✨│ai-help

[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 💖

[plush dew]

select vb cable as your input

[spark ridge]

it doesnt work

I try to say something but doesnt work

[plush dew]

https://rentry.org/VoiceChangerGuide#audio-setup

either it hasn't started, you haven't set up vb cable properly or your mic is too quiet

[spark ridge]

uhh

I tested it but it's just loud and nothing like I said

I started it

[plush dew]

screenshot your cmd

[plush dew]

screenshot the gui also

w okada

[spark ridge]

this?

[plush dew]

it should be going albeit with a huge delay

[spark ridge]

so what sohuld I do/

[plush dew]

have you set vb cable as input on discord?

[spark ridge]

Mhm

yes

is it because I didn't install pytorch?

[rare goblet]

Ayo? @spark ridge level 3 !!!

[plush dew]

no it is not

idk

[spark ridge]

it just sounds static

like radio static'

[plush dew]

lower the gain

[spark ridge]

still sounds like static

is it my rvc quality being low?

[plush dew]

try enabling monitor on okada check if you can hear yourself

[proud elbow]

try another better model

[plush dew]

i think he's using the stock ones

[spark ridge]

[spark ridge]

howt o fix

It's not working I pressed start but I can't hear anything

[brittle wing]

yoo i how can my voice be clear using a virtual mic anyone?

i keep having some electrical sound when i talk

on okada its good in discord its not

[worldly lodge]

the file it should be reading is in the logs but nothings coming from it to the web page

[buoyant sphinx]

anyone got some goot voicechanger which works on discord?

[old oracle]

-help

[azure marsh]

Command has expired

[old oracle]

.

how do I install rvc locally

[wild bobcat]

pressed "install dependencies" and i got this error message

how do i resolve this

[wild bobcat]

you can look up a tutorial theres plenty i think

[old oracle]

there all outdated for me at least its all tts videos that pop up

[wild bobcat]

then i dont know im sorry

i dont do local rvc i use the colab

[old oracle]

oh

[near spruce]

i get an error everytime i try to download the model

[wild bobcat]

@acoustic scarab could you please help me on this

[near spruce]

like theres always an error for me in step 2 of the colab

[rare goblet]

Ayo? @near spruce level 1 !!!

[carmine hearth]

outdated colab

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot 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

How to use?

🤗 Hugginface Spaces

• 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

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

[near spruce]

thank you

[acoustic scarab]

can't do, i dont use colab

[wild bobcat]

oh sad

[worldly lodge]

where does applio put index files when generating index? i just finished a model but i cant find the index

console is throwing this error

[carmine hearth]

I am also facing the problem that the index file is not being created in applio colab... In addition to that, the sync graph option doesn't seem to be applied properly to tensorboard... Is there a problem with applio colab?

[hot willow]

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot 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

How to use?

🤗 Hugginface Spaces

• 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

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

[arctic gazelle]

hey! i mixed the voice of taylor swift & ariana grande in the hope of making clean covers with that.
however, even after using UVR, i still have voice bugs etc.. anyone able to dm me & help me out? idm paying i just rly need to have a clean voice model and a way to make 100% clean covers

[wheat rivet]

U know the name of program of how to separate two speakers. I tried speaker diarization few months back. But now I couldn't really find the tutorial for it

[pine pivot]

-realtime

[azure marsh]

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

[rare goblet]

Ayo? @pine pivot level 1 !!!

[proper shale]

https://huggingface.co/spaces/Delik/pyannote-speaker-diarization-3.1 should do it i think

[wheat rivet]

thanks

[proper shale]

yw! :)

[lunar quartz]

how to suggest a model?

[brittle wing]

Does Rvc perform better on a RX 7900 GRE compared to a RTX 4070 Super

[proper shale]

#1159289738314919936

[lunar quartz]

thx

oh

[proper shale]

id say probably rtx just because it's NVIDIA

[lunar quartz]

i meant i already have the model done i just want to upload it

[proper shale]

ah

#1159514067187277865 to get the model maker role so you can post in #1175430844685484042

[brittle wing]

It's somewhat in pair with the RTX 4000 series cards

[mortal anchor]

Rtx 3060 ti

[rare goblet]

Ayo? @mortal anchor level 1 !!!

[brittle wing]

Did you download Pytorch and Python correctly

With the latest CUDA version 12.1

[marsh heath]

guys

pls help me

im stuck on: [Voice Changer][RVC]: update_settings gpu:0
[Voice Changer][RVCr2] Initializing...
gin_channels: 256 self.spk_embed_dim: 109
[Voice Changer] generate new embedder. (no embedder)
2024-06-09 14:17:57.7771898 [W:onnxruntime:, session_state.cc:1030 onnxruntime::VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-06-09 14:17:57.7825447 [W:onnxruntime:, session_state.cc:1032 onnxruntime::VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
[Voice Changer] Loading index...
Try loading... model_dir\5\added_IVF304_Flat_nprobe_1_K9GorillaTagYT_v2.index
GENERATE INFERENCER<voice_changer.RVC.inferencer.RVCInferencerv2.RVCInferencerv2 object at 0x000001A490463D00>
GENERATE EMBEDDER<voice_changer.RVC.embedder.OnnxContentvec.OnnxContentvec object at 0x000001A4B8384B50>
GENERATE PITCH EXTRACTOR<voice_changer.RVC.pitchExtractor.RMVPEOnnxPitchExtractor.RMVPEOnnxPitchExtractor object at 0x000001A4B8384B80>
[Voice Changer] [RVC] Initializing... done
[Voice Changer][RVC]: update_settings serverReadChunkSize:256
[Voice Changer][RVC]: update_settings modelSlotIndex:1705728161005
[Voice Changer] VoiceChangerManager initializing... done.
[Voice Changer] MMVC_Rest initializing...
[Voice Changer] MMVC_Rest initializing... done.
[Voice Changer] MMVC_SocketIOApp initializing...
[Voice Changer] MMVC_SocketIOApp initializing... done.

