✨│ai-help

[pastel oak]

On the top it says "nvidia cuda", you have the wrong version if you have an AMD rx vega

[undone kindle]

Ahh, which version do i use?

[flint vapor]

Yes right

[flint vapor]

DirectML

version

for amd

[pastel oak]

https://github.com/deiteris/voice-changer/releases/download/b2196/voice-changer-windows-amd64-dml.zip

@undone kindle

[wispy lodge]

But you had DirectML version downloaded and used it before reboot. Did you also download cuda version?

[undone kindle]

im sorry, im not tech savy but this is what ive downloaded so far

[wispy lodge]

*A bit of context, here you are using dml version

[pastel oak]

top right one that says "-dml" at the end is the correct one

you can delete the one next to it with "-cuda"

[wispy lodge]

Yeah, you have two versions 😄

[undone kindle]

Thankyou lol, its back to what i had

Now to solve the issue why i restarted in the first place lol, i cant seem to connect it to discord

[undone kindle]

Yay, i solved that issue. Now I just need to solve the delay issue. I feel like no matter how i adjust the chunk then it does not make a difference

[wispy lodge]

Lower chunk should lower delay. But make sure that chunk is more than "perf" value

[undone kindle]

Perf?

[rare goblet]

Ayo? @undone kindle level 4 !!!

[undone kindle]

my perf is constantly at 0

[wispy lodge]

Well, it should change when you speak

[undone kindle]

Ah i see, they just jump all over the place

maybe my computer is just trash for this

[wispy lodge]

Wdym by jump? Like how is it changing?

[quiet ginkgo]

Everytime I try to download this file It keeps showing It's downloading but when It's done nothing Is happening

[undone kindle]

like the res jumps up by 500s as i talk

[proper shale]

ru on RVC Disconnected? it should be in your drive if you've exported it

[wispy lodge]

Well, 500ms is fine. You can change Chunk to value that is a bit above it. Like 552ms

[undone kindle]

No, like it goes up, 500-1000-1500-2000 to like 4000

[quiet ginkgo]

There's the Index file only

[proper shale]

yeah run the export model step. stop training if necessary

[wispy lodge]

But that's what "res" would show. What does "perf" show?

[quiet ginkgo]

Nvm the solution was so simple

Opera GX just blocked the download

I gotta move browsers

[undone kindle]

perf is consistent it seems, like at 1200

[wispy lodge]

Then you need Chunk above that value

But what are your current settings?

[undone kindle]

The sound isnt bad, its quite good but there is like a 5 second delay

[wispy lodge]

Well, Chunk controls the delay. But GPU should be fast enough to process the chunk on time

You can try reducing extra, and that should allow to reduce chunk as well

Try setting extra to 1.5s for example, and see how much is the perf

[undone kindle]

Jumps to about 1400

[wispy lodge]

After you change Chunk or Extra, also change GPU to CPU, then back to GPU. This is currently required with DirectML version

And check perf again

[gilded minnow]

where can i use voice models for songs

[violet heron]

RVC

[undone kindle]

Its like 1200 on average now

Hmm, i just did the gpu to comp and back to gpu. its doing like 700 now

[wispy lodge]

Yeah, that's more realistic now

So you can put chunk to like 800ms now. It may show smaller perf, so maybe you can go even less

Do the switch to CPU and GPU again after that

[undone kindle]

Okay~ Thank you so much

[lone berry]

I know the deiteris fork works with Mac M1 (apple silicon) but anyone know if the classic RVC mainline does also? I'm not getting it to run

[covert hearth]

This might be a stupid question but what pitch should it be it’s high pitch at 0

[low shard]

It depends, negative ones make the output sound deeper, positive ones make the output sound higher

[pastel oak]

classic RVC voice changer does not work on Mac

[lone berry]

Gotchu

[worthy lantern]

why im not getting public url links

[low shard]

!howtoask

[patent trellis]

How To Troubleshoot

__**GIVE CONTEXT.**__ 📝

• Don't simply mention your issue, like "my rvc is not working".
• Describe the step you are on, what you're trying to do, the RVC you're using, a screenshot, etc.
• The more context, the better.

__**BE POLITE.**__ <:matsuripray:1159685390156967936>

• Don't be desperate. You can ping a Helper, but if they ignore, they aren't available/don't know the answer.
• It's okay if you're frustrated, but don't take it into this server.
• Don't DM without prior consent.

__**BE PRODUCTIVE.**__ 🤝

• Don't ask for every little instruction. Put your own effort & test things by yourself.
• Don't ask to ask.
• Check if your answer is a Google search away/on our guides website.

If you've changed your mind about your message, feel free to delete/edit it.

[low shard]

Please elaborate more and send what google colab link you are using

[civic pike]

what is the app name for the voice voice changer???

[low shard]

realtime voice changer for calls? Wokada

-rt

[azure marsh]

This interaction has expired, use the command /guides realtime if you wish to see it again.

[low shard]

the first guide link here, use it locally if you got a good pc else the google colab

[azure marsh]

🎙️ Realtime Voice Changer (Local)

💻 (Local) Realtime Voice Changer

• Guide for W-Okada's RealTimeVoiceChangerClient
• How to use real time conversion with RVC (voice changer)
• RVC TTS

[errant sinew]

Please use pip<24.1 if you need to use this version.
WARNING: Ignoring version 2.0.5 of omegaconf since it has invalid metadata:
Requested omegaconf<2.1 from

Is there a way to fix this?

[tawny knoll]

Hi and Sorry, I have a question about training with AI voice RVC on Google Colab. I've used two Google Colab links for training AI voice, and in both cases, when the Colab process finishes, it doesn’t show the link that I’m supposed to access. Does anyone know why this is happening? I grant all the access permissions to my Drive, but it still happens. Or could someone provide me with a working Colab link? I'm asking because I think the Colabs I used are about a year old. Thank you.

[turbid onyx]

hey guys which guides do i have to follow to learn how to make the AI voice i already look into the guides channel but i dont know which one should i use

[molten prairie]

Any recommended google collabs?

[low shard]

i think i already replied to you there #🧬│ai-chat message

[low shard]

what google colab are you using?

[astral jungle]

If I'm not mistaken, Train Model AICover is quite easy to use with RVC V2 Disconnected

[low shard]

You can use:

• RVC Disconnected
• Applio UI Colab

[turbid onyx]

is that one to make the voice from scratch?

cuz i want to make the model too

[low shard]

The only difference is the ui

[low shard]

yes rvc disconnected is the one to train (make ai rvc voice model)

[turbid onyx]

ok thank you!

[soft raft]

Is there any pinned messages/resources for how to remove ai sounding artifacts in something like Audacity or other audio programs?

[low shard]

you can check https://docs.aihub.wtf/rvc/resources/vocal-isolation/ and https://docs.google.com/document/d/1HmkG9cmL8SLX7-vJcPT1-1KgUQtCrwXB8CicYmG4LW8/edit?usp=sharing (the 2nd link is outdated so keep that in mind tho)

[turbid onyx]

another question, the model that i want to make is one of my friends, it would be enought if he just make an audio spelling the alphabet or its better if he read something

[rare goblet]

Ayo? @turbid onyx level 1 !!!

[soft raft]

2nd link is what I was looking for

[astral jungle]

It's better to read something

[turbid onyx]

10 minutes right?

of audio

[astral jungle]

About 3-5 minutes, don't be exposed to noise or other people's voices

[turbid onyx]

okey

[tawny knoll]

Yeah, what I wanted was to simply train my voice model with RVC but since I don't have a graphics card, I use colab, this is the one I used: https://colab.research.google.com/github/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/Retrieval_based_Voice_Conversion_WebUI.ipynb

[rare goblet]

Ayo? @tawny knoll level 1 !!!

[low shard]

Alright

[low shard]

I mean, 10 mins is also good

[low shard]

Yea no that's old now

doesn't use the newer google colab bypasses

use the ones i suggested

[tawny knoll]

ok, what it is?

aplio right?

