#✨│ai-help
1 messages · Page 103 of 1
Spectral Noise adv+extr
No there was no room reverb in the dataset so I don't have to dereverb it
But u still left noise in the
E.g = exampli gratia
And leaving noise in the dataset could cause mode collapse
Idk what that is
If u can go into rx and zoom in on a small portion of the dataset
The dataset has little noise but I removed it with RX spectral noise, also no room reverb at all so I didn't need to use dereverb, then I truncated silence.
I did everything right to my knowledge
Spectrogram ofc
bump
Ayo? @fading drum level 2 !!! 
U need to click on train index
What problem that I particularly find here
I wanna check for artifacts, noise and other stuff
Hello. I'm using Applio for RVC model training. I get the error "Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass" when training. How to fix it?
It isn't a problem ignore it
Should I wait when the training gets finished?
Yes
U should let it train normally
And just let it be
Once it's done u can go ahead and test out the model
Yeah there's noise
Is it the middle? Also are they reverb trails?
RVC doesn't care if it's apperant
GANs can pick up the slightest of noise
And then can go crazy over it
That's why it's collapsing
And yes
U need to decide again
what do you think of this?
Yeah it's the middle part, I don't think those are trails, might be a prolonged note that just went lower and lower
Ayo? @red kayak level 73 !!! 
Muted
Yeah mode collapse means that the model hasn't read thru the whole dataset
I thought it's a thing that randomly happened
I trained an actual vocal stems before and it has mode collapse regardless
95 epochs, sounds pretty great
I'm trying to create Carl Casagrande AI voice model
No copy right audio please
Ty, just had to confirm something. The blue looking trails are basically dead audio
It's usually noise, there's no such thing as dead audio
Gotcha
No it isn't
U should instead do labels and export them
Wdym by labels
-audio
Suggestion for @lament eagle
- Perfecting Audio Isolation on Low-End Rigs: A Practical Guide, by Litsa The Dancer and Faze Masta Google Docs
- Gathering and Isolating Audio, by SCRFilms Google Docs
- Creating Datasets for RVC using iZotope RX, by LUSBERT
Rentry - Vocal Mixing Tutorial, by Roomie YouTube
Audio Separation/Isolation
Look at lusberts guide and scroll down
I just discovered you can drag this. It's going to make spectral editing a lot easier instead of guessing what's noise you should keep
U can see if it's noise or not regardless
Do labels anyway
4070 laptop
if you have finished training yes you can. but if you want to train more. then Don't do this
it works until 7 epochs then it breaks
hello, can anyone maybe help me find voice model that can work with polish language in decent way?
I think every model with good training will work
you can find a lot of decent models on weights.gg
ok so im using replay to do ai voice stuff, i HAVE .pth files, but replay says "there is no .pth file"
actually you don't have it
Error creating song: No valid .pth files found for mi-test
bu-bu-bu-but there iiisss...????
G_21060 or D_21060 is not correct files
what.. would be?
(i ctrl z'ed alot)
I think it's not a RVC model
RuntimeError: CUDA error: an illegal instruction was encountered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
does anyone know what this means
I think you forget to export actual pth
find mi-test.pth
there is not one, do you know why?
is there like a setting i need to turn on?
no. idk why this happened. czz I think I automatically save the final pth file
there is only g and d, could i rename it?
That will not work
okay .. go to weights
how do you do that
okay .. go to weights
GASSPP
now i just gotta compress 3 gig to under 2 lol
select pth in replay and index too
everything is useless except .pth and .index [if you don't want to train more]
yes .. If you don't want to train your model more. only .pth and .index [are] important
ok, thank you very much for all the help
remember don't delete .index from logs
yw : )
so, final thing to clarify, only the mi-test.pth, and the "trained_IVF1614_Flat_nprobe_1_mi-test_v2.index" is all i need
Ayo? @fading drum level 3 !!! 
yah.. but you have other models too
ok ok, core memory made
ye but in the folder im zipping
so remember these models also have index files and those index are also important
ye
just zip these files
else nothing
ok
the google collab keep saying access denied when i try to downloand my model
-colab
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
-rvc
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
Model training with Mainline RVC
Link: Rentry
credits: Raven (ravencutie21)
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How To Make an AI Cover With Ilaria RVC
Link: Rentry
Credits: 👽 Julia (ailen2091)
Hola!!! Buenas.
Tengo una pregunta, si termine de entrenar mis modelos, pero olvide exportarlos y la página se le acabó el tiempo de espera, puedo hacer algo o le toca entrenarlos otra vez?
[Voice Changer] Pipeline is not initialized.
[Voice Changer] Waiting generate pipeline...
help
what if i want to use a model found on applio in a google colab file? how can i do it (if it's even possible)?
RVC Guides (How to Make AI Cover)
Translation by country
I get the traceback when trying to convert a vocal using RVC GUI
I used it for a while and didn’t have any problems before. Maybe I'm doing something wrong?
might be that the fork is breaking down, cause it's outdated
Explain properly plz.
Tendrás que entrenarlos otra vez.
Porque una vez que se te acaba el tiempo de colab y se te desconecta, pierdes todos los archivos
if i found a rvc model on applio that i want to use, do i necessarily need to download applio and use it there or is there a way to import it in huggingface so as to use it on google colaboratory scripts
In that case, download the model. If it's an zip file, check the file so it has the correct model structure for usage
(It must have a folder with the model name and in that folder there should be the .index and .pth)
Una lastima, gracias por la info
alright thx
Please, use https://discord.com/channels/1159260121998827560/1159595680659157132 channel.
heyo
do you guys record and put autotune on vocals, then isolate them and swap voices, or do it the other way around?
