)

Hindi guide

Link: https://rentry.org/AI_hub_guide_hi
Credits: Enes

Turkish guide

Link: https://rentry.org/AI_hub_guide_tr
Credits: Enes

[molten pecan]

It says it at the end of the guide

[vapid gust]

@proven hill

what was i supposed to do with this graph

[silver anvil]

@vapid gust ???

[vapid gust]

what

like what do i do with it

[silver anvil]

y u tagin admins for shit?

[vapid gust]

calm down

[silver anvil]

im calm?

[vapid gust]

so why u pressed

all i did was tag to ask

why u in here

[silver anvil]

tag a helper

[vapid gust]

no

[silver anvil]

[hot lagoon]

[pearl swift]

-train

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[jaunty iris]

so like
its on them

because ive been trying everything for the last couple hours

still cant get past this no module tensorboard issue

this thing is driving me crazy

[rare goblet]

Ayo? @jaunty iris level 2 !!!

[honest pond]

omg you scare me

[lean owl]

thats why i told you to turn down ur volume

[fiery bridge]

I was using RVC V2, but I keep getting connection errored out.

[chilly tiger]

机器人

[molten pecan]

What

[molten pecan]

Are you using Colab?

[jaunty iris]

Figured it out Went back to applio github, installed go applio bat 'outdated' version and now i can launch

Finally lol

[latent kettle]

Can someone explain how i can analyse the tensorboard ?

[mighty laurel]

is there any model for creating a chorus in rvc ? plz send if any

[viscid moss]

Check this guide:
https://rentry.org/rvc_making-models

[viscid moss]

hmmm search por a specific model in #1175430844685484042 and #🔍│find-models

[spare saffron]

I'm trying to train a model without pitch guidance and now it's giving me this error.
`INFO:cmbsoldier:Loaded pretrained assets/pretrained_v2/f0G40k.pth
Process Process-1:
Traceback (most recent call last):
File "C:\Users\strin\Desktop\Applio-RVC-Fork\lib\infer\modules\train\train.py", line 280, in run
utils.latest_checkpoint_path(hps.model_dir, "D_*.pth"), net_d, optim_d
File "C:\Users\strin\Desktop\Applio-RVC-Fork\lib\infer\infer_libs\train\utils.py", line 212, in latest_checkpoint_path
x = f_list[-1]
IndexError: list index out of range

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "multiprocessing\process.py", line 315, in _bootstrap
File "multiprocessing\process.py", line 108, in run
File "C:\Users\strin\Desktop\Applio-RVC-Fork\lib\infer\modules\train\train.py", line 318, in run
net_g.module.load_state_dict(
File "C:\Users\strin\Desktop\Applio-RVC-Fork\runtime\lib\site-packages\torch\nn\modules\module.py", line 2041, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for SynthesizerTrnMs768NSFsid_nono:
Unexpected key(s) in state_dict: "enc_p.emb_pitch.weight", "dec.m_source.l_linear.weight", "dec.m_source.l_linear.bias", "dec.noise_convs.0.weight", "dec.noise_convs.0.bias", "dec.noise_convs.1.weight", "dec.noise_convs.1.bias", "dec.noise_convs.2.weight", "dec.noise_convs.2.bias", "dec.noise_convs.3.weight", "dec.noise_convs.3.bias".`

How do I fix this?

[zealous falcon]

-colab

[azure marsh]

All working colabs / spaces ☁️

• AICoverGen-NoWebUI, by Ardha Google Colab
• RVC Disconnected, Kit Lemonfoot Google Colab
• easyGUI, by rejects Google Colab
• Ilaria RVC, by thestingerx Hugginface Spaces
• 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

Credits to Faze Masta and Antasma for compiling these links.

[gusty kestrel]

I have found 14 results for the search Peter...

Peter Griffin

Owner: skippy_lean
Uploaded: N/A

Epochs

100

Technology

RVC

Algorithm

Mangio-Crepe

Requested by rd8 · Dec 5, 2023, 7:14 AM

I have found 9 results for the search Peter Griffin...

Peter Griffin

Owner: skippy_lean
Uploaded: N/A

Epochs

100

Technology

RVC

Algorithm

Mangio-Crepe

Requested by rd8 · Dec 5, 2023, 7:14 AM

[tranquil cliff]

use https://discord.com/channels/1159260121998827560/1159290666019471421

[boreal mango]

ValueError: mutable default <class 'fairseq.dataclass.configs.CommonConfig'> for field common is not allowed: use default_factory

im using applio rvc fork bat file

anyone know how to help?

[spare saffron]

I've tried to install Applio 4 times now and everytime it fails.

[flat harness]

why cant i hear the voice changer??

[regal hill]

can i get access to #✦│chat @red kayak btw

is there a specific requirement

[red kayak]

Keep chatting

[regal hill]

oh ok

[red kayak]

Show me ur tensor board

[regal hill]

looks fine right

[red kayak]

Click the thingy box

[regal hill]

my models usually never just does straight lines ahaha

[red kayak]

Below the graph

[regal hill]

these?

[red kayak]

3rd one

[red kayak]

Keep going

[regal hill]

keep training?

[spare saffron]

No matter what I do whenever I launch go-applio.bat it says file not found and the only options are conda.

[robust quartz]

how to use mp3 vocals only on yt to RVC?

[proper shale]

just use a Huggingface Space and upload that mp3

[robust quartz]

how pls?