[proper shale]

anything on the GUI?

[marsh heath]

the voice changer tab?

[brittle wing]

What's your GPU

[marsh heath]

one time it showed me a blank white screen

[marsh heath]

how do i check? (sorry im kinda new to tech)

[brittle wing]

Task Manager

[marsh heath]

and then what

[brittle wing]

Click performance

And then screenshot the whole tab

[mortal anchor]

Yep

[marsh heath]

OOHH

i cant

[rare goblet]

Ayo? @marsh heath level 1 !!!

[marsh heath]

now i can :D

[brittle wing]

RTX 2060

[prime pond]

is the gtx 1060 good?

[brittle wing]

That's very good enough to run RVC

I think you installed something wrong lol

[marsh heath]

NOOOOOOOOOO D:

bro can u like look at my screen at vc

and tell whats wrong?

[steep moth]

are the any RVC voice models on this server

[marsh heath]

nice pfp

[lunar quartz]

how to upload rvc model to huggingface?

[marsh heath]

GUYS PLS HELP

[brittle wing]

I don't think I have the time but if I did I would gladly help

[marsh heath]

do u at least know what may be the problem

[brittle wing]

You might need to do lots of trial and error because I don;t know what the situation you're in now lol

[marsh heath]

oh ok thx

do i need pytorch?

[brittle wing]

Maybe you installed some incompatible version of Python with the rvc voice changer or you didn't have not yet installed pip on the rvc voice changer folder

It could mean many things

[proper shale]

no, alr comes with the changer

[marsh heath]

how do i do the pip part?

[brittle wing]

Don't you need like to manually install Pytorch first?

[brittle wing]

You can read the GitHub page here https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/docs/en/README.en.md

[rare goblet]

Ayo? @brittle wing level 6 !!!

[proper shale]

nope

installed okada on my mac with no python/pytorch

abt rvc realtime, im pretty sure it's prepackaged as well

but i might be wrong

[marsh heath]

https://www.youtube.com/watch?v=fJKdIf11GcI is this video helpful @brittle wing ?

[brittle wing]

https://www.youtube.com/watch?v=wCuJncQsXxI

This is what I watched before

[marsh heath]

ive already watched it

but this guy is bad at explaining

[brittle wing]

I very well understood what he explained

I think it's your vocabulary lol

[marsh heath]

what did he mean by ADD PYTHON TO THE SYSTEM PATH

and also

when i try to do this, pip says that it cant be found

[rare goblet]

Ayo? @marsh heath level 2 !!!

[marsh heath]

@brittle wing it worked yesterday

but when i restarded the pc and came back today, it didnt work

[turbid wolf]

i used to be able to do ai covers on one of the colab link but rn all colab links i see is for to model training does anyone still have the link of one of them ? i really dont wanna download 5 or 2 gb videos for to do something so little

[brittle wing]

-overtrain

[azure marsh]

Overtraining

You can detect if a model is overtraining if the TensorBoard graph starts to rise and never comes back down. An overtrained model will sound robotic, muffled, and won't be able to articulate words well.

Check these resources to learn more about this topic

• Epochs & TensorBoard from AI HUB Docs
• TensorBoard from 🍏 Applio Docs

[brazen willow]

W

[spark ridge]

Hello, can Intel UHD Graphics run?

[turbid wolf]

does someone have a rvc gui colab link that still works ?

[mortal anchor]

can any1 know why the res ping is so fucking high

[pastel oak]

Whats your GPU

[mortal anchor]

rtx 3060ti

[pastel oak]

show full screenshot of wokada

[pastel oak]

not only did you download an old version but you also downloaded the AMD version lol

[mortal anchor]

wait what

[pastel oak]

nvidia works on this version but would still be smarter to get the nvidia version

https://huggingface.co/wok000/vcclient000/blob/main/MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a.zip

[mortal anchor]

oh shit

thx

[pastel oak]

then

f0 det: rmvpe
dont use sup 1
dont use echo unless you need it

chunk and extra are good

[pastel oak]

do -colab

in this channel. then probably select Applio

[pastel oak]

you wont get any realistic results, the delay will be 10-20 seconds

[odd shale]

🐢

I don't recommend using Intel UHD Graphics for anything AI related

[rustic snow]

can i get help

i cant start

---------------------------------------------------------------------------
JSONDecodeError                           Traceback (most recent call last)
<ipython-input-7-851cea461113> in <cell line: 31>()
     31 if os.path.exists(config_path):
     32     # File exists, proceed with creation of creds and client
---> 33     creds = Credentials.from_service_account_file(config_path, scopes=scope)
     34     client = gspread.authorize(creds)
     35 else:

5 frames
/usr/lib/python3.10/json/decoder.py in raw_decode(self, s, idx)
    353             obj, end = self.scan_once(s, idx)
    354         except StopIteration as err:
--> 355             raise JSONDecodeError("Expecting value", s, err.value) from None
    356         return obj, end

JSONDecodeError: Expecting value: line 1 column 1 (char 0)

--colab

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot 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

How to use?

🤗 Hugginface Spaces

• 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

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

[odd shale]

Which colab are you trying to use?