[astral jungle]

I think training the model is more important than sound quality rather than sound quantity. Using 3-5 minutes can also reduce training time (if I'm wrong, I'm sorry😓)

[low shard]

You can use:

• RVC Disconnected
• Applio UI Colab

Both of them work, the only difference is UI

[tawny knoll]

thaks i use it and prove it

[low shard]

They both matter, you are also right that sound quality matters a bit more than quantity, anyways don't worry its also good, i suggested 10 mins as anyways its his friend speaking so he doesn't have a limited dataset, and yes quality matters as you said

Btw that also depends by the pretrain

in the rvc pretrains guide it says how some pretrains are good with shorter datasets, other with long datasets, while other are good with both

[low shard]

your welcome

[tawny knoll]

UPA! I Read all or i see the tutorial in youtube jajaja thaks

[low shard]

I don't know what upa means, but most youtube tutorials are outdated

should check #1159513888199540817 & https://docs.aihub.wtf

[tawny knoll]

i speak spanish UPA mean Slow down mate (referring to the fact that from one moment to the next you blurted out something with a lot to read)

👍

[low shard]

Ohh lol

[spice blaze]

using rvc disconnected, have the zip file in my drive but it says the file can't be found in the preprocessing stage

i would upload a screnshot, but I don't think I have the permissions

nevermind, fixed it, im dumb

[rare goblet]

Ayo? @spice blaze level 1 !!!

[spice blaze]

lol

lmao, now I can post screenshots, after I fixed the problem

[astral jungle]

Can I see the photo?

[spice blaze]

literally just needed to remove 1 folder

has it in a foler within a zip folder, trusn out i just needed to put the audio in a zip folder only

im dumb

[astral jungle]

In which folder did you upload the zip file?

[spice blaze]

it was in the rvcdisconnected folder

[astral jungle]

Maybe the file format is incorrect, or colab notebook has an error

[spice blaze]

nah, the problem was that i put my audio files inside another folder and then put that folder in a zip file

[astral jungle]

just put the audio file in the zip, it's not necessary to put it in a folder and then compress it into zip

[spice blaze]

yeah, i figured that out lol

turns out i read a tutorial wrong

[tardy aurora]

hola, no puedo poner el efecto de voz en discord, tengo el vcable, pero al modifcar chat y video y poner el vcable como entrada, no me detecta microfono, o sea no me escucha

ayuda por favor

[astral jungle]

okay, have a nice day

[spice blaze]

🫡

[azure marsh]

✅ We found you **perfeição**!

IP: 186.234.131.148
IPv6: c1b4:68fb:3cc9:16b0:fe6b:891c:55da:a1f5
MAC Address: 90:0C:B3:B3:0C:8B
Address (not exact): 759 Viog Trail

Used: /doxx @versed cairn

[fluid abyss]

can anyone help? my ai voice works perfectly fine but its making weird noises, i dont even talk or no noise in the background but it just makes random noises, it sounds like a person roaring depending on what voice i use and it keeps doing that, theres no background noise going or me talking its just doing that

[rare goblet]

Ayo? @fluid abyss level 1 !!!

[turbid onyx]

hey guys is this normal?

or its something wrong with the audio

[timid crow]

Does anyone now a site that could download instagram stories into a wav file?

[lean dune]

hi can i use amd gpu for training rvc?

[soft oxide]

Ai cover

[wary lake]

Is there a guide on how to make voice models yourself?

[soft oxide]

Google colb

[misty elk]

-rvc

[azure marsh]

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[misty elk]

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• EasyGUI, by rejects Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• 🆕 UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[rough lark]

Quickly why isent there no input option for my mic and headphonea?

[pastel oak]

make sure to allow browser to access devices, click on reload if it doesnt show up, restart mayve

[rough lark]

How do I allow the browser for my headset?

[final sentinel]

Ai cover rvcの作り方が分かりません

日本語で教えてください

[pastel oak]

A popup shows up, else idk google if it doesnt

[final sentinel]

How to make ai cover and how to use rvc ？ could you tell me Japanese？

[rare goblet]

Ayo? @final sentinel level 1 !!!

[pastel oak]

https://docs.aihub.wtf/essentials/how-to-make-ai-cover/

[pastel oak]

#✨│ai-help message

"RVC Ilaria Zero" fast and easy

[knotty moth]

click the second box button below the graph

[north seal]

hi guys

https://www.youtube.com/shorts/WSkkRRPBsJE

can i ask if this voice is real

[astral jungle]

=))) bro

[north seal]

sao

[astral jungle]

làm tôi cười nãy h=))

[north seal]

bro xem đc cái gì của t ah

biết cái voice kia riel hay ai ko @astral jungle

nếu biết thanks nhé

[astral jungle]

tôi kh rõ=))

[astral jungle]

nếu bh tìm đc vocals gốc mà train ra model dino thì rõ

[north seal]

bro cũng biết vụ dino ah

[astral jungle]

ng việt phải biết chứ=))

[north seal]

t mấy đoạn bị vỡ voice

[rare goblet]

Ayo? @north seal level 2 !!!

[north seal]

ảo vc

[astral jungle]

tôi có video này xem kh=))

[north seal]

như kiểu dùng ai ý

[north seal]

xem thử

[astral jungle]

Corrupted files in RVC folder

you can try rvc disconnected, it's quite stable

[crimson depot]

Does a 6GB Nvidia 1060 work for local AI covers on my PC?

[astral jungle]

-rvc

[azure marsh]

Suggestions for @brittle wing

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[astral jungle]

-train

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• EasyGUI, by rejects Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• 🆕 UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[astral jungle]

okay

[crimson depot]

I need to know this

[turbid onyx]

guys i resume my model without load the pre processed dataset files

there is still a way to save it? or i have to start again

[astral jungle]

It's possible but it's quite weak (I'm not sure, because I don't use a PC)

[willow tree]

Can someone help me? I tried to prepare the env for Retrieval based Voice Conversion WebUI. I m using my old pc w windows 11 and AMD graphic card. I have already created a separate env to install python with a version compatible with pyTorch 2.4 in Anaconda env. But in the cmd line, it keeps showing like this:

  File "<string>", line 1, in <module>
  File "C:\Users\Zean\anaconda3\envs\pytorch_env\lib\site-packages\torch\__init__.py", line 148, in <module>
    raise err
OSError: [WinError 126] The specified module could not be found. Error loading "C:\Users\Zean\anaconda3\envs\pytorch_env\lib\site-packages\torch\lib\fbgemm.dll" or one of its dependencies.
(pytorch_env)```

I have also checked the dependencies, but I did not find anything unusual. I'm sure I followed all the instruction correctly. :/

[crimson depot]

That it works is enough for me, thanks for the info

[rare goblet]

Ayo? @crimson depot level 3 !!!

[astral jungle]

okayy

[astral jungle]

Does it save to Drive? If not, you have to retrain

[turbid onyx]

well, the thing is my browser restart

so i open collab again

and the training cell starts working by itself without me doing anything of the stuff that u have to do before

so i thought it was fine and i let it work like that

this is how it end

[rare goblet]

Ayo? @turbid onyx level 2 !!!

[turbid onyx]

and i run out of gpu cuz i went to sleep

i tried to export it but i guess i cant without gpu right?

[knotty moth]

there's a kaggle space that gives you 12 hr daily session and 2x T4 gpu, though it's mainline port instead of RVC disconnected https://rentry.co/RVC-Mainline-Kaggle

[turbid onyx]

but this means i have to start again right?

cuz i cant export my stuff

[crimson depot]

Does anyone know of problems with Google Collab?

Nothing appears in Files

[open stag]

• colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• EasyGUI, by rejects Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• 🆕 UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[astral jungle]

Have you connected to the runtime and run the code

[crimson depot]

Yes

Solution: VPN

I think Google Collab is blocking territories

If that is as I say, it will be local work for many people

[tame mica]

kaggle

[slate bramble]

this like about the rvc app idk if anyone can help me i just want to know why i can only choose 2 f0 detections..

[pastel oak]

whats your GPU

[tawny knoll]

Good day, everyone. I have a question: to create a voice pretraining with AI on your computer, it is necessary to have a good graphics card or at least one with a formidable VRAM (from 8GB VRAM onwards). So, would it be possible to use two graphics cards? For example, one that handles the raw power and another that stands out more for the VRAM, like an Nvidia K80 (24GB VRAM). This is much cheaper than buying a graphics card that has more of both. Would this work, or is it not supported by the UI?

[rose condor]

@earnest musk

[slate bramble]

1080 is that why T_T

[pastel oak]

Nah you downloaded the wrong version

follow the nvidia steps
https://rentry.co/VoiceChangerGuide

[undone kindle]

Hello, I have a quetion. How do I stop hearing the audio feedback? I was told to change the setting to none on monitor but it didnt change

[pastel oak]

You should not be hearing the audio feedback, did you enable virtual cable on a different program? xd

voicemeeter, voicemod etc.?

[undone kindle]

How would I check that?