-guide
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
Model training with Mainline RVC
Link: Rentry
credits: Raven (ravencutie21)
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How To Make an AI Cover With Ilaria RVC
Link: Rentry
Credits: 👽 Julia (ailen2091)
-rvc
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
Model training with Mainline RVC
Link: Rentry
credits: Raven (ravencutie21)
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How To Make an AI Cover With Ilaria RVC
Link: Rentry
Credits: 👽 Julia (ailen2091)
evry time i get this error, can you help me?
You would probably have to go to windows control panel and set that audio as the main
what do you mean?
Type in the windows search bar and look up control panel
okey, en than?
and then you go to audio and check what your main input is
that are my speakers
Go to microphone
yes
What's your default input
Hmm that's strange if it's not working try using rmvpe instead of crepe
what is the difrence?
And turn the sample length up to 1.5 or 2 that will reduce the voice cutting out
It's the newer better version than crepe
anyone else getting this message when downloading models in easyGUI? the permission is already set to ' Anyone with the link' - Access denied with the following error:
"Cannot retrieve the public link of the file. You may need to change
the permission to 'Anyone with the link', or have had many accesses.
You may still be able to access the file from the browser:
https://drive.google.com/uc?id=https: "
okey, i try
Alright
downloading from Huggingface doesnt work either
You should also probably turn off harvest rate since you aren't using harvest
okey wait, i try
C:\Windows\System32>runtime\python.exe infer-web.py --pycmd runtime\python.exe --port 7897
The system can not find the path specified. Please help
The shitty problem is that it worked before and now it doesn't
i think thats an outdated colab
What happens if I use 1 hop length?
with rvmpe? absolutely nothing
with mangio-crepe? it wont sound great, at all
why is my tb graph not steep?
are ya using ov2? that could be why
does it take so much time
with mangio-crepe
uyes why
yeah thats why
honestly dunno why
probably one of ov2's quirks
thanks
oh okay but it isnt wrong?
...you're welcome
i mean... it ain't good, but if the model sounds good after the training...
what should ive used instead
probably the normal pretrains
Ayo? @dire vapor level 2 !!! 
I think it would be better if I reduce the hop length.
yeah probably
it aint a good idea to use 1 hop length
sorry
thanks
yw
what is the eta for a 10 minute database, using colab and ov2
probably 2-4 hours, but id recommend stopping after 1 or 2 hours and exporting model
then continuing later
im so dumb i was in timeseries and not in scalar
ah
is there a place that i could train AI voices
-colab
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
RVC Disconnected ;)
i keep getting this
Cannot connect to GPU backend
You cannot currently connect to a GPU due to usage limits in Colab. Learn more
To get more access to GPUs, consider purchasing Colab compute units with Pay As You Go.
i was trying to use easyGUI awhile ago but
i keep getting problems
ran out of gpu, wait 24 hours
https://aihubdocs.github.io/en/rvc/resources/inference-settings/#pitch-extraction-algorithm check this out
Last update: Feb 25, 2024
RMVPE should be your go-to. Use Mangio-Crepe/Crepe if you know what you are doing
how to make cover?
Ayo? @modest sparrow level 1 !!! 
Last update: Mar 9, 2024
thank you
you're welcome :)
any video tutorials on how to make models? i want to make a voice model out of scratch
i wanna make a voice model of "noongar" https://www.youtube.com/watch?v=rt2LHE3oeks&ab_channel=dymes25
Last update: Mar 9, 2024
check our docs
RVC Guides (How to Make AI Cover)
Translation by country
shts too confusing, can i pay someone to make this model for me? 😭
what does INDEX file do in RVC? its functions and stuff
helps with accent stuff
bro everything was working and then suddenly even stable diffusion stopped working 😭
this is rvc
which colab has the TTS(text to speech)
Ayo? @young zenith level 1 !!! 
-rvc
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
Model training with Mainline RVC
Link: Rentry
credits: Raven (ravencutie21)
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How To Make an AI Cover With Ilaria RVC
Link: Rentry
Credits: 👽 Julia (ailen2091)
why my model doesnt work?
-colab
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
I think you should consult that on https://discord.com/channels/1159260121998827560/1185670409270607993
🐢
well it was rvc but I've got that fixed, thanks
Oh, i got kinda confused since you said "Stable diffusion stopped working"
My wrong
how do i use a model
I'm still getting this error
I reinstalled RVC GUI, no result
Traceback (most recent call last):
File "C:\Users\tmand\Desktop\RVC-GUI-pkg\rvcgui.py", line 129, in vc_single
audio_opt = vc.pipeline(
File "C:\Users\tmand\Desktop\RVC-GUI-pkg\vc_infer_pipeline.py", line 340, in pipeline
self.vc(
File "C:\Users\tmand\Desktop\RVC-GUI-pkg\vc_infer_pipeline.py", line 243, in vc
(net_g.infer(feats, p_len, pitch, pitchf, sid)[0][0, 0])
RuntimeError: CUDA error: the launch timed out and was terminated
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
what to do with that?
Ayo? @iron cobalt level 3 !!! 
Anyone there who had the same error when trying to make an AI cover using RVC-GUI? If yes, can you explain how you fixed this error?
im needing help with setting up voice mod for my discord and willing to pay someone to help me
when will it be fixed
colab or locally?
colab
How can I help you?
VoiceChangerParams.init() missing 1 required positional argument: 'whisper_tiny'
I get this error and I don't know what to do
- Use this colab https://colab.research.google.com/drive/1X8YR4Ruv7zzY8YAMPTfC7hkxqT_d4Q5d
- Here RVC A For Voice Changer -> #🔍│help-w-okada
okey
Ayo? @hasty frost level 1 !!! 
any K-pop fans here pls tell me we can still mention/credit the model makers' username/nickname on YouTube even if we no longer in KAU?✌️
bro, use your own discretion since you keep spamming it. You chose the most miserable fandom to piss off regardless of what you do
No I'm not spamming yesterday I just thought my message yesterday didn't send so that's why that my previous message accidentally sends again and again but I'll be careful next time✌️
Ayo? @subtle cedar level 2 !!! 