[rustic snow]

idk one i used while ago

https://colab.research.google.com/drive/1Gj6UTf2gicndUW_tVheVhTXIIYpFTYc7?usp=sharing#scrollTo=F6zbsuNs6xZF
this one

[amber zephyr]

does it still wor?

[rustic snow]

i dont think so

[rare goblet]

Ayo? @rustic snow level 1 !!!

[amber zephyr]

do you know which ones are still usable and the ones that aren't

[rare goblet]

Ayo? @amber zephyr level 5 !!!

[mortal anchor]

@pastel oak should i use these settings

[odd shale]

That Colab is outdated

Use Applio Colab instead.

https://colab.research.google.com/github/iahispano/applio/blob/master/assets/Applio.ipynb

Read these docs to know how to use it: https://docs.aihub.wtf/rvc/cloud/applio-colab/

[pastel oak]

i said the ones you use previously were fine

80 chunk 16384 extra

[pastel oak]

Yea its outdated use the one leo linked

[brittle wing]

Does the voice blender work in Apillio?

[odd shale]

I'm not sure.

[thorn inlet]

hey is there a tutorial or something for the settings ? there's just so many idk what to do

like i installed a model but sounds horrible, ig it has to do with all the settings im seeing on my screen ?

sorry whats a colab ?

[rare goblet]

Ayo? @thorn inlet level 1 !!!

[thorn inlet]

🙏

[granite falcon]

im using applio to train a model but it keeps saying tensort not found

can anyone help?

[scenic gull]

-pretrain

-guides

[azure marsh]

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• Model training with Mainline RVC by Raven (ravencutie21)
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[granite falcon]

colab

im using chrome

didnt see no cookies popping up tho?

[sudden kettle]

Does anyone here know any good RVC AI TTS?

[low shard]

rvc is made for Speech To Speech (STS), but applio & ilaria use Edge TTS (of microsoft) built in them, basically you use their built in tts that generates the tts audio and then uses it as an input for the rvc voice model

else you can just use any tts generated audio and use it as an input

tip: use a tts model with similar language and gender to the rvc model

[sudden kettle]

Tysm!

[low shard]

yw! also remember to play around with the settings when using tts

if you don't wanna use edge tts, you could check the other tts in https://discord.com/channels/1159260121998827560/1212368971127590922 (they might be harder to use tho)

[sudden kettle]

Ok, thank you

[low shard]

yww

[odd shale]

melocoton.

[low shard]

?

[odd shale]

🐢 👍

[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 💖

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 💖

RVC Colabs and Spaces

⠀

Local Forks 🖥️

⠀
Mainline RVC
Original project, suggested for advanced users,
by the RVC-Project team.

Applio
Simplified, suggested for all, by the Applio team.

RVC Studio
Simplified, suggested for all, by SayanoAI.

Mangio-RVC
Simplified, may not be supported anymore, by Mangio621.

AICoverGen
Simple yet great way to make covers, by SociallyIneptWeeb.

Replay
From the greators of weights.gg, excellent product for everyone.
⠀

Made with love by Ilaria 💖

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 💖

[wheat bolt]

How do I make friends?

[dark burrow]

make friends

[wheat bolt]

Ok, thanks. Also, I'm new here and I'm scared that everyone's gonna think I'm a threat when in reality I'm just here for fun.

[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 💖

[neat compass]

Is there anyway to do AI voices on mobile

[odd shale]

You can try using the Applio Colab.

-applio

[azure marsh]

🖥️ 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

[neat compass]

OK thank

[odd shale]

https://colab.research.google.com/github/iahispano/applio/blob/master/assets/Applio.ipynb

Use this one.

https://docs.aihub.wtf/rvc/cloud/applio-colab/

And read these docs

[neat compass]

Thanks

[brittle wing]

is a little over 1 minute of audio enough to make a model

[odd shale]

It depends.

Normally we recommend over 10-15-20 mins of audio

But if you got only 1 minute but it's insanely clean, you can test training it

[brittle wing]

not really insanely clean theyre phone recordings outside

damn

[proud elbow]

💀

[brittle wing]

skull emoji me again and im touching u bro

[proud elbow]

like the results will be

[brittle wing]

that doesnt make sense

im so confused r u just trying to say it will be bad

😭

[proud elbow]

yes, also good for streaming rig

[thorn musk]

my voice keeps lagging while I speak, idk why
my gpu is the rx 7600
my cpu is the ryzen 5 5660
sample length is .15
fade length .08
extra interface time 5.00
prob a good idea to mention I am using RVC (Retrieval-based-Voice-Conversion)

[pastel oak]

do fade lenght max

are you using the dml version (which you should)

if you are running go-realtime-gui-dml and everything else fits, then its probably CABLE thats shit

https://rentry.co/RVCRealtimeGuide download the virtual cable thats linked in there its better

[tall warren]

Is there a tutorial for this one? :oo

[odd shale]

That's Ilaria Mainline?

[tall warren]

Yup

[odd shale]

Has probado el colab a ver si te sirve?

Si no te sirve puedes usar el colab de Applio

[tall warren]

Nop, dice que está desactualizada

[odd shale]

Usa el colab de Applio.

[tall warren]

Applio?

[odd shale]

Si.

https://colab.research.google.com/github/iahispano/applio/blob/master/assets/Applio.ipynb

[tall warren]

Hay tutorial para ese?

[odd shale]

Si, tenemos una guia.

No te recomiendo ver videos de ningun tipo en YT con respecto a RVC

[tall warren]

Vale, vale

[tall warren]

¿Dónde la puedo encontrar? ♡

[odd shale]

https://docs.aihub.wtf/rvc/cloud/applio-colab/

Aquí tienes.