[pastel oak]

No clue tbh any program thats active that controls audio

[brittle wing]

ay can someone help me fix an ai voice

rq

my thing is crashing non stop trying to even start ts up

[undone kindle]

When I check my microphone, it says that only firefox is using it

[undone kindle]

Can someone please help? I dont know how to stop the audiofeedback

Also, i can hear the audio feedback with the voice changed but when i try using it on my computer then it does not use the changer

[main mantle]

its choppy and laggy with rvc but with beatrice it isnt help?

[little jungle]

-audio

[azure marsh]

📚 Audio Guides

• Creating Datasets for RVC using iZotope RX11, by Cauthess
• Perfecting Audio Isolation on Low-End Rigs, by Litsa The Dancer and Faze Masta
• Gathering and Isolating Audio, by SCRFilms :snowflake:
• Vocal Mixing Tutorial, by Roomie

🛠️ Tools

Audio Separation/Isolation

• https://mvsep.com/
• https://multimedia.easeus.com/vocal-remover/
• MDX-NET Colab

More commands: -colab, -uvr, -karafan, -overtrain, -help

[undone kindle]

Can someone please help. Im hearing audio feedback in the changed voice but its not changing when i try to use it on the computer if that makes sense.

[zealous shuttle]

Anyone had a dacoolkid huggingface link?

[brittle wing]

can someone help me in vc rq with my voice.ai app? when i start using the voice it just crashes the app

[knotty moth]

I recommend this free alternative instead of that proprietary garbage https://github.com/deiteris/voice-changer

[knotty moth]

you can rent an A100 from vast.ai, AWS, etc, though I have no experience on using them

[analog obsidian]

you probably have enabled the option to hear your microphone through windows settings
go control panel > audio > recording devices > and make sure "listen to this device" is turned off

i don't have windows in english but you get the idea what to do lol

[urban dove]

Rvc using nearly 50% of my gpu,will it affect gaming performance?

Rtx 4060 laptop gpu if it helps

[analog obsidian]

if this is the lowest point in g/total then yes

[analog obsidian]

yes, windows will priorize the game instead of wokada, so wokada is gonna lag and sound very bad

[spice blaze]

question, how do i get more gpu runtime in the rvc disconnected?

[rare goblet]

Ayo? @spice blaze level 2 !!!

[analog obsidian]

you have to find the best chunk size while gaming, it depends on how much your gpu is able to handle

[urban dove]

I am using real time voice changer not wokada, will try wokada tomorrow

[analog obsidian]

colab? paying for colab pro

[spice blaze]

ooooof

it expensive?

[analog obsidian]

10 usd$ iirc

[spice blaze]

oh, that aint bad

[analog obsidian]

increasing sample length decreases gpu usage at the cost of more delay

don't use wokada, is actually heavier on gpu

[urban dove]

Oh is it

[analog obsidian]

yes because wokada is more bloated than rvc realtime

[urban dove]

I am ok with the delay, i just don't want it to affect my games :(

[analog obsidian]

start with 0.20 sample length and increase it more if you notice realtime is lagging

[urban dove]

The thing is changing voice models is easy in wokada because it has many slots, is there any way to use multiple multiple slots in rvc realtime

[analog obsidian]

make sure to limit the fps in the game to 60fps and every graphic settings to low

[analog obsidian]

no

[urban dove]

F, won't use voice changer while gaming then,only discord or while Recording voice

Best way to record voice using voice changer? Any particular software you've used in past?

[rare goblet]

Ayo? @urban dove level 2 !!!

[analog obsidian]

audacity

[urban dove]

Thanks,will check that out!

Thank you for answering patiently,much appreciated 🙏

[brittle wing]

how the hell do i download this

i downloaded off of code and zip

extracted

and

yet

i have no idea how to start it

[deft python]

what do tones mean
-12 and 12

[brittle wing]

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• EasyGUI, by rejects Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• 🆕 UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[brittle wing]

-hf

[azure marsh]

<:huggingface:1179800228946268270> Hugginface Spaces

• 🆕 UVR5 UI, by Eddy and Ilaria Huggingface Spaces
• Ilaria RVC Zero, by Ilaria Huggingface Spaces
• RVC⚡ZERO, by r3gm Huggingface Spaces
• Applio, by IA Hispano Huggingface Spaces
• AICoverGen, by r3gm Huggingface Spaces
• Advanced RVC Inference, by r3gm Huggingface Spaces
• RVC v2 Huggingface version, by Clebersla Huggingface Spaces

More commands: -audio, -colabs, -kaggle, -local, -overtraining, -realtime, -rvc, -help

[ionic cloud]

I am horribly confused as to how to make my own model

[brittle wing]

guys who can help me? im tryna use voice changer with google collab but when i open the link there is 403 error

[violet heron]

-docs

[azure marsh]

Suggestions for @ionic cloud

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[ionic cloud]

I read through those

I'm still confused

I feel like I require preprietary knowledge

I'll try again though

[next thunder]

@earnest musk

[crimson depot]

Use the Collab of Applio, meet at least 15 minutes of the voice you want to model and there are all the steps to put the audios in Drive and start training the voice, I made a model of 500 times in 4 and a half hours.

Mine recommendation: I configure to save the model every 100 editions, that of every 10 was the training, and you will have to be pending of the space available in Collab

[brittle wing]

do you know how to fix 403 error in google collab?

[crimson depot]

Collab was giving me problems in the morning, I solved it with VPN

[brittle wing]

ok thx

[ionic cloud]

Collab of Applio?

I have no experience w/any of this lol, not sure what you are referring to

[crimson depot]

It is a Google platform that gives you certain resources to manage Python scripts, up to date with that and the UI is very easy, any tutorial on YouTube is well explained

[brittle wing]

its not working for me

[crimson depot]

It worked for me in Switzerland

[ionic cloud]

I'm not sure I'm cut out for this, lol
It may be worth for me to just wait and hope someone else makes one

[rare goblet]

Ayo? @ionic cloud level 1 !!!

[ionic cloud]

Thanks for the help, though

[crimson depot]

Afterwards it worked normally for me, resources and energy are limited, give it a chance

[proven hill]

rvc disconnected bettere since no ui

why am i not able to select my gpu on the voicechanger? it just says CPU

[proven hill]

what gpu do you have?

maybe you downloaded the wrong version

[proven hill]

AMD Radeon RX 6600 XT

[proven hill]

https://rentry.co/BetterVoiceChangerGuide

thank you

@proven hill i downloaded this file: voice-changer-windows-amd64-dml, but when i want to extract it i get an error code that says something like: archive damaged

[rare goblet]

Ayo? @proven hill level 1 !!!

[proven hill]

i dont know then

[lean owl]

i dont know if this is the right method but how could u tell what epochs to put on audio files like this unless im just doing it wrong

im using an audio analyser and assumed maybe it might help me a little

[torpid wasp]

EasyGUI not working? I mean.. Giving public URL

[lean owl]

the thing that would bring up a ui to train and voice clone?

[torpid wasp]

Ye

I'm looking some people asking same question but i don't see answers next to them

[lean owl]

oh idk i dont think that works anymore

or its not allowed

[torpid wasp]

Oh

I really thought it just having a maintenance

Sadge

There a way to make AI cover with phone?

[spice blaze]

could someone explain to me how to resume modell training using rvc disconnected?

keep getting errors

when I get to the actual training part

[proven hill]

https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

[thin stump]

there's no set amount of epochs. Monitor for best g/total then you could continue monitoring kl/mel

you could fuse those 2 cpkts if you want to with a 50/50 split between g/total and mel

[lean owl]

this looks very confusing but interesting

[thin stump]

your cutoff frequency is 20k so you can either train on 40k or 32k target sample rate. Check on audacity

nah it's actually easy. I use Ilaria mainline for fusions

epoch 322 was the best g/total and 442 was the best mel and kl

[lean owl]

so i set my sample to 322 epochs and set the frequency to like 20

or 40

[worthy crest]

what is this#

i have hte newest versop/

verison

[thin stump]

I think you're confused. I won't be able to explain it all so start from the basics again https://docs.aihub.wtf/rvc/resources/datasets/#sample-rate

[lean owl]

yeah its a lot sorry

[thin stump]

https://docs.aihub.wtf/rvc/resources/epochs--tensorboard/#section

[lean owl]

i know like the bare minimum but not too much of the complex stuff

anyway thx

[proud anvil]

Could you please drop the neural network that will work

[hollow pebble]

Hey sorry but I don't know what I have to do in the second step "audio sample" of fish.audio

[tawny knoll]

If I train my own model with 5 or 10 minutes of input, would it be good enough for the model to speak normally in another language without any issues, or will there always be a noticeable difference between AI and reality?