RVC Guides (How to Make AI Cover)
Translation by country
What are the best models on UVR?
oh ok but still crediting is the least anyone could do unless it's a meme-y video
Okay yeah thanks
/me message:
I'm also getting this error, I've tried everything I can think of and nothing is fixing it. Any ideas? I've been using EasyGUI on and off for months now, I'd consider myself decently familiar with it. March 1 is the last time I successfully used EasyGUI, now I just get this error. I don't know how it's not finding the link. It's already set to public, as all my models are.
Anyone know what a "Supertrain" is?
Do you use UVR5 for the song you are trying to clone into as well? or just for the data set
can someone give me like a really basic runthrough of how to use the models
@steel forge
Ayo? @spring orchid level 4 !!! 
-colab
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
@livid galleon
Should this go up or down? does this matter?
no
Alright
how do you get model maker ?
only mel,kl, g and d loss matter
You have to submit a model in #1159514067187277865
I know. but you don't know about grad_norm_g
All these seem fine there isnt any collapsing
Did you mean how did you get model maker?
yes ..
I installed tensorboard 2 days ago
Just submitted a model and that was good enough to get the role
yes...
Suggestion for @rugged stirrup
English guide
Link: https://rentry.org/fdg_guide_newer
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
ohh.. I think we have to grab a lot of information to get this
Sry I was just wondering if that was good the tensorboard graph
I think it's kinda easy
Oh
training is going well . don't worry.
@novel plume if you are testing commands go to #🤖│bots
Oh ok
hello there! using the cover gen collab, everything's done and finished. hence, i was given a waveform audio file format but how do i access it? sorry, it's my first time dealing with AI models so i'm a little oblivious atm lol.
It should be there in your files. I was intimidated by seeing how big wavs can be at first but now I use that format for everything ai-related
RVC Guides (How to Make AI Cover)
Translation by country
Oh sorry, I meant in your folders to the left of the colab
thanks for your response! this is the first time i've used this specific ai model collab, but i've done some other really simple things and typically i get a pop up link within the file if that makes sense but this time the text isn't coming up as a url so i am unable to download it to my files.
And make sure google chrome or firefox doesn't block consecutive downloads
ohhh that makes a lot more sense! thank you so much! sorry for the follow up question but, how do i find it within the files cause for me there's a lot of stuff showing up there?
Ayo? @shy siren level 1 !!! 
I think it shows up when the function is running if that makes sense. You didn't click out of the page right?
Is there a folder called cover gen or something. If not you have to redo it. Idk if it's in your google drive either
nope, i didn't. i think i might need to redo the process just to start fresh now that i know what to do, thanks to your help, and hopefully i'll find that folder this time since i don't see it anymore😭 thank you very very much for your help! means a lot 😄
yw, it happens. I lost my ai model when google kicked me out
Hello guys can someone help me , im new in this type, i finded an model voice ai on hugging face:https://huggingface.co/sail-rvc/CapoPlazaV3 , how can i use it on googlecollab rvc? THX for help
copy the link for both of them
or download it and put it into a folder
the index is the accent btw. important stuff
-colab
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
Anyone around to help?
`ModuleNotFoundError Traceback (most recent call last)
<ipython-input-2-236cfd0b6a58> in <cell line: 4>()
2 #@markdown Link the URL path to the model (Mega, Drive, etc.) and start the code
3
----> 4 from mega import Mega
5 import os
6 import shutil
ModuleNotFoundError: No module named 'mega'
NOTE: If your import is failing due to a missing package, you can
manually install dependencies using either !pip or !apt.
To view examples of installing some common dependencies, click the
"Open Examples" button below.`
how can i solve that?
It usually appears during a train
What’s the best advance option for Ilaria RVC?
I tried googling this traceback error line 129, but no fix I have found
just use a newer local fork, might fix it
check https://aihubdocs.github.io/en if ya want download links and guides for those
Last update: Mar 10, 2024
old colab, use a newer one, check https://aihubdocs.github.io/en ;)
Last update: Mar 10, 2024
RVC Guides (How to Make AI Cover)
Translation by country
i need help when i put my audio and my model it still says to put those in
What tool you using
rvc gui
U mean the local fork?
i wish i could sent what i mean
I think thats outdated, check our docs https://aihubdocs.github.io/en
Last update: Mar 10, 2024
-colabs
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
Guys
What colab is the best for epoch training?
for rvc ai voice model training, i personally use rvcdisconnected, you can find more about it in our docs https://aihubdocs.github.io/en
Last update: Mar 10, 2024
hey I downloaded the program locally and there is no model visible in the web ui, can someone help me in which folder do I need to put the downloaded model to use it on web ui
if i have a shit mic will rvc still sound good"?
First of all what fork r u usin
Shit mic will sound shit with rvc
I can't pick up where I left off training a model. Is there a way to fix that?