[tall warren]

OH

Graciaaas ♡♡♡

[odd shale]

Leela amiguita

[tall warren]

Voy

Hmm... Parece que es como Ilaria

[odd shale]

Aunque el colab de Applio está mucho más actualizado.

[rare goblet]

Ayo? @odd shale level 73 !!!

[tall warren]

¿La interfaz es parecida?

[tall warren]

Oooh

[odd shale]

No.

Si tienes cualquier duda consultame.

[gaunt citrus]

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot 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

How to use?

🤗 Hugginface Spaces

• 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

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

[odd shale]

Deja que me encargue light.

[tall warren]

Install Applio, Start Applio y luego Public URL, no?

[odd shale]

Si.

Así es.

Ya borré ese instrumental, gracias por notificarme.

[tall warren]

Una pregunta que no tiene nada que ver, ¿Para qué es el local URL?

[odd shale]

De eso no te preocupes, es la ip del gradio.

[tall warren]

Ohhhh vale

Ya estoy aquí, ¿Dónde debería poner el link de la voz?

¿Inference?

[odd shale]

Pero no te preocupes, son seguros.

[odd shale]

Vete al apartado de Download.

Pasame capturas que sé que hacer pero para tener referencia para guiarte.

[rare goblet]

Ayo? @tall warren level 3 !!!

[tall warren]

Gracias Arclame. ♡

[odd shale]

Pulsa en Download model ahora y espera a que se te descargue.

[tall warren]

Vale

[odd shale]

Me avisas.

[tall warren]

Listo

[odd shale]

Ahora vete al apartado de Inferencia

(Inference)

[tall warren]

Uhum

[odd shale]

Y en donde te lo indica, donde dice "drag and drop" arrastra el archivo de audio que quieras usar

Ojo, el audio que quieras usar debe estar limpio

Sin ruido ni instrumental

[brittle wing]

how would I go about making a web ui that you can only use predefined models on like how uberduck.ai used to be

[tall warren]

"Upload Audio"?

[odd shale]

Muestrame una captura del apartado de Inference.

[brittle wing]

for example I have a folder in db/Garfield (Lorenzo Music)/ that contains main.pth and main.index

[odd shale]

I don't know how to reply to that.

All you can do is installing RVC locally and put the .pth's on the weights folder and on the "logs" folder put folders with the same names as the pth's and on these folder put their respective index files.

[odd shale]

Si, en upload audio arrastra el audio que quieras usar.

Toma en cuenta que no debe tener ruido ni instrumental o sino el resultado en conversion va a salir feo.

[tall warren]

Sip, lo sé, es una acapella ♡

[odd shale]

Muy bien, aunque tiene eco y reverb, verdad?

[tired mason]

applio is so hard to use

[odd shale]

It's not that hard.

[tired mason]

bro what

[odd shale]

It's just matter of reading the docs bud.

[rare goblet]

Ayo? @tired mason level 1 !!!

[tired mason]

I tried

But I can

[tall warren]

Hm... Tal vez un poco de eco (reverb no sé qué es) pero es para probar el rango vocal del modelo

[tired mason]

t get to train something

Whenever I download it doesn't do anything

[tall warren]

A ver si debo buscar cosas más graves o agudos, voz de pecho o de cabeza

[odd shale]

El reverb es algo así como un efecto de delay en un audio.

Y supongo que esta bien, prueba con esa acapella.

[odd shale]

Download what?

Explain properly your issue please.

[tall warren]

¿Cómo sé si está cargando bien?

[odd shale]

Si tu audio te aparece seleccionable como se supone y no aparece ningun error, entonces está bien.

@analog obsidian ayuda a F por favor.

[analog obsidian]

Que paso?

[tired mason]

yea I have it open

But i try to install my .pth and nothing happens

[odd shale]

Amiguito, ese usuario llamado F tiene problemas usando el colab de Applio. Te invito a que lo ayudes como tu primer día de Helper.

[tired mason]

It just says succsessfully downlaoded

[odd shale]

Show a screenshot.

[tall warren]

Vale, entonces solo quedaría esperar, en las otras páginas que he usado suele tardar porque estoy en una casa con mala conexión, así que cuando cargue paso captura de pantalla

[odd shale]

Todavía está cargando el audio?

[tall warren]

Uhum, suele tardar por la mala conexión a internet

[analog obsidian]

if you're having issues in applio colab try to restart the session

[tired mason]

ok thanks

[analog obsidian]

http://docs.aihub.wtf/rvc/cloud/applio-colab/

colab tutorial

[odd shale]

Okie, me avisas.

[odd shale]

Y así se ayuda.

No quiero sonar como un NPC de Tutorial pero te vas a encontrar con gente con problemas así.

[analog obsidian]

si me ha tocado algunos con problemas raros en making models xD

colab es algo buggy no es culpa de ellos

[odd shale]

Si ves que alguien tiene problemas con un colab de Kalomaze o Easy GUI V2 (desactualizados), diles que usen el colab de Applio.

Eso es algo basico que un helper hace.

[analog obsidian]

si de hecho les recomende applio

[tall warren]

Lo he intentado varías veces y si no muestra eso simplemente me muestra Error

[odd shale]

Si hay un problema de alguien que no puedes resolver (como cosas de W-Okada o etc), dale ping a Shad.

[vast pond]

how do i verify?

[analog obsidian]

okei

[odd shale]

Pero eso es cuando estás subiendo el audio o cuando estás intentando ya convertir el audio?