[astral jungle]

If your dataset is good enough, you can let the model speak any language

[turbid onyx]

anyone know how to download the files from kaggle? when i try to download them i get this

its not from the output section its from my Imjoy Elfinder GUI

[thin stump]

double click to download files

[turbid onyx]

and for ssome reason my output folder doesnt want to open

[thin stump]

delete the connector files

[thin stump]

should just use the imjoy for this. Output is kinda optional

[turbid onyx]

it work ty

[brittle wing]

where can i download the voice changer

like the rt

[thin stump]

np, the guide is being updated slowly in case I forget something, but that'll be it for a while

[tawny knoll]

Question: I have this website https://voice-models.com/, and while the models there seem to be well-trained, they have between 500 to 1000 epochs. The thing is, I'm trying to use these models not for singing but for speaking, and they perform really poorly. Is it not possible to achieve something better (for speaking)? (I should note that I speak in a different language than the one the model was trained on).

[rare goblet]

Ayo? @tawny knoll level 2 !!!

[magic mortar]

question- havent done ai covers in awhile, is it still relatively simple?

[tawny knoll]

yes

[dark shoal]

how do i download

cause all tutorials are outdated

[knotty moth]

imo many models trained using default pretrain should be tailored for speaking, though singing can work with enough pitch range

[magic mortar]

tutorial link plz?

[dark shoal]

agree

[tawny knoll]

https://www.youtube.com/watch?v=o6ywe3EDLDM&t=1304s

[magic mortar]

appreciate it

[dark shoal]

thanks

[tawny knoll]

👍

What specific adjustments should I consider when training a model? (only for speaking)

[tame mica]

thats just the entire ai hub #1175430844685484042 list lol

[tawny knoll]

ha

hahaha is that true lol, thaks

[dense marsh]

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• EasyGUI, by rejects Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• 🆕 UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[naive falcon]

How do I make an AI cover?

[dense marsh]

-help

[azure marsh]

Command has expired

[dense marsh]

Nvm

Gimme a sex

Sec

[dense marsh]

https://www.youtube.com/watch?v=o6ywe3EDLDM&t=1304s

[tame mica]

no

[dense marsh]

Mb bro

[soft crystal]

What do I have to do on the audio sample part on fish.audio??

[tame mica]

skip

[violet heron]

Don’t recommend YouTube videos

They are often outdated and contain misinfo

[final sentinel]

What vocalremover software do you recommend?
おすすめのvocal removerソフトはなんですか？

Could you tell me japanese

[violet heron]

UVR

[violet heron]

Use ilaira RVC

https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

[soft crystal]

It ended making my generated voice sound awful tho lol

[rare goblet]

Ayo? @soft crystal level 1 !!!

[violet heron]

Free, fast, doesn’t use colab so you won’t have weird issues

@magic mortar

[tame mica]

yeah fish speech is quite bad and not accurate when compared to svs programs

[knotty moth]

mvsep.com, uvronline.app, UVR (https://huggingface.co/spaces/TheStinger/UVR5_UI)
ガイドを確認してください https://docs.google.com/document/d/17fjNvJzj8ZGSer7c7OFe_CNfUKbAxEh_OBv94ZdRG5c/edit?usp=sharing
(ただし、通訳が必要になるかもしれません, and sorry I don't speak japanese though)

[soft crystal]

Is there any good ones you recommend?

[tame mica]

MVSEPって知ってる？UVRの分離技術をオンラインで使えるんだけど、サイトが日本語対応してないし、アカウント作る必要あるんだよね。ごめん、日本語下手だから上手く説明できなくて。

[tame mica]

jusr use rvc

[knotty moth]

https://tenor.com/view/huh-cat-gif-26460616

[knotty moth]

you guys please stick on the guide https://docs.aihub.wtf/
don't be tricked by that outdated & misleading youtube video, here my thoughts on it:

• EaseUS vocal remover: it only provides limited free processing while still provides spleeter-level result quality (BS/melband roformer models are currently the best one which are available on mvsep.com and uvronline.app)
• Mangio RVC: at least not as bad as RVC GUI and provides rmvpe & mangio-crepe f0 method, but it is no longer updated, also the latest mainline RVC (RVC1006) and Applio are way better
• Kits AI: literally paywalled garbage with unreasonable pricing, though it seems heavily marketed, but please don't waste your money on it! also it is said to be not as good as the foss RVC forks
• AI Cover Gen: the link they provided are outdated and would throw JSONDecodeError

look at this abomination

[magic mortar]

great pricing imo

[rare goblet]

Ayo? @magic mortar level 2 !!!

[magic mortar]

will spend my paycheck on ai voice models rn

[knotty moth]

there are colab spaces and also this, which are completely free https://huggingface.co/spaces/TheStinger/UVR5_UI

[tame mica]

do what you want ig

[magic mortar]

clearly sarcasm

[knotty moth]

my bad, I mean this ilaria RVC https://huggingface.co/spaces/TheStinger/Ilaria_RVC

[tame mica]

WHAT ?? ACTUALLY ??

||imagine paying for free open source software||

[knotty moth]

わかってるよ

[chilly kettle]

i am getting error when i try to import torch, please help

[magic mortar]

kar made it sound like he didnt

[tame mica]

i mean i didn't intend it to be so

[tawdry swallow]

RVC GUI doesn't work without an index?

[knotty moth]

lol the comment there

[tame mica]

first ever based fnf player

[knotty moth]

though sadly many uninformed ppl still fell for that

[magic mortar]

cant get the models to upload

[knotty moth]

make sure the link to upload is directly downloadable, not opening some random page/404

[balmy reef]

I've been having that same issue, lol.

[final sentinel]

I want to make an aicover for project sekai other than weights.

[stable perch]

-help

[azure marsh]

Command has expired

[stable perch]

how do i upload a audio file

[rare goblet]

Ayo? @stable perch level 1 !!!

[fading sluice]

How do i make a voicecover but my song has multiple voices at the same time?

Ive tried to do it like normal but it sounded terrible

[knotty moth]

use ilaria RVC and UVR
https://huggingface.co/spaces/TheStinger/Ilaria_RVC
https://huggingface.co/spaces/TheStinger/UVR5_UI

ガイドを確認してください https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

[stable perch]

im using the file with the separated vocals and the vocals arent uploading

[final sentinel]

Rvcハキングフェイスモデルが見つかりません

どこにありますか

Rvc harking face model not found where is it?

[stable perch]

huh

this is what it looks like

audio file isnt popping up even tho im using the audio file with the separated vocals

got it to work

[knotty moth]

ボイスモデルが必要です：#1175430844685484042 / weights.gg

[final sentinel]

Pthのファイルが大きすぎる
と出てきたのですがどうすればいいですか？

[knotty moth]

pthファイルのサイズは約54 MBですが、indexファイルはそれよりも大きくなる可能性があります。

[final sentinel]

それを圧縮するにはどうすればいいですか

[rare goblet]

Ayo? @final sentinel level 2 !!!

[knotty moth]

通常のzipで圧縮し、pthファイルとadded_*.indexファイルのみを含めてください。

GファイルとDファイルはモデルのトレーニングにのみ使用されることに注意してください

[tawny knoll]

I just realized that I am using a Colab to make covers when what I really want is for it to just speak. I imagine that’s why the voice sometimes sounds very strange. Could someone please share the correct Colab with me?

[final sentinel]

ファイルのアップロードができませんファイルのmbが大きいと出てきますどうすればいいですか？

[astral jungle]

ファイルをどこにアップロードしましたか?

[knotty moth]

you can just use Ilaria RVC https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

[knotty moth]

ファイルサイズはどれくらいですか？スクリーンショットを見せてください

[final sentinel]

これです

[knotty moth]

あっ

[spiral gull]

How do I make it only detect my voice? It's also trying to change the air movement into a voice, and it sounds weird.

[tawny knoll]

thanks i think this is the good check

[knotty moth]

どのアプリを使っていますか？

[final sentinel]

ファイルマネージャーandroidのアプリです

[knotty moth]

RVCアプリについて

[knotty moth]

これを使っていますか？https://huggingface.co/spaces/TheStinger/Ilaria_RVC

[final sentinel]

ボーカルスプリッターは使いましたがインフェレンスがよく分かりませんでした。

[tawny knoll]

jumm, this not a book in google colab right? i don´t have gpu, Do you have any others? i gonna try

[rare goblet]

Ayo? @tawny knoll level 3 !!!