RVC Guides (How to Make AI Cover)
Translation by country
hello guys
can rvc and openvoice work together combined? and if such thing is possible, is there a written guide ot going about this? thanks!
how long does it take to train 300 epochs with 8gb vram on a nvidia rtx gpu?
i am on rtx 3060 12gigs and it takes around 40 seconds an epoch
but that's leaving it run at night
damn bruh
is 200 ten second clips too much for training
Hi I am searching for someone who has a lot of knowledge about AI Voices and that can help me to convert some Scripts in AI Voice. Also willing to pay
Please DM me
how do i fix this error after trying to use ilaria rvc
Ayo? @frosty lichen level 4 !!! 
did you install first
yes
hm
Just restart the download again
ill try that
hay nuevo metodo de cpu?
por que no tengo cpu, y vi varios videos que se podia, o se puede, no se si ahora no
How in the actual fuck am I trainning an epoch every 4 seconds
Yes I do have an rtx 4050 laptop but this makes me concerned
Doesn't matter
mk
i must be doing something totally wrong then
I had normal trainning times, then it sparked into this state of very wonky and short trainning time
maybe because I have done too many runs with the same dataset trying to improve upon my model
im trying to run it through collab and i got it to run but it closes after like 5 minutes
what do i do
How do I do text to speech talking and songs
-colabs
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
Error: Unable to connect to Rich Presence. Could not find Discord installed and running on this machine.
Theme Loaded: Applio
What to do?
I'm trying to make an AI cover using Applio but my PC completely freezes every time I'm trying to do it. The only thing I can do in this situation is forcefully turn off the PC. My GPU is Nvidia Geforce 940MX, CPU is Intel Core i3. My laptop is Acer E5-575G, RAM is 20 GB with 2 DDR4 (16 GB + 4 GB)
help i keep getting errors
Ayo? @regal plaza level 1 !!! 
Hey yall
Can i get some producer tag using A.I collab voice ?
Like dirty sexy sauce tag
yo y'all
Ayo? @viscid pecan level 1 !!! 
pls help me
It says my client is wrong
spent 6h today trying to figure out
how to wok this then i edited the code and got f****
what tool are u using?
send the link of which colab you are using
no not the gradio public link, i mean the google colab link where you runned the code
this one
I ruined the code
then i tryed again from scratch
and put in the link
that colab is outdated
kinda cus you prolly followed an old yt tut, that colab cant work anymore, use only the updated ones in our docs https://aihubdocs.github.io/en
Last update: Mar 10, 2024
!howtoask
How To Troubleshoot 
- 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.
- 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.
- 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.
yw
How long does it usually take to run dependencies with RVC? It feels like it has been at least 40 minutes. I don't know anything about it but I assumed the loading circle should stop when it's done with each dependency? I know it's still going because eventually it got to "Mount Drive" but it's been sitting on that one alone for at least 20 minutes. I'm using RVC Disconnected V2 from a Google Collab. Maybe im doing something wrong. I run a Geforce GTX 1650 with a i5, Windows 10. I'm not sure if there is any other details I need to show so ill show a screenshot as well. Edit: Nevermind it won't let me post screenshots?
Hello, I'm new here, nice to meet you all.
Question: Which open source can I use to convert a song's voice (for example Adele) to an AI voice with all the music and background?
RVC Guides (How to Make AI Cover)
Translation by country
Ok, thanks.
Ayo? @unreal rain level 1 !!! 
Update: I decided to refresh and run it again. This time it ran all of the dependencies in no time flat. Now there is a issue. It says my GPU was not detected. What do I do?
Ayo? @torn bluff level 1 !!! 
Google colab not working
I found out how to change my runtime but the only GPU option I have is T4 GPU. So what am I supposed to do?
Nevermind, it's saying my Collab GPU Limit is reached. So when can I run it again?
Is it possible to convert MP3 with Voice Changer Client like voice to voice?
12 hours, or use another google account. But the thing is you need that gpu to stop the training so it saves to your google drive https://aihubdocs.github.io/en/rvc/cloud/rvc-disconnected/#rvc-disconnected
@thin stump When will I know to stop the training?
Ohh, sorry my reading comprehension
I think you can view the gpu usage?
The problem im not sure where right now.
The screenshots used in the link you sent me is not what I am using
I am using something called "Google Collab", it's one called "RVC v2 Disconnected"
Ok, they will give you an amount of compute units by default as a free user. You should monitor that. I only have 89 because I bought some
under your T4 tab
The down arrow?
sure, you could click either one
View resources
Yup, you would wait until tomorrow to resume training and monitor that next time
Okay so I won't need to reset disk space or anything. Just wait about 12 hours?
Did you stop the training or is the tensorboard still going?
Nothing is running right now.
Ayo? @torn bluff level 2 !!! 
Great, you can resume tommorow as long as you did the step where you saved the preprocessed datset files to google drive https://aihubdocs.github.io/en/rvc/cloud/rvc-disconnected/#section-17
Oh then you have to restart. Because it didn't save 
Okay. So I should just wait for 12 hours then?
Could just use another google account and follow the steps of the guide. Monitor the compute units, stop the training and let it save, then transfer to another google account via google drive
Would show 0 compute units for free accounts no?
i'm not sure I can check
Nah it's fine. I have time. No rush. I wouldn't be training a voice if it was available online but this specific type of vocals can't be found unfortunately. It's samples of vocals from multitude of songs I snipped together in Audacity and used a program to remove instrumental, no reverb, echo etc. and now i want to use that WAV track I created to do a cover of another song.
AI cover to be more precise.
Thanks for your help.
You're right, man google is so greedy
yw
apparently free users have 12 GB of usage and I remember getting kicked out at around 140 epochs. Just my anecdote
Free users get kicked after account 2-3 hours
Around*
Yo, how long does an epoch take for you guys who train locally?
Training a new model and each epoch takes like 50 seconds, versus 19 on my old model
The only difference is my audio is separated into multiple files instead of one big one
I'm new how do I get an ai to sing a song or something?
it depends on dataset length and batch size, besides the gpu performance
doesn't really affect in training performance, but I'd suggest splitting the audio using Analyze -> Label Sounds, in Audacity to help getting rid of mode collapses
Oh I didn't know what mode collapses were. Huh. Learn something new everyday.