[tall warren]

Subiendo el audio

La verdad creo que es por la mala conexión

[odd shale]

Es probable.

[analog obsidian]

estas en el celular o en el pc?

[odd shale]

ella está en celular

[analog obsidian]

tal vez le funcione si usa chrome

[tall warren]

Estoy en Chrome

[odd shale]

Es que ella tiene mala conexion Lierito.

[analog obsidian]

oh

si es el lag

F

[tall warren]

Lo suponía jsjs

Cualquier cosa yo he usado otras RVC antes así que supongo que cuando cargue el audio elijo el modelo y convierto, no?

[odd shale]

Si, en efecto.

[analog obsidian]

sip, seleccionas el pth y listo

[tall warren]

Vale, ¿El modelo como aparece? Porque a diferencia de otras no me dejó ponerle nombre al descargarlo

[analog obsidian]

anda al lugar de la inferencia, luego dale a actualizar

deberia salirte el .pth

en el apartado de seleccionar modelo

[tall warren]

Refresh, ¿No?

[rare goblet]

Ayo? @tall warren level 4 !!!

[analog obsidian]

sip

[tall warren]

Ok se me cerró y ya no me quiere dejar entrar, voy a pagar otro wifi

[analog obsidian]

[odd shale]

Hicimos lo que pudimos mijita.

[tall warren]

Soy ese

[analog obsidian]

nada que podamos hacer con eso sdfjk

perdon:c

[odd shale]

Reinicia el colab.

[tall warren]

Lo sé jsjs, no se preocupen, más bien gracias :3

[analog obsidian]

sigue intentando hasta que te deje

reinicia el colab como dijo leo

[tall warren]

Refresh a la página?

[odd shale]

Si.

[tall warren]

Sigue mostrándome lo mismo

[wheat hearth]

I clicked one of the links in RVC Links collection and one thing says "Inferencing voice" what would i put there?

[tall warren]

Ah otra cosa, ¿Se pueden crear modelos de voz en teléfono?

Por ejemplo, si quiero hacer un modelo de un personaje de una serie que me gusta, de 0

[odd shale]

First than anything, i suggest you to use Applio Colab.

[wheat hearth]

Ok thank you do you know where i could get a link for that?

[odd shale]

These links are kinda outdated

[odd shale]

https://colab.research.google.com/github/iahispano/applio/blob/master/assets/Applio.ipynb

There you go.

[wheat hearth]

Thank you!

[odd shale]

You're welcome buddy.

[vast pond]

i am getting a error

---

JSONDecodeError Traceback (most recent call last)
<ipython-input-21-56a9da7fba12> in <cell line: 31>()
31 if os.path.exists(config_path):
32 # File exists, proceed with creation of creds and client
---> 33 creds = Credentials.from_service_account_file(config_path, scopes=scope)
34 client = gspread.authorize(creds)
35 else:

5 frames
/usr/lib/python3.10/json/decoder.py in raw_decode(self, s, idx)
353 obj, end = self.scan_once(s, idx)
354 except StopIteration as err:
--> 355 raise JSONDecodeError("Expecting value", s, err.value) from None
356 return obj, end

JSONDecodeError: Expecting value: line 1 column 1 (char 0)

[odd shale]

Which colab are you using?

[vast pond]

idk i am watching a random toturail on yt first time

[vast pond]

the voice is a minecraft anvilif that helps

[odd shale]

Nop, don't use YT tutorials.

[vast pond]

ok

[odd shale]

These are pretty outdated

[vast pond]

what should i use?

[odd shale]

Use the applio colab instead.

https://colab.research.google.com/github/iahispano/applio/blob/master/assets/Applio.ipynb

[vast pond]

what is new one i am using the "Easy GUI (for RVC v2, with crepe) (with improved downloader)"

[odd shale]

That one is outdated too.

[odd shale]

This one is the most updated.

[vast pond]

do i need to download it?

[surreal prawn]

bro where do yall find yalls datasets at i cant find anythinggg

[vast pond]

how does it work?

[odd shale]

No, just execute the cells.

[odd shale]

Yeah, it works.

http://docs.aihub.wtf/rvc/cloud/applio-colab/

Read these docs to know how to use it.

[vast pond]

k

[vast pond]

ty

[analog obsidian]

si pero necesitas limpiar el dataset en el pc

[odd shale]

I suggest you to gather audio from any source.

Idk who you want to make a model of.

[vast pond]

where do i put in the link of the ai voice?

[odd shale]

First of all, execute the Install Applio cell.

Then the Start Applio cell.

[vast pond]

it is currently

[odd shale]

Wait till it finishes downloading.

[wheat hearth]

I ran the 2nd one and it says "No dashboards are active for the current data set." on a thing that says TensorBoard that came up is that normal?

[vast pond]

it has

[odd shale]

Yeah, it's normal if you haven't put to train anything.

Don't worry. You're just going to test a model, right?

[odd shale]

Then execute the Start Applio cell.

And click on the public URL whenever it appears.

[vast pond]

k

[rare goblet]

Ayo? @vast pond level 2 !!!

[surreal prawn]

im just trying to look for females with nice voices

like there impossible

to find

[vast pond]

its downloading loads of stuff is that normal?

[surreal prawn]

with no backround noises

[odd shale]

Yeah, it's normal.

[wheat hearth]

I was looking to make an AI song

[odd shale]

Yeah, that means you're going to use a model.

[analog obsidian]

uhm people denoise their dataset before making the model

[vast pond]

i am in the applio do i pput the link in voice model?

[odd shale]

There are some female voice models on the #1175430844685484042 channel.