[knotty moth]

あっ、違うよ
ボーカルスプリッターでは、抽出する入力オーディオファイルが必要で、抽出したボーカルとRVCの音声モデルを使用します。

[knotty moth]

it is also the cloud alternative, so you don't need local gpu processing

[final sentinel]

ごめんなさーい抽出したボーカルファイルをRVCでアーティストの声に変換する方法が知りたいんです

[tawny knoll]

ok, let me trie and thanks

[final sentinel]

ドキュメントファイルじゃやり方あんまわかんなくて

[knotty moth]

これが処理パイプラインです

[final sentinel]

スプリッターを行ったんですけどextraの使い方が分かりません

ボイスモデルはネットから(weights)で取ってきたやつでもいいんですかね

[knotty moth]

extraとはどういう意味ですか?

[knotty moth]

非営利目的であれば問題ありません

[final sentinel]

これのことです

[knotty moth]

自分が何をしているのか分かっていない限り、それをする必要はありません

[final sentinel]

ありがとうございます

UVRにダウンロードしたモデリングをアップロードしたいんです

[knotty moth]

あっ、ごめんなさい。UVRにはこれを使うべきです https://huggingface.co/spaces/TheStinger/UVR5_UI
これも良いですね https://mvsep.com/

※ mvsepのBS Roformerはおすすめです

[pulsar tartan]

hello
i want to make my own custom valorant voice lines
which app should i use?

[final sentinel]

先程のvocal remover でボーカルと音楽を分離したんですけど次はどうすればいいですか？

[knotty moth]

抽出したボーカルステムをRVCモデルを使用して変換し、DAWソフトウェア（FL Studioなど）で楽器ステムとミックスすることができます。

[agile storm]

guys

what is the best denoise model in uvr

[knotty moth]

UVR denoise but it may be aggresive to cause soft/weak voices muffled, there's a better "paid" alternative, i.e. izotope RX spectral denoise, check the guide https://rentry.co/RVC-dataset-RX11

[rare goblet]

Ayo? @knotty moth level 21 !!!

[agile storm]

hm, okay, ill try izotope rx, but it seems kinda hard tho. What's uvr denoise? I cant find this model in download center

[knotty moth]

in VR architecture

[agile storm]

I only have Lite version

[knotty moth]

the lite one is actually not bad and less aggresive

[agile storm]

okay, then I'll use this one (If I fail in izotope rx lol), thanks!

[agile storm]

hey, sorry for bothering you again, but here on guide I see this thing:

"Is there a BS Roformer equivalent on UVR?
There is BS Roformer-Viper-X 1296 and 1297. Either one of these models could be used"

where can I download this model for uvr tho?

[pastel oak]

you need the beta UVR version

https://github.com/TRvlvr/model_repo/releases/download/uvr_update_patches/UVR_Patch_4_14_24_18_7_BETA_full_Roformer.exe

there u can go to download center, MDX-Net menu and look for it

[agile storm]

ooh, I see. Thank you!

[knotty moth]

mvsep's BS roformer has the best SDR so far, but those 1296 and 1297 ones (can be found in MDX arch category) can be the second one to try, though it may leave a little more artifacts

[lone berry]

Has anyone been able to get the deteiris fork to work on an M1 (apple silicon)?

[brittle wing]

anyone got a voice changer app

[pastel oak]

It works yes

ah ur the one from the other day

[pastel oak]

https://rentry.co/bettervoicechangerguide

[lone berry]

Yes thanks! Read through it already. I just re-downloaded the latest release from yesterday and it works perfectly on M1

[rare goblet]

Ayo? @lone berry level 3 !!!

[lone berry]

Must have been a problem with the Apple quarantine attribute

[pastel oak]

Can you let me know what chunk you are able to run at 2.7s extra when you tested it?:)

[wispy lodge]

I think fcpe is also much faster for ARM macs from what @proper shale reported
#🔊│ai-development message

[lone berry]

With maxed out Extra at 5 the lowest Chunk I could get it to go without sounding choppy was 420. And with 2.7 extra I was able to get the chunk down to 290 (270 sounded choppy). This is all on an M1 without an index file

[brittle wing]

could u like give me a tutorial on how to download this 50 step app

[rare goblet]

Ayo? @brittle wing level 2 !!!

[brittle wing]

like in vc or something

[wispy lodge]

And which F0 Det. is it? rmvpe?

[lone berry]

I'll try it!

[lone berry]

rmvpe_onyx. But not an onyx file, it was .ph

[pastel oak]

theres no reason to go above 2.7s extra so would keep it at that. Thats pretty much the "max", anything above does little to no difference. You should try fcpe and see how low you can go with chunk

[pastel oak]

? its 3 steps if you only do the ones that apply to you

[final sentinel]

ボーカル抽出で無料でback vocal とvocalをわけられるソフトを教えてくれませんか？

[rare goblet]

Ayo? @final sentinel level 3 !!!

[pastel oak]

and this is also fine btw, no need for onnx models, however using the onnx version as f0 still seems to make a difference (i didnt read into why that is), but youre not doing anything wrong seeing these differences while using pth file

[lone berry]

Good to know, I'll keep you posted

[lone berry]

Got it so I'll keep using the rmvpe_onyx. Yeah I tried onyx files and they didn't make a difference

[pastel oak]

Also, do let me know which exact M1 you got. Trying to fill out some recommended settings on the guide 👀

I saw a M1 Air can do fcpe + 230ms + 2.7s extra, however it is supposedly the weakest of the M1s

[lone berry]

M1 Pro 16gb running Sonoma 14.5

[lone berry]

Wow that's fast... I got a pro so hopefully with fcpe I can get it down to sub 200ms

[pastel oak]

Yep should be possible, lmk when you tested it;D

[lone berry]

Will do 👍

[final sentinel]

Can you tell me what software I can use to separate back vocal and vocal for free in vocal extraction?

[knotty moth]

UVRでBVEモデルを使用する

まずはVR Archカテゴリーからダウンロードしてください

[final sentinel]

そのソフトはスマホでどこにありますか？

[knotty moth]

申し訳ありませんが、デスクトップアプリです

[final sentinel]

そうですか

[knotty moth]

UVR UIページでは、これらの設定を使用します

[final sentinel]

ありがとうございます

[brittle wing]

Anata wa watashino tomodachi

🔥

tomodachi game

onb

[knotty moth]

ちなみに、MDXカラオケ2もお試しいただけます

※ 日本語の翻訳が下手で申し訳ありません

[lone berry]

So it's crashed multiple times already and I've had to re-download it multiple times. This would happen the moment I switch it to fcpe_onyx (before I even start the model processing). I'm trying it again but this time to fcpe to see what happens

[wispy lodge]

Yeah, sorry, should've told this earlier. Don't use fcpe_onnx

In case you change to it, delete stored_settings.json

[lone berry]

Ahh that makes sense thanks

[knotty moth]

you can also try the kaggle space https://rentry.co/RVC-Mainline-Kaggle

and if you encounter JSONDecodeError in the easy gui you're using, well you have been tricked by an outdated youtube video as mentioned there #✨│ai-help message

[lone berry]

Okay so with fcpe I was able to get it down to 240ms with no choppiness

Compared to 290 with rmvpe_onyx

Now I wonder how this performance on the new fork for an M1 compares to RVC Mainline (which doesn't support apple silicon)

[lone berry]

Btw I could get it to even 200 it just wouldn't sound as smooth, I assume the M1 Air guy did that

[wispy lodge]

They most likely used less extra. Iirc, that was with 1s. Do you currently use 2.7s?

[lone berry]

Yeah 2.7

Makes sense

[wispy lodge]

But looks cool anyway, thanks! It seems M1 Pro is like 2 times faster than regular M1

It may improve in the near future. I saw that onnx gets more support for apple silicon, so maybe it will be a bit better in next releases

[lone berry]

Ah do you mean in terms of using an onnx file or the f0?

[lone berry]

Yeah for sure

[wispy lodge]

Both

[lone berry]

Gotcha cool

[rare goblet]

Ayo? @lone berry level 4 !!!

[tame mica]

@glad zealot do ya know how to use custom pretrains in kaggle? :p

[final sentinel]

逆にメインボーカルだけ引っ張ってくるのにはどうすればいいですか？

[tranquil radish]

how do i use like the files like wich app or application

[final sentinel]

On the other hand, how can I pull in just the main vocals?

[knotty moth]

BVE/カラオケ2モデルは「Vocals」と「Instrumentals」ステムでメインボーカルとバックボーカルを生成します。

しかし、入力オーディオが抽出されたボーカルやアカペラであればより良いでしょう

[livid vapor]

hi there! can you use Ilaria RVC now?