-Colab
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
@proper shale I’m sorry to ping you but I’ve been trying to figure colab out and I can’t
Can you help me?
⠀
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.
⠀
Anybody know how to fix this? Is Colab busted for voice conversions now? Any alternatives? Tried Chrome and Edge with different models I know to be working.
⠀
HuggingFace Spaces 🤗
⠀
Ilaria RVC
EasyGUI port with some improvements, by Ilaria.
RVC-HFv2
Applio port, by r3gm.
AICoverGen
AICoverGen port, by r3gm.
Advanced RVC Inference
Extended version of the GUI with advanced settings, r3gm.
⠀
thats a super super old colab
RVC Guides (How to Make AI Cover)
Translation by country
why can't i make voice model on easygui anymore?
outdated colab
When it comes to training, can you put multiple audio files for input in the folder or does it only accept one audio?
These Two Correct Options You Can Use 1 File But Minimum: 8 Minutes Or Multiple Files Minimum: 30 Seconds
(But that’s what I think + you need it to be clean and noise-free)
wrong information
the correct response is
technically:
- you can put 1 or different audio files in the training folder (usually its "dataset")
- the duration of one or more audios needs to be 1/2 seconds at least to be working.
pratically (suggestion):
- doesnt make any difference, for my personal comodity i usually use only one single audio but it wont affect the training.
- i suggest you using at least 5 minutes (total) of data for a good model, of coursethe audio needs to be clean (you can use different programs such as UVR5 and RX10 if youre advanced)
What Colab
akshually 🤓
Found out that more audios can cause training to slow down... a 30-40 min dataset separated in 1,2k files was considerably slower than a normal dataset. Even at 10 batch, every epoch took a minute and 30 seconds
⠀
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.
⠀
Full AI Voice Model Training Guide (Local)
Link: YouTube
Credits: Christopher Villanueva
Model training with Mainline RVC
Link: Rentry
credits: Raven (ravencutie21)
AICoverGen Colab Guide
Link: Google Docs
Credits: Eddy (Spanish Helper)
Create a model with RVC disconnected (colab)
Link: Google Docs
Credits: Angetyde
How To Make an AI Cover With Ilaria RVC
Link: Rentry
Credits: 👽 Julia (ailen2091)
logs
did you train the index?
and train index=
is that gonna take a while too
no, only 10-15 seconds
alright
how do i use the voice changer?
doesnt matter what the epoch count is
Ayo? @brittle wing level 1 !!! 
nope
do i do the preprocess again
Ayo? @hazy thunder level 2 !!! 
not needed
cause thats just gonna create a new log folder of my model
so i just put the location of my already trained model here?
then click train index
ye
g path and d path are the same thing
just like the example did
or is there something else i have to put
no, like the original
am i supposed to wait after clicking train index because there is nothing happening in the command prompt
wait a few sec
the output information box was doing the loading thing then stopped shortly after
check the log folder
nothing there
show me a screen of rvc
the gui?
yup
i noticed it creates
a mi-test folder
in logs
whenever i click train feature index
i got it to work
theres a trained index and an added one
How do I know if my pc is good enough to train epoch?
list me ur specs
RVC Guides (How to Make AI Cover)
Translation by country
@proper shale https://colab.research.google.com/drive/1Gj6UTf2gicndUW_tVheVhTXIIYpFTYc7?usp=sharing Ive been using this colab and every time I paste a model into the model link thingy I get an error. It’s been like this for a month.
Ayo? @feral saffron level 3 !!! 
hi i face a big problem any one can help?
if u got a good enogugh pc just use it locally
colab kinda sucks
That colab is outdated
Dont use it
Use the updated ones in our docs
Last update: Mar 10, 2024
!howtoask
How To Troubleshoot 
- 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.
- 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.
- 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.
Yeah but I don’t have a pc. That’s why I use colab
you do you ig
Can I use it on iPad?
Yes
Awesome! How do I do that tho?
Just follow the guides in the docs
There are some colabs you could try, for making ai covers i personally use ilaria rvc which is in the docs
Yw
one q, why Ilaria RVC no Google Drive mount?
thestingerx you should add Google Drive mount python script
What would you use the drive mount for ?
I want to save my voice models and output
in colab without T4
Is downloading them not an option ?
models import really quickly so I'm not sure it's even worth it
on audio set server, i not working even i set all input and output device
and what is S.R. ?
where is the pitch change setting
Ayo? @hazy thunder level 3 !!! 
im tryna turn a female voice into a man
for what, real time voice changer ?
If that's what you're using, it might be called "TUNE"
no
for the regular rvc gui
go-web.bat
i swear there was a setting for pitch before but now there isnt
Anyway to get .pth working on amdM
Without tinkering no
Maybe pytorch-directml could do something, but I think it's outdated and useless
Or if you happen to have a linux partition lying around...
Fair. I'm trying to get sillytavern 's rvc extension working and i am thinking it may be a bust due to no .pth support
I don't wanna have to get an Nvidia graphics card
....are there even gpus that fit on x8 pcie slots?
Yeah
And the one underneath is the same size right ?
It's just that it gets less lanes ?
As long as it fits it should work
I'm running my 6600xt at x8 3.0, and then another rx 580 at x4 3.0
from ubuntu
is ffmpeg installed ?
I just uploaded ffmpeg
seems like something went wrong with the versions of the ffmpeg python packages it uses
-colab
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
So how does fine tuning a model work?
I've so far relied on just using a model out of the box, but I think this one might not be plug and play
thank you problem solved
Are you using the hugging face space?