[odd shale]

Have you clicked on the public URL?

[surreal prawn]

i tried long ago but it does backround noises when i did so idk if i was doing it right

[vast pond]

yes

[surreal prawn]

id rather make my own tbh

[analog obsidian]

u can also check #1175430844685484042 and click e-girl tag if you want an egirl voice

[vast pond]

and do i upload the vocals to audio?

[surreal prawn]

even if i look in there there never realistic enough

[odd shale]

Ya se lo dije Liery

[wheat hearth]

How would i import that

[analog obsidian]

[odd shale]

😵‍💫

(me voy a petatear por ayudar mucha gente a la vez pero puedo resistir)

[analog obsidian]

best u can do is train a voice yourself

[odd shale]

Actually, you must go to the Download tab.

[vast pond]

umm i uploaded the file and nothing changed to the vocals

[vast pond]

k found it

[odd shale]

Then paste the model link on the "Download model" thing.

[surreal prawn]

thats what ive been trying to do but the data sets r hard

[vast pond]

i drop the sound in "drop files"

[surreal prawn]

how do i denoise properly

like when u denoise did it ever hear like static

in the backroun

[odd shale]

You would need tools like UVR.

[surreal prawn]

d

[odd shale]

Install UVR and it's denoise model.

Or use the MVSEP page.

[vast pond]

where do i put the vocals?

[analog obsidian]

denoising is a bit hard and tricky task, since this is ur first time stick to uvr denoise for a time

[odd shale]

Have you finished downloading the model and selecting it on Inference tab?

[analog obsidian]

use mvsep or uvr5, both have the same denoise model

[vast pond]

no, oh i need to download it

[odd shale]

Yeah, you must download it.

[vast pond]

how, i am so sorry

[odd shale]

Go to the download tab

[vast pond]

yes

[surreal prawn]

its not my 1st time i used uvr be4 but the problem was id hear a little bit of static

for some reason

[odd shale]

And paste your model link on the "Download model" thingy.

[vast pond]

👍

[odd shale]

It's because you didn't denoise it properly.

[vast pond]

it said model downloaded successfully, what next

i never downloaded a physical file

[odd shale]

Go to the inference tab and press "refresh" on model selection.

[vast pond]

not sure if its supppose to

[vast pond]

yes

K

[analog obsidian]

uvr-denoise model doesnt do this, denoise model will do just that xd, remove your noise

[surreal prawn]

i used it ab 8 or 9 months ago so maybe bc it was outdated?

[analog obsidian]

well yea

[odd shale]

The UVR denoise is still the same.

[analog obsidian]

en realidad el uso el modelo de separacion de acapella

[odd shale]

Cuál?

Porque yo uso MVSEP.

Y para quitar ruido uso X Minus ya que su denoise está más actualizado.

[analog obsidian]

osea en el pasado, ya que dijo que sus audios tenian un ruido metalico (el tipico artifact que dejaba kim voice antes)

[odd shale]

Y si quita la estatica.

[odd shale]

Entiendo.

[analog obsidian]

dijo que lo uso hace 9 meses ya

[surreal prawn]

what do u use here

[analog obsidian]

click this

and go to download center

vr tab, search for denoise model

[odd shale]

🐢: Te voy a evaluar. lo observa

[analog obsidian]

"uvr de-noise"

[analog obsidian]

no puedo ver uvr porque kaspersky me lo bloqueea lol

[vast pond]

what next?

[odd shale]

Click on "convert"

And that's all.

[mystic steeple]

Hi friends. how can I make my own sound model?

[odd shale]

You did mean "voice model" right?

Well, take a look at our docs.

[vast pond]

how do i download the finished work

OH

ty

[vast pond]

TYSM

[odd shale]

You're welcome.

[analog obsidian]

You have to record audios of yourself (10 minutes or more)
then clean them to remove any unwanted noise/sound, etc
and much more, i can't explain it with only one message

[odd shale]

No sé si el especificó que quería hacer un modelo de sí mismo.

Pero dale esta guia igual.

https://docs.aihub.wtf/essentials/how-to-make-voice-models/

[analog obsidian]

click that ^

[analog obsidian]

si entendi mal xD

[mystic steeple]

thanks broskis

[odd shale]

Ur welcome.

[odd shale]

🐢: Bien hecho Lyery.

[analog obsidian]

para la otra leere mas lento el mensaje jaja

[surreal prawn]

i seen denoise then i downloaded it but it was stuck at 100% so i restarted and no i dont see it

[analog obsidian]

oh... wait, is that uvr installation recent or is the old one you told me?

maybe it got corrupted

[surreal prawn]

nah its new im on a new pc

[analog obsidian]

try to download it again, and when it reaches 100%, wait a bit

like 2 minutes or so

[surreal prawn]

thats the thing its gone lol

[analog obsidian]

reinstall uvr xd

and remove uvr folder from appdata > local > programs > ultimate vocal remover

[surreal prawn]

i dont see a uvr folder anywhere in there

[rare goblet]

Ayo? @surreal prawn level 5 !!!

[surreal prawn]

and i redownloaded it its still not there

[analog obsidian]

uhm, well, anyways we can try to download uvr de noise again

[analog obsidian]

o

wait wait

go to the main window

select vr arch

uvr denoise should appear

[surreal prawn]

OHH

I GOT IT

do i use the lite version or no

[analog obsidian]

no

copy these settings

[surreal prawn]

alr and i denoise it before or after i edit and trucante in audacity

[surreal prawn]

^

[analog obsidian]

don't truncate

use audio labels

im gonna send u the tutorial

wait

[surreal prawn]

okay

[analog obsidian]

#1203830360878743612 message

do this instead

[surreal prawn]

okay so im confused

i did the uvr

then i did what u told me

and nothing happened

i think

like i dont notice a difference

[analog obsidian]

there still noise present in your audio?