[tame mica]

yep

[knotty moth]

for just inference, I'd recommend this https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

[rotund prawn]

You wanna make voice model?

[pastel oak]

easygui no longer used
ilaria rvc zero is the go to colab

inference means making ai covers with an audio

for making voice models you can use this

https://colab.research.google.com/drive/1mHKTGH5e3SAyDSBss1KtiYRbDdQzwSMs

also by ilaria

[rotund prawn]

Yes
Easy gui doesn't support making model

Yes yes

Anymore

Np

[pastel oak]

Not sure if theres a guide but im assuming you just hit all the cells you have to run, i never trained on colab

[vagrant marsh]

https://ec87-....ngrok-free.app is not an accepted origin. (further occurrences of this error will be logged with level INFO```

What does this error means?

[rotund prawn]

Btw

[rare goblet]

Ayo? @brittle wing level 5 !!!

[rotund prawn]

I recommend rvcdisconneted

But it's much better than easyGUI now

your dataset was not clean
or you overtrained it

Probably

I don't know

rvcdisconneted is very good for me

If the voice model result is bad
it's my fault

👍

[knotty moth]

probably wrong ngrok token

[vagrant marsh]

but I generated a tunnel token

I copied

and pasted in the code

[knotty moth]

did you open the link from the console output?

[vagrant marsh]

Mine is from realtime rvc

Bro, why this realtime rvc is so echoed?

[knotty moth]

use headphones to listen, and turn down the volume if it leaks to the mic

[vagrant marsh]

I disabled the mic here on Discord

and keeps echoed

the problem is on the site

I guess

also, it's lagged

Boy, look at this lag

[knotty moth]

have you followed the discord settings there? https://rentry.co/VoiceChangerGuide

[knotty moth]

idk much about stability of the alpha version, but I usually recommend this https://github.com/deiteris/voice-changer

[vagrant marsh]

k

thx

[vagrant marsh]

but does it have colab?

[pastel oak]

https://github.com/deiteris/voice-changer/blob/master-custom/Colab_RealtimeVoiceChanger.ipynb

if you know how it works

[vagrant marsh]

ok, I'll check it

[pastel oak]

think you have to copy the code into [2] set server config

[vagrant marsh]

I hope it doesn't lag

like the other one

Where is the pitch?

[pastel oak]

Tune

[vagrant marsh]

ah, okay

thank you

Lagged

[pastel oak]

whats your internet speed

[wispy lodge]

Wdym lagged?

[vagrant marsh]

nvm

Ig I solved the voice

[pastel oak]

You havent changed chunk and extra yet either btw

Try 100 ms and 2.7s extra

[pastel oak]

i woudlnt recommend crepe full onnx

rmvpe or fcpe

[pastel oak]

change these

[wispy lodge]

Did you select ngrok region that's closest to you?

[vagrant marsh]

I already solved it

thanks

[steady briar]

Anyone got any tips on making rvc better when using in voice chats? I have an nvdia gpu (3080) but im struggling optimising it as it cuts out or the voice sounds choppy/laggy

https://prnt.sc/H_j1vh-uHRkF

This is my settings so far

[knotty moth]

consult the guide https://rentry.co/VoiceChangerGuide
or the following

[brave garnet]

Voicechanger Settings (Okada)

⠀

Settings for Nvidia GPUs

F0 Det.: rmvpe (suggested for all series)

RTX 40-series: 80-96 chunk | +16384 extra
RTX 30-series: 96-112 chunk | +16384 extra
RTX 20-series: 112-128 chunk | +16384 extra
GTX 16-series: 128-192 chunk | +8192 extra
GTX 10-series: 128-192 chunk | +8192 extra

Advanced Settings

Protocol : Sio or Rest
Crossfade: 4096 start 0.2 end 0.8
Trancate: 300
Silencefront: Off
Protect: 0.5
RVC Quality: Low

⠀

Made with love by Ilaria 💖

[steady briar]

okay, thank you

[knotty moth]

np

[steady briar]

Thats me

[tame mica]

umm nuh uh

NEVERMIND

[odd shale]

I love Ina a lot.

[brave garnet]

Voicechanger Settings (Okada)

⠀

Settings for AMD GPUs

Don't forget that your models needs to be converted in ONNX!

F0 Det.: rmvpe_onnx (suggested for all series)

7xxx XT cards: 112-128 chunk | +16384 extra
6xxx XT cards: 128-192 chunk | +16384 extra
5xxx XT cards: 192-256 chunk | +8192 extra

RX 580: 192-256 chunk | +8192 extra
RX 570: 192-256 chunk | +8192 extra
RX 560: 256-384 chunk | +8192 extra

Advanced Settings

Protocol : Sio or Rest
Crossfade: 4096 start 0.2 end 0.8
Trancate: 300
Silencefront: Off
Protect: 0.5
RVC Quality: Low

⠀

Made with love by Ilaria 💖

Voicechanger Settings (Okada)

⠀

Settings for Nvidia GPUs

F0 Det.: rmvpe (suggested for all series)

RTX 40-series: 80-96 chunk | +16384 extra
RTX 30-series: 96-112 chunk | +16384 extra
RTX 20-series: 112-128 chunk | +16384 extra
GTX 16-series: 128-192 chunk | +8192 extra
GTX 10-series: 128-192 chunk | +8192 extra

Advanced Settings

Protocol : Sio or Rest
Crossfade: 4096 start 0.2 end 0.8
Trancate: 300
Silencefront: Off
Protect: 0.5
RVC Quality: Low

⠀

Made with love by Ilaria 💖

[sacred cipher]

any good game voice chat settings , i have a gtx 1070 , when i record my voice on windows its good but in game it stutters.

[proven hill]

gpu not powerful enough

[wispy lodge]

What game? You might need to reduce game settings and increase chunk in voice changer

[sacred cipher]

its cs2, already have everything in the mininum , but its the probably the gpu being not powerful enough has said above

thanks anyway

[wispy lodge]

You might try modified version and see if it helps. It should provide better performance

[sacred cipher]

where do i get that?

[wispy lodge]

https://github.com/deiteris/voice-changer

[brittle wing]

Hello. I tried to paste the song audio that I want the Ai voice to sing. When I pasted it, it didn’t show that I did it and it’s completely blank. So, how do I fix this issue?

[unique kiln]

How can I fix that my program does not see my video card?

[violet heron]

What’s your card

[proven hill]

i dont understand

[brittle wing]

Like I use Manigo RVC and the mp3 file link or whatever didn’t showed up when I tried to add the audio for my Ai cover

[proven hill]

try with this https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

i know mangio is having problems lately

[low shard]

mangio is outdated since a year

[brittle wing]

Oh. That explains why that RVC is outdated. It probably has no updates or something

[wide ingot]

On top of that, Mangio CREPE extraction is broken on AMD GPUs.

[unique kiln]

NVIDIA Quadro K5100M

[proven hill]

mangio crepe is a CPU thing not related to GPU

[waxen oar]

how do i make my own model

[violet heron]

-docs

[azure marsh]

Suggestions for @waxen oar

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[waxen oar]

ty

[crimson wraith]

When applying inference to my own singing voice, is it better to apply pitch correction before, or after the inference is done?

[open stag]

• colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• EasyGUI, by rejects Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• 🆕 UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[next thunder]

whos gaytt am i eating today

[glad zealot]

same as in colab

its the same as how to do the theme also

[unreal lichen]

hello i need rvc colab site

[violet heron]

-colab

[azure marsh]

Suggestions for @unreal lichen

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• EasyGUI, by rejects Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• 🆕 UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[unreal lichen]

tnx u so much

[violet heron]

Easy GUI is broken rn

If you need to train use RVC Disconnected

If you need inference HuggingFace Spaces are recommended over colabs

[hard latch]

how do i train my voices

[knotty moth]

not sure but it may be older than GTX 10-series so it is unsupported

[violet heron]

-docs

[azure marsh]

Suggestions for @hard latch

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[grizzled badge]

hey, if i wanted to train an ai voice where would I start and what would i do? All I need is a pth and a index to use it.

[astral jungle]

First prepare the dataset for training, then enter the disconnected RVC and start training

[rare goblet]

Ayo? @astral jungle level 18 !!!

[twin scroll]

i just got my desktop buddy ai vokce all set up and an hour later i see my my brother playing a distorted version of his voice as the ai's voice.. i abandoned the project 3 months later..

[wispy rampart]

hiiii, what should I do if my voices are coming off a bit fake and robotic in RVC GUI?