Like, are you doing it online or locally
Ohh
Does it happen with every models?
Its your own interaction right? Like u didnt clikc the button of someone else doing the / command right
Try with another model
Thats not applio bot
Thats model ai bot
Applio is /search
To know how to make kodels, check out our docs https://docs.aihub.wtf
Last update: Mar 10, 2024
Ayo? @brittle wing level 3 !!! 
Yw
Are there differences between models during training? To explain in terms of Colab, is there a difference between RVC Disconnected Colab and Applio Colab?
there shouldn't be any differences tbh
end result should be roughly the same
I'm sure I heard somewhere that the quality of Applio is generally a bit low, but I guess that's incorrect knowledge.
If you have a good dataset, all RVC forks will do a good job on the training
thx
RVC Guides (How to Make AI Cover)
Translation by country
if no work try again
Hi! Where should I go if I need a voice that is not on the server. Can I create one? I need the voice of a spider man, but from a Russian dub (in Russian) or is there someone who can make one?
Hey Guys i Want to make my own RVC model How I Can Do it ? | oh nothing im so F blind
- To find more models you can check on https://weights.gg or https://applio.org/models
- If you want to make one yourself, check out our docs https://docs.aihub.wtf/
- You can request a model in #1159289738314919936. Keep in mind, Paid Requests get responses quicker.
- Or, check #1191429836321849435
GUYS PLEASE HOW DO I USE RVC LIVE CHANGER ON TELEGERAM WITH THE VB CABLE IT ISNT WORKING ITS JUST Glitching
@jade mantle dude quit with the caps
are you okay?
😭
for realtime voice changer ask in #🔍│help-w-okada
Thank you:)
Ayo? @stiff cliff level 1 !!! 
yw! ^^
Don’t be rude
are there any RVC training colab that i could use right now?
yes. check the tutorials on our docs https://docs.aihub.wtf/
for training you have RVC Disconnected, Applio Colab and EasyGUI
I'll finish the EasyGUI guide at some point
And how hard is that to make my own one? How much time on average does it take for a newbee?
hmmm well the duration depends on a lot of factors
making my first model took me around 4 hours. but that includes the training step, which is just waiting
Okay, thank you. Because I was thinking between buying or making it. Because it will be cheaper to buy rather than making it yourself and spend a lot of time struggling.
guys why rvc voices doesn't work but beatrice work?
uhhh the voice models do work. maybe you are doing a step incorrectly
try to give much more context so we can help
So, I downloaded the voices from this discord server, installed them, but they don’t work. no one. but this voice on beatrice (default in voice changer) works as it should work
and default RVC voices doesn't work too
I’ll say right away that the version of voice changer for NVIDIA is what it should be, I downloaded everything as it should be
btw in console:
File "voice_changer\RVC\pipeline\PipelineGenerator.py", line 30, in createPipeline
File "voice_changer\RVC\embedder\EmbedderManager.py", line 25, in getEmbedder
File "voice_changer\RVC\embedder\EmbedderManager.py", line 48, in loadEmbedder
File "voice_changer\RVC\embedder\FairseqHubert.py", line 11, in loadModel
File "fairseq\checkpoint_utils.py", line 425, in load_model_ensemble_and_task
state = load_checkpoint_to_cpu(filename, arg_overrides)
File "fairseq\checkpoint_utils.py", line 315, in load_checkpoint_to_cpu
state = torch.load(f, map_location=torch.device("cpu"))
File "torch\serialization.py", line 797, in load
with _open_zipfile_reader(opened_file) as opened_zipfile:
File "torch\serialization.py", line 283, in init
super().init(torch._C.PyTorchFileReader(name_or_buffer))
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory
[Voice Changer] pipeline create failed. check your model is valid.
[Voice Changer][RVC]: update_settings serverReadChunkSize:128
[Voice Changer][RVC]: update_settings f0Detector:dio
[Voice Changer][RVC]: update_settings silentThreshold:0
[Voice Changer][RVC]: update_settings enableServerAudio:0
[Voice Changer][RVC]: update_settings extraConvertSize:16384
[Voice Changer][RVC]: update_settings modelSlotIndex:1710271697015
[Voice Changer] Pipeline is not initialized.
[Voice Changer] Waiting generate pipeline...
Ayo? @faint cradle level 1 !!! 
Hi guys, i'm using coqui XTTS (with webui project on github) i have cool result but i want to try RVC+TTS, do you have advice of top you use project work's locally ?My goal is to have natural voice for translate my video
Ayo? @undone heath level 1 !!! 
Ah shite, looks like I have to download the game to provide me self with the optimal sound files
my bad
does anyone know where i can find older models that were on here before the original server shutdown?
guys should i improve the dataset using FL Studio or just train it after the seperation?
can i change the names of index files?
i dont want all that stuff before it
just the model name
Hey all Im having some trouble with Applio whenever I try to download a model. it says its successful but when i hit refresh under inference it comes up with a bunch of random model names i don't know. This is with every model so i don't know what I'm doing wrong .
Ayo? @turbid pagoda level 4 !!! 
And none of the names are of the model i want to use. ive tried like 5 different models and it all comes up with the same random names when i hit refresh
just to prove it says successful
which file should i use if i wanna download the version of rvc that lets you use custom audio
normally, rvc dont work
do ov2 super models not work on ilaria or something?
how do i download rvc
what's the error you get ?
i either get the ERROR download failed one
or it says it worked and to refresh the thing
but it doesnt show up
if i import a separate older model it works fine tho
ive tried uploading the model to my own huggingface and google drive
and copying those links
unzipped the file, rezipped the file then tried as well
I get the same thing most hugging face models don't work for me. All I've heard is some Will work some won't.
no idea why this is happening
freddiemercury
Ayo? @hot apex level 1 !!! 
nice one detective
That's why I switched to applio and I'm getting kinda frustrated with no models working.
thank you
The only ones
I've gotten to work is Hank Hill and plankton lol
WOAH okay so i think ive found a fix
have to change /blob/main/ to /resolve/main/
in the link name
i compared the older models link to the newer ov2 one
figured it out
never knew this
I need help, I can no longer do ai cover, my google colab rvc no longer works
Wats the link for this?