[surreal prawn]

wait hold on i was using the wrong audio to begin with

i hear 0 noise

[surreal prawn]

but what about there being empty spaces in the audio

dont u trucant it to close it

[analog obsidian]

what labels does, is ignore every long silence in your audio and only exports where the person is speaking

[analog obsidian]

nice

use labels on the no noise audio

(theres too much to do besides this but u dont want to use rx lol)

[frosty pawn]

yo can anyone help me, its not loading up anymore

and i have no pic perms so i cant even show

when i load it up its permenantly on a white screen and the command thing says "wait web server" for infinity of time

[surreal prawn]

So for the ending.. is the noise its making when it cuts off fine

it does that on some of them not all

[surreal prawn]

wait dm me a ss rq

[analog obsidian]

its normal dw

[frosty pawn]

ok

[surreal prawn]

dont this mean the server temp down?

[frosty pawn]

its been happening for the past week now

[rare goblet]

Ayo? @frosty pawn level 1 !!!

[surreal prawn]

what gpu do u have

could be on the wrong version

[frosty pawn]

idk its called like amd ryzen 2600 or som

[violet heron]

That’s a cpu

[surreal prawn]

yea...

[violet heron]

check task manager under GPU 1

[frosty pawn]

shi

[violet heron]

If it’s not there check GPU 0

[frosty pawn]

geforce rtx 2060

[violet heron]

K

Just making sure

[analog obsidian]

yeah huggingface is down

[frosty pawn]

ts been down for a week?

what r yall using then

[violet heron]

https://status.huggingface.co

[analog obsidian]

if ur having issues with okada always check that site

to see if huggingface is up

if hugginface is fine then is another issue

[frosty pawn]

its up

it says on the website

[surreal prawn]

yo im trying to figure out how much batch size i need to use and to do that i need to know the length of the video when cutted up into all those files do u know how i can do that?

[analog obsidian]

how many minutes of audio? if more than 10 minutes do batch size 8

[surreal prawn]

nah i had 12 but the thing is idk how long the thing is when it cut all the empty audio out

[analog obsidian]

select all audios, right click, properties, details and scroll to see how many minutes you have

[surreal prawn]

alr thanks

i got it

[violet heron]

Try deleting the model dir and pretrain folders

Then run start_http again

[surreal prawn]

btw aside from rvc do yk why my models wont transfer on newer update

its like its broken

[analog obsidian]

where are your .pth files located?

[frosty pawn]

idk wher allat is

[surreal prawn]

in model_dir

[violet heron]

It’s in the voice changer folder

[analog obsidian]

weird u cant see them hmm, try to reupload them

[surreal prawn]

the bottome one is the latest one right?

[analog obsidian]

use win_cuda version

i suppose ur using windows + nvidia

so thats ur version

[surreal prawn]

whats the std used for then

[analog obsidian]

cpu only

u dont need that

[surreal prawn]

when i open it

[analog obsidian]

did you downloaded cuda version?

[surreal prawn]

yes

[rare goblet]

Ayo? @surreal prawn level 6 !!!

[analog obsidian]

if u click close, you're able to use okada?

[tall warren]

Gracias a los que me ayudaron y me tuvieron paciencia TT ♡

[surreal prawn]

it dont let me transfer any of my models but i think idk

[analog obsidian]

upload them again

felicidades xD

oh cierto

no subas instrumentales porfis

[tall warren]

OHHHHH VA

Lo iba a borrar y ya lo habían borrado JAJSJAJ

Perdón, no sabía TvT

[surreal prawn]

when i switch my input and output device it says internal error

and it wont let me switch from client to server

[analog obsidian]

#🔍│help-w-okada

[warm iris]

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot 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

How to use?

🤗 Hugginface Spaces

• 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

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

[umbral mango]

what should be the settings like for delay and everything cuz its laggy asf

[rare goblet]

Ayo? @umbral mango level 1 !!!

[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 💖

[plush dew]

@umbral mango

[brave garnet]

RVC Colabs and Spaces

⠀

Local Forks 🖥️

⠀
Mainline RVC
Original project, suggested for advanced users,
by the RVC-Project team.

Applio
Simplified, suggested for all, by the Applio team.

RVC Studio
Simplified, suggested for all, by SayanoAI.

Mangio-RVC
Simplified, may not be supported anymore, by Mangio621.

AICoverGen
Simple yet great way to make covers, by SociallyIneptWeeb.

Replay
From the greators of weights.gg, excellent product for everyone.
⠀

Made with love by Ilaria 💖

[brave garnet]

RVC Colabs and Spaces

⠀

Local Forks 🖥️

⠀
Mainline RVC
Original project, suggested for advanced users,
by the RVC-Project team.

Applio
Simplified, suggested for all, by the Applio team.

RVC Studio
Simplified, suggested for all, by SayanoAI.

Mangio-RVC
Simplified, may not be supported anymore, by Mangio621.

AICoverGen
Simple yet great way to make covers, by SociallyIneptWeeb.

Replay
From the greators of weights.gg, excellent product for everyone.
⠀

Made with love by Ilaria 💖

RVC Colabs and Spaces

⠀

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.
⠀

Made with love by Ilaria 💖

[dull ledge]

I have a question. If I am aiming for a realtime model, does it matter if its 32k, 40k or 48k? If it matters which one should I aim for?