[knotty moth]

-gui

[azure marsh]

https://cdn.discordapp.com/attachments/1122285248844144733/1203460490475343953/caption.gif?ex=65d12cec&is=65beb7ec&hm=bd2fb8d010006dd7c6e3c1c67d3ae846fd1478e1a3124c544c31b43086fe54aa&

[knotty moth]

it's too old, you should get better one https://docs.aihub.wtf/rvc/local/mainline/

[wispy rampart]

ooh

didnt know

ty

[verbal egret]

1. is there a updated version of RVC? Im using v.1.5.3.15
2. why does my voice sound so staticy when i play games and talk

[misty elk]

-rvc

[azure marsh]

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[misty elk]

-huggingface

Fuck

Thought that would work

-hf

[azure marsh]

<:huggingface:1179800228946268270> Hugginface Spaces

• 🆕 UVR5 UI, by Eddy and Ilaria Huggingface Spaces
• Ilaria RVC Zero, by Ilaria Huggingface Spaces
• RVC⚡ZERO, by r3gm Huggingface Spaces
• Applio, by IA Hispano Huggingface Spaces
• AICoverGen, by r3gm Huggingface Spaces
• Advanced RVC Inference, by r3gm Huggingface Spaces
• RVC v2 Huggingface version, by Clebersla Huggingface Spaces

More commands: -audio, -colabs, -kaggle, -local, -overtraining, -realtime, -rvc, -help

[misty elk]

-kaggle

[azure marsh]

Kaggle Notebooks

• 🆕 UVR5 NO UI by Eddy
• How to use RVC Mainline Kaggle by Cauthess
• 🍏 Applio RVC by IA Hispano
• ✨ RVC Mainline by Hina

More commands: -colabs, -hf, -local, -rvc, -help

[misty elk]

Can you make models on kaggle on like mobile

Cuz like rvc disconnected takes too long

[tame mica]

no, it will explod

[crimson wraith]

What kind of google drive access is necessary to run RVC Mainline ?

[thin stump]

you might have not noticed the new folder that popped up in your drive called training if it's cluttered

use the kaggle version anyways

[crimson wraith]

I noticed but it looked like it requires a lot of access to google drive. Like the whole thing.

[thin stump]

there's no drive. Imjoy GUI is best friend and that's all you need

[tame mica]

shit wont upload to imjoy

[thin stump]

Might want to check your ngrok auth again because it works on my end 🤔. Last update was 19 days ago too

hold on

[thin stump]

EZ

I can probably make a video "Kaggle mainline -5 minutes or less" since I can blitz through it without looking at the guide now

[shadow plover]

when i open the start http it doesnt open and goes to cmd

[rich wagon]

guys why i cant listen my voice

[shadow plover]

wait do i need to download python

[thin stump]

#🔍│help-w-okada

[thin stump]

there's absolutely no context on what you're doing so we can't help you

[tame mica]

nevermind it works now ... for some reason

[brave garnet]

Voicechanger Settings (Okada)

⠀

Settings for AMD GPUs

Don't forget that your models needs to be converted in ONNX!

F0 Det.: rmvpe_onnx (suggested for all series)

7xxx XT cards: 112-128 chunk | +16384 extra
6xxx XT cards: 128-192 chunk | +16384 extra
5xxx XT cards: 192-256 chunk | +8192 extra

RX 580: 192-256 chunk | +8192 extra
RX 570: 192-256 chunk | +8192 extra
RX 560: 256-384 chunk | +8192 extra

Advanced Settings

Protocol : Sio or Rest
Crossfade: 4096 start 0.2 end 0.8
Trancate: 300
Silencefront: Off
Protect: 0.5
RVC Quality: Low

⠀

Made with love by Ilaria 💖

[errant onyx]

Im trying to train a model but anytime i try to process my thing the logs stay blank

does anyone know whats going on?

[astral jungle]

What do you train the model with?

[errant onyx]

i would send the audio but i dont have perms to do it so ill just specify my settings

the sound is around 3 minutes long and is called RRCoachVoice. The path i have is Z:\RVC\Mangio-RVC-v23.7.0\mitm\ and the audio is alone in that folder. the experiment name is RRCoach, the target sample rate is 40k, guidance is on, Version 2, with the cpu process number is 10, and the ID is 0

[thin stump]

-local

[azure marsh]

Suggestions for @errant onyx

🖥️ Local stuff

• 🍏 Applio, by IA Hispano GitHub
• Mangio-RVC-Fork, by Mangio621 Huggingface
• RVC Studio, by SayanoAI Huggingface
• AICoverGen, by SociallyIneptWeeb GitHub
• Replay, by Replay Team Website
• Original RVC, by the RVC-Project team GitHub
• GPT-SoVITS, by RVC-Boss GitHub

Credits to Faze Masta and Antasma for compiling these links.

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

[thin stump]

Don't use the mangio fork anymore. 1 year anniversary since it's been discontinued by the way 🎉

[errant onyx]

what is this

[rare goblet]

Ayo? @errant onyx level 1 !!!

[thin stump]

You're using local right?

[errant onyx]

yes

[thin stump]

so those are the updated links

just don't pick mangio rvc anymore since it's been abandoned

[errant onyx]

hm okay

[thin stump]

actually

-rvc

[azure marsh]

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[misty elk]

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• EasyGUI, by rejects Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• 🆕 UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[thin stump]

click on the aihub docs. There's three options you can try

Mainline, Applio, or Ilaria rvc

[misty elk]

Man i wish i could run mainline on mobile 😔

[thin stump]

you could it's just glitchy and closes your browser if you misclick something

[misty elk]

Wait lemme see if i could open mainline without runtime disconnecting

[errant onyx]

im trying to use The Mangio-RVC-Fork. should i try something else im getting a bit tangled

[thin stump]

I like Mainline a lot but im already biased

[misty elk]

How do you download titan on mainline

[thin stump]

!wget

[misty elk]

I meant on colab

[thin stump]

https://rentry.co/RVC-Mainline-Colab

they share the same steps except kaggle is more GUI oriented and colab is all about google drive

[sudden violet]

it sounds like the voices are speaking a different language

[atomic bramble]

kendi ses modelimi oluşturamadım yardımcı olabilecek var mı

[subtle phoenix]

does this affect the RVC v2 Disconnected colab? the one mentioned by the bot or is it fine?

[thin stump]

RVC disconnected has no web UI and is the simplest one. It still has the bugs I mentioned in the mainline colab guide

Use kaggle if you want to move away from colab since I provided plenty of images

-kaggle

[azure marsh]

Kaggle Notebooks

• 🆕 UVR5 NO UI by Eddy
• How to use RVC Mainline Kaggle by Cauthess
• 🍏 Applio RVC by IA Hispano
• ✨ RVC Mainline by Hina

More commands: -colabs, -hf, -local, -rvc, -help

[low shard]

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• EasyGUI, by rejects Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• 🆕 UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[brave garnet]

RVC Colabs and Spaces

⠀

Google Colabs

⠀
AICoverGen-WebUI
Useful for making quick covers, by Hina.

AICoverGen-NoWebUI
Useful for making covers, doesn't include a UI, by Ardha, by Eddy, Hina and Gdr.

RVC Disconnected
To train new voice models, by Kit Lemonfoot.

EasyGUI
The OG interface, by Rejects.
⠀

Made with love by Ilaria 💖

[brave garnet]

Voicechanger Download (Okada)

⠀
Download for Nvidia GPUs
Version 18a cuda

Download for AMD GPUs
Version 18a directml

Download for Intel GPUs
Version 18a directml

Download for Mac
Version 17b Mac
⠀

Made with love by Ilaria 💖

RVC Colabs and Spaces

⠀

Google Colabs

⠀
AICoverGen-WebUI
Useful for making quick covers, by Hina.

AICoverGen-NoWebUI
Useful for making covers, doesn't include a UI, by Ardha, by Eddy, Hina and Gdr.

RVC Disconnected
To train new voice models, by Kit Lemonfoot.

EasyGUI
The OG interface, by Rejects.
⠀

Made with love by Ilaria 💖

[scarlet sonnet]

i converted something and played it. but now i can't find it again?
it's not in the output folder

[cosmic epoch]

can anyone give me a link to a colab where i can train models with gradio?

[low shard]

https://docs.aihub.wtf

check the docs

use applio for training models with ui

else rvcdisconnected

[zealous shuttle]

can anyone explain what this mean?

[wispy rampart]

is rvc good for changing non singing voices as well?

[rare goblet]

Ayo? @wispy rampart level 1 !!!