I had problems too. You'll have to switch to llaria. Scroll up on the chat for a bit and you will find a list of Links Google colab no longer works
for mobile ?
Did 2 of my models I uploaded get deleted?
Anyone know how to make harmonies in fl for the ai vocals
yes
For what
It didnt
prob have dms off
Am I about to get banned
its just that the models you uploaded were deemed to be low quality and was deleted
not at all lol
I don't understand how they are That bad
Please, don't ping staffers.
calm down
Can you calm down?
look at tuto
This isnt even the voice changer help channel
Did you got pissed or something?
Also, please go to the help-w-okada channel and explain your problem there.
It is necessary to get angry and insult others?
Even if you did put * it isn't okay to use slurs only because you don't get immediate help.
Hey, if you're not going to be respectful, please step back.
All you have said is "can anyone help me set up W-Okada" but you didn't specify your problem.
Unless you don't know how to use W-Okada, you could have said this politely.
please remember helpers are real people and have things to do
Ayo? @tribal spire level 2 !!! 
Install the VB Audio Cable driver.
I want to make Eminem music with my own voice, how do I do it?
im at school bro
@dawn nebula I want to sing Eminem's music with my own voice, how do I do it?
u train ur model
from where
Btw Delik we moved the docs to https://docs.aihub.wtf/
Last update: Mar 10, 2024
The aihubdocs one has nothing in it
yeah just saw that
Yes
yeah i will use the new link
The tensorshould board page should be a priority. It doesn't have a redirectory
Redirectory?
We moved everything to the domain. It should be there
In RVC → Resources → Epochs & TensorBoard
OH SHIT I FORGOT TO CHANGE HYPERLINKS
I can do that in a literal second is fine
Thing is, they forgot to give me editing perms 
That's why I didn't mention anything since it's a quick change. I thought someone was editing it
I understand. You are always free to bring suggestions btw
A?
Bro doesnt have edits of her own docs😭
Right now on Ilaria RVC if I use the download voice models tab, they don't show up as an option after hitting refresh in the inference tab, I am stuck with the same single voice model as an option
I want Ronnie James Dio instead of Dio from JoJo's Bizarre adventure lol
you forgot to click Refresh
Pink button on top
It's not working
I am using the gradio.live rather than the huggingface.co website, does that make a difference
Ayo? @robust locust level 1 !!! 
I can't add an Image to my model request.
I can't upload an image to my model request.
Hm...
Send me a DM I'll send it to your Channel
The Image?
yep
I have no knowledge of any of this sorry
Any chance anyone knows how to get more quota on elevenlabs for free
u gotta to u will love it bro
I will I will
Ayo? @urban silo level 1 !!! 
thanks nd sorry for the ping
Np. Good luck
You need to be Level 5
is rvc down? 
Hey, quick question. Where would one go for help with a voice not being able to be downloaded?
Treying to download the Neuvillette rvc, but it says the password or username is invalid.
Is there any guide docs for gpt sovits?
is lliara rvc fixed?
Genshin models are usually taken down because it's requested by the VA's
Fair enough
Ayo? @golden quest level 1 !!! 
I swear it's only them because they have a strict contract 
regarding the index feature that u can slide from 0 to 1 is there any way to make this higher in future updates to improve model tone accuracy?
-colab
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
hello everyone
I want to introduce real time speech to text in my django project. it use English, Hindi and Marathi. how to do? plz help me
which model is suit for this project
have chinese people here?
check #1212368971127590922
Ilaria RVC doesn't work
Ayo? @ancient wasp level 1 !!! 
how can i fix it?
!howtoask
How To Troubleshoot 
- 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.
- 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.
- 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.
Screenshot plz.
it seems Google Drive dosn't work anymore.
How to use models?
what does it mean if g/total stalls out and dosent decrease or increase lol
I think ure using an outdated colab, use the new ones
-colab
- Applio, by IA Hispano 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
- RVC Disconnected, by Kit Lemonfoot Google Colab
- easyGUI, by rejects Google Colab
- Applio, by IA Hispano Huggingface Spaces
- Ilaria RVC, by thestingerx Hugginface Spaces
- RVC-HFv2, by r3gm Huggingface Spaces
- AICoverGen, by r3gm Huggingface Spaces
- Advanced RVC Inference, by r3gm Huggingface Spaces
- RVC v2 Huggingface version, by Clebersla Huggingface Spaces
U can find guides here:
https://docs.aihub.wtf/
Last update: Mar 10, 2024
screenshot 👀
I think you can stop the training, the model is not improving
👍
Ayo? @bright jackal level 1 !!! 
Can some make a Blitzo from Helluva Boss voice model
is there a currently working colab for iiaria rvc
i may be stupid
doesn’t seem like iiaria colab is in the #1188551486620303442 though idk
Ayo? @teal dome level 1 !!! 
The tempfix one works.
Idk what happened to the main Ilaria RVC Colab that it stopped working again.
yeah just tried the main one and got some torch module error. i was using this tempfix one for a while, is it the same one? https://colab.research.google.com/github/poiqazwsx/RVC-Notebooks/blob/main/Ilaria_RVC_Tempfix.ipynb
i got the same torch module error with it
That's weird, since i tested It some moments ago and it worked for me.
It's bugged for some reason
I guess It's possible to fix it, but it really is weird
someone can help me? i got trouble with pipeline is not initialize
i cant find the answer at github okada. what must i do??
I am on the "Load Dataset" under "Preprocessing" step. I had to create a rvcDisconnected folder under my one drive and put my file there unfortunately I went into some other folder by accident and don't know how to back out of it... Please help....
filename
change it to somthing without spaces and with only "regular" letters & numbers
although I also have an rtx 3060, rvcui sees 0 cuda how do I fix this
Can someone help me?
What's your issue ?
Scroll up
yo, i have a problem it says this
Traceback (most recent call last):
File "/content/Ilaria-RVC/app.py", line 1, in <module>
import subprocess, torch, os, traceback, sys, warnings, shutil, numpy as np
ModuleNotFoundError: No module named 'torch'
and doesnt show the Gradio
should be under a folder called mydrive or drive from there
relaunch, it bugs out randomly somehow
@sand iris Nope. Don't see a folder that says that
or maybe under content then
okay let me check
i only have the install illaria rvc and start
@sand iris I found it. Thanks, however the zip file i put in there doesn't come up underneath it?
Maybe it didn't refresh ? Not sure if it requires a restart or not, I don't think it does
might be the file explorer being bugged
let me see, perhaps
@sand iris Was clicking upload a mistake? because that is what I did under the rvcDisconnected folder, chose the file you see in the screenshot, went to load the "DataSet" and continuing to get this error in screenshot 2...
I don't use colab much, not sure I'll be able to help
naive guess : check the exact filenames ? The start of the filenames do match...
Okay so it's weird... doing this process created a new zip folder i had to rename to ..."TESF2.zip" and i also changed the file name in the dataset to the same with a 2 at the end and now this... im beyond lost at this point...
Ayo? @torn bluff level 3 !!! 
im retrying with the first folder that is actually in the rvcdisconnected folder as it asks me to do, see what happens. I even renamed it back to "...TESF.zip
idk anymore... nevermind, the moment i select the file. It will give me the loading circle, ill wait until it's done but then it will ask me to select a file from my computer... im beyond lost at this point.
OMG I THINK I FIGURED IT OUT! IM SUCH A IDIOT. Let's see if it works this time first before i get too excited
now it says sed: can't read app.py: No such file or directory
/bin/bash: line 1: .venv/bin/activate: No such file or directory
python3: can't open file '/content/app.py': [Errno 2] No such file or directory
Yup im the idiot.... The folder I was trying to upload was called "CoreyTaylorTESF" when the zip folder should be called "CoreyTaylor_TESF" (yeah didn't... name it with a underscore)....
gotta launch the installation cell first
yeah it reset
@sand iris It worked... im a idiot... lol look at my last comment.
thanks bro, now it works, is this error common ?
Kinda is ?
the green check mark means it's done on each of the steps right?
To reduce setup times, I made it use another tool that is much faster at setting up python. But it's quite young, and apparently, if it fails to download, it still continues
so at the end it's missing one thing to run properly
but that happens at random
Okay well "Feature extraction" says it's done but now im running into this error under "Index Training"...
OH NO! WHAT!? I haven't left it for that long!
OMG! Now I have to start all over!? BROOOO
Dude, i left it inactive for maybe 2 minutes at most, there's no way
@sand iris Does this mean, I don't have to do it all over again?
you might be using one of the older colabs
if you got your runtime ended this quickly
and considering the weird errors
Any suggestions? I'm just trying to train vocals i ripped from a few songs into one wav file so I can train the AI to do covers of other songs.
How do I eliminate voice cracks from my models
They are noticeable but they are not that many
Imagine running inference on a clip of your own dataset
just for it to return shrouded with voice cracks
You guys know where the Ilaria page went?
@sand iris still same
OH
you mean that cuda 0
that's normal
0 is the first device
1 is the second and so on
Whether the model has pitch guidance (required for singing, optional for speech):
For a rap model true or false ?
since its not really speaking or singing xD
Anyone knows more about this ?
Still waiting on my question
Ilaria page? U mean the kofi shop?
Or the google colab
What colab are u using send link
It was the one recommended in a YouTube video I found that was uploaded 2 weeks ago. Either way, there is new problems now. Now it's saying my GPU wasn't detected while restarting the project when I ran the dependencies.
U may have finished ur daily free gpu
That is what I was thinking but wasn't sure. Now I have to wait 12 hours?
When u got ur gpu back, do this step for inactivity issues, then re try using the rvcdisconnected
If your daily Gpu ended
U have to wait around 12-24 hours
Or either do everything again on another google acc
im going to screenshot this. Because I'll probably forget to do that. Thanks. Yeah ill just wait the 12 to 24 hours. I don't want to create another Google account. Thank you.
Yw
@low shard What was the steps again? You edited the comment. I clicked on the workaround link. If I do the steps in inspect elements that will get rid of the timeout problem?
The Colab
Well, that will work for when you got ur GPU back, but yea that inspect elements thing will help you to jot get that timeout
What problem you got with it
The link I'm using to get to it is a 404 page now
Ayo? @lament wagon level 3 !!! 
Last update: Mar 8, 2024
Thanks
Hi ı cant hear myself,ı dont think ıts also changing the voice when ı click the rec ıts not also record any voice ??
my microphone is working no problem but just this program not changes?
is there a colab version of RVC 2
I downloaded the song that i wanted and prepared it to make the ai cover. But now I'm stuck, I don't know what I need to do. I tryed simple GUI smth like that and it gives me JSONDecodeError on the model selection. Can anyone help me?
yes
Ayo? @delicate skiff level 1 !!! 
I think i fixed it
yes