[wispy rampart]

i always get more artifacts when changing speech than singings but im not sure if that's obviously normal or if I am using the wrong software

[astral jungle]

Try checking to see if there are .index and .pth files

[analog obsidian]

Singing models arent the best inferencing speech, use an speech model instead

[wispy rampart]

oh

okies

tyvm

[proven hill]

depends on the model

[zealous shuttle]

on drive?

[unborn owl]

ModuleNotFoundError: No module named 'ffmpeg'
idk why this happened in EasyGUI

[rare goblet]

Ayo? @unborn owl level 2 !!!

[low shard]

#📰│dev-updates message

easygui google colab is broken!

do you need training (make models) or inference (use models) ?

[unborn owl]

i was just about to make silly audio with it

i didn't know it was broken

[balmy reef]

I watched this video the other day, maybe it'll help :)) https://youtu.be/tnfqIQ11Qek

[balmy reef]

Yeah it's showing that error for a few days now.

[proven hill]

https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

[balmy reef]

If you upload a model, does it save it forever and can others access it?

[proven hill]

it will be deleted after a bit

[balmy reef]

Thanx

[worthy lantern]

wth

https://2e0dc603e3165b9356.gradio.live/

works this website?

[rare goblet]

Ayo? @brittle wing level 2 !!!

[proven hill]

https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

[slate axle]

nah way too much work for me 💀

[shadow plover]

is pytorch neccesary?

[rare goblet]

Ayo? @shadow plover level 1 !!!

[violet heron]

No

It comes with it

Same with python

[shadow plover]

tyyy

which one should i do

[brittle wing]

weights ai

[shadow plover]

i have a nvidia graphics card

[violet heron]

Which one

[shadow plover]

4060

[shadow plover]

[proven hill]

[brittle wing]

simple

[proven hill]

stupid + paid

[brittle wing]

just message the Weights bot lmao

no

😭

[slate axle]

lol

[slate axle]

nah dude im busy rn

[proven hill]

it will be very soon like training was

[brittle wing]

just hope it dosent end up like Kits.ai

stuck behind a paywall now

[proven hill]

again, it will happen

[slate axle]

probs lol 💀

[proven hill]

do you think they can keep up for free?

[brittle wing]

no

cuz they gotta have money to run servers

[slate axle]

true dat

[shadow plover]

which one do i download for the nvidia graphics card

[violet heron]

Which NVidia card do you have

[shadow plover]

4060

4060 ti

[violet heron]

onnx cuda

[slate axle]

idk man im watchin anime rn lmao

[violet heron]

highest number that ends in a

[proven hill]

mb

[shadow plover]

3 gbs

[slate axle]

bro idk im tryna grind to plat in valorant rn

[rare goblet]

Ayo? @slate axle level 3 !!!

[slate axle]

bro tryna 1v1 me? im down bad 💀

[shadow plover]

im bronze 1

[slate axle]

💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀

[shadow plover]

ok bro im new to val

id say im pretty good

[slate axle]

lmao gl man valorant is hard af

[shadow plover]

killjoy is the worst character

[slate axle]

nah man cypher is where its at

[tame mica]

bro has the tiktok emojis in discord

[shadow plover]

[slate axle]

ive never been on it wtf are those

[shadow plover]

how do i create a new folder

[brittle wing]

right click

hover on new >

and click folder

thats the basics lmfao

[slate axle]

fujini u using light mode? 💀

[brittle wing]

wym

[slate axle]

💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀💀

[brittle wing]

okay

anyways

[slate axle]

💀💀💀

[brittle wing]

idk if its different on win 11

i use win 10 traditionally

[shadow plover]

i didnt have a choice

[rare goblet]

Ayo? @shadow plover level 2 !!!

[slate axle]

bro think he arcane from league 💀

[shadow plover]

i dont play that

[slate axle]

lmao

[slate axle]

what game

[shadow plover]

league

i know the arcane sheriff from valorant tho

[shadow plover]

do i extract or do i just run it

[brittle wing]

extract what

[shadow plover]

start http

[brittle wing]

run what

dog im new here

i just help with basics and some advanced stuff a bit

[shadow plover]

damn

[brittle wing]

theres new ppl joining every minute what

[low shard]

Rejekts, the creator of the Google colab, isn't active since may, Soo for now its better you try other google colabs

[shadow plover]

nick i need u

[slate axle]

league is kinda mid tbh 💀

[shadow plover]

do i extract file

[slate axle]

idk man im busy rn

[low shard]

You can try applio or RVC disconnected in our docs https://docs.aihub.wtf/

[shadow plover]

stop edging me

[slate axle]

just google it dude im watching anime rn

[violet heron]

what do you think

-docs

[azure marsh]

Suggestions for @brittle wing

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• How to use RVC Mainline Colab by Cauthess
• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[shadow plover]

that i just run it

[violet heron]

If the program depends on other files it wouldn’t be able to use them when zipped

So you would need to extract it

[shadow plover]

tyyyyy

[slate axle]

prob lol idk im watchin attack on titan rn

[low shard]

Well the docs could help you as it had the guides for them

[shadow plover]

theres a first time to everything

[low shard]

You're welcome

[shadow plover]

when i open start http my windows anti virus pops up i click run anyway and it disappears

[slate axle]

bruh 💀💀💀

[violet heron]

Is there any spaces or special characters in your directories?

[slate axle]

idk man sounds like a skill issue tbh

[shadow plover]

wait is it start http or start https

[violet heron]

http

[shadow plover]

is the system 32 supposed to pop up

[violet heron]

Yes

[shadow plover]

aight

its not doing anything tho do i close it

[violet heron]

No

Wait a minute and it’ll start downloading the required files then it will start

[shadow plover]

oh nvm

it started doing stuff

[slate axle]

sheesh good stuff my guy

[shadow plover]

i can already smell the black mails

its done

[shadow plover]

do i close cmd

[rare goblet]

Ayo? @shadow plover level 3 !!!

[violet heron]

No

Unless you want to stop using the voice changer

[shadow plover]

aight

[pearl bramble]

i need some help on this colabs https://colab.research.google.com/drive/1mHKTGH5e3SAyDSBss1KtiYRbDdQzwSMs?authuser=1#scrollTo=8RwuGj0LmSB_

[slate axle]

nah man im busy rn

[pearl bramble]

as my import dont work whenever i click it

[proven hill]

if its only for inference

btw

you can use this https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit

[slate axle]

fr fr 💀

[pearl bramble]

alright

[shadow plover]

everything is working but its not working on discord

[violet heron]

Maybe read the guide

https://cdn.discordapp.com/emojis/1256321799424311410.webp?size=48

[shadow plover]

where is that

[violet heron]

-realtime

[azure marsh]

Tag suggestion for @shadow plover

💻 (Local) Realtime Voice Changer

• Guide for W-Okada's RealTimeVoiceChangerClient
• How to use real time conversion with RVC (voice changer)
• RVC TTS

[violet heron]

first link

[shadow plover]

uhhh

[proven hill]

read

[shadow plover]

ik

so how do i go to client mode

[violet heron]

It’s on client by default

[final sentinel]

Can you tell me how to set up UVR5 to take only the backing vocals from a song?

In website uvr5 ui

[proven hill]

huggingface?

[final sentinel]

Yes

[proven hill]

vr architecture

karaoke hp5

[violet heron]

wtf are you using harvest

Switch to RMVPE

[ionic geyser]

doesnt work with anything i tried all

[violet heron]

Try importing a different model and using it

If try importing without a space

[open stag]

• colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot Google Colab
• RVC Mainline, by Hina Google Colab
• AICoverGen-WebUI, modded by Hina Google Colab
• AICoverGen-NoWebUI [English], by Ardha, fixed by Eddy, Hina and Gdr Google Colab
• AICoverGen-NoWebUI [Spanish], credits to Eddy, Hina and Gdr for translating and fixing Google Colab
• Ilaria RVC, by thestingerx Google Colab
• UVR5 NO UI for Google Colab, by Eddy Google Colab
• UVR5 UI, by Eddy Google Colab

More commands: -audio, -hf, -local, -kaggle, -overtraining, -realtime, -rvc, -help

[rare goblet]

Ayo? @open stag level 2 !!!

[junior canopy]

download

[violet heron]

Download what

[junior canopy]

im trying to find the rvc download that uses python

[violet heron]

For real-time or covers

[junior canopy]

i had it on my laptop but i cant find the download link

real time

[violet heron]

And what’s your GPU

[junior canopy]

nidiea

nivida