[brittle wing]

Hmm...a Colab version of rvc studio?

[flint seal]

quick question, is there any rvc websites or apps that would allow me to train a model overnight? i've tried weights.gg which is good, but it only allows me to use up to 5 minutes of audio to train my model.

[brittle wing]

Use RVC DISCONNECTED colab

-colab

[azure marsh]

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot 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

How to use?

🤗 Hugginface Spaces

• 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

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

[analog obsidian]

32k models are better
40 and 48k training is not in a good spot

[flint seal]

i've used rvc disconnected, but how do i train overnight with it?

[brittle wing]

While you're sleeping?

[flint seal]

uh yeah i guess

or so i can save my progress?

[brittle wing]

Well no way but while the training cell is pressed you should press the export cell so it will save while you're sleeping

[flint seal]

ah ok, thank you

[brittle wing]

I also want to train overnight...but it's almost impossible there are always captchas to resolve or it'll automatically disconnect

[dull ledge]

@analog obsidian Thank you! I'll stick with 32k then.

[proud elbow]

it has up to 4 hours daily limit or less depending on frequent usage

if u want more freedom overnight, get a decent Nvidia gpu to train locally

[analog obsidian]

even pro colab users have a time limit iirc

[fading jewel]

if you have 44k audio but it only goes up to like 16k on the spectrogram, should you use 32k or do you need to resample the audio

[analog obsidian]

resample to 32k

[fading jewel]

can i do this using audacity?

[analog obsidian]

its recommended to resample in rx

but audacity can also resample too

[fading jewel]

does encoding matter

[unborn tangle]

hello guys! I have ryzen 5 3600x and AMD 5700xt which version of RVC should I download fromt the github? Thank you! Pretty new to this ❤️

[rare goblet]

Ayo? @unborn tangle level 1 !!!

[proud elbow]

no need to resample the dataset audio, just select the sample rate in preprocessing stage

[fading jewel]

O ic

yea I selected target sample rate 32k

[analog obsidian]

audio encoding? like 32 bit float? it does matter, however if you dont know the actual value of your audio, nothing bad happens if you set this to 32 bit float

sry if i didnt understand well

[fading jewel]

so dont need to resample audio dataset

my b

[analog obsidian]

rvc does this too

rvc normalizes and resamples automatically

[proud elbow]

there's no codec processing for wav format at all due to being uncompressed

also 32 bit float is preferrable

[fading jewel]

is the default pretrain suitable?

assets/pretrained_v2/f0G32k.pth or should I use one of the user-created ones

[analog obsidian]

titan is better than og pretrain

but og pretrain itself is already good enough

[fading jewel]

word

thank

u

[rare goblet]

AI HUB's leaderboard

Want to view more than the top 10 users? Click here

[dull ledge]

Do you guys agree that a good realistic voice REQUIRES a merge of at least 3 voices?

I've read it more than once from different people

[plush dew]

it is also not as reliable, it has been down whole morning/afternoon .-.

[rare goblet]

Ayo? @plush dew level 5 !!!

[proud elbow]

I'm still considering model fusion as experimental double-edged sword, though u can research it in the guide
#✨│ai-help message

[rough carbon]

guys is there somthing wrong with my acc i tried every voice model but when i speak it just says nonsense pls help me

[snow hazel]

Can I remove Harmonies?

[west nymph]

why doesnt my rvc open

[snow hazel]

Try using administrator

[violet heron]

hey that’s me

[violet heron]

Which file are you opening?

[west nymph]

go-web

[violet heron]

What happens when you open it

[west nymph]

a black screen that says "The system cannot find the path specified." shows up and it says "Press any key to continue" and when i do it just closes and nothing pops up

[rare goblet]

Ayo? @west nymph level 1 !!!

[violet heron]

Did you unzip it

[west nymph]

no what do i click for that

[violet heron]

When you open the bat file does it show a little pop up menu that has a “extract all” button

[west nymph]

yeah

[violet heron]

Click that

Then a new explorer tab should open with the same thing, just unzipped

Now run the file

[west nymph]

it dont work it keeps being weird

[violet heron]

Send a screenshot

[slow marsh]

Anyone on here looking to do paid work for me? I need someone who can help get some of our track vocals processed with AI

Or does anyone know where I could go to find someone who'd be able to help?

[fluid topaz]

i merge 5 models and now my friends believe me that im an actual vtuber

[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 💖

[dreamy jolt]

why does the audio need to buffer when i use a local install?

Isn't it like... a file in the temp folder?

[umbral mango]

how do i fix high ms in the downloaded models cuz the 4 original model doesnt have the problem it gives like 400 res ms at max but whenever i use a custom one it reaches like 20000 res ms

[proper shale]

i suppose it's because it's trying to reach the file

since it's hosting via gradio

buuuut im not sure

[proper shale]

export to ONNX

https://rentry.co/VoiceChangerGuide - uploading models to AMD section

[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 💖

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 💖

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 💖

[small briar]

when i tried to drop the file it says "You can't upload files of this type"

how do i make it work

it says only .mp3 or .wav files

but everything i found here is a .zip file

[proper shale]

.zips are for models

[proper shale]

this, is for audio files

the audio you wanna convert

[brittle wing]

how do i get drop down for voice model and index file in applio

[brave garnet]

RVC Colabs and Spaces

⠀

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.
⠀

Made with love by Ilaria 💖

[tidal iron]

I have a problem with downloading the model :( when I do it it shows me the error "JSONDecodeError" I can show a screenshot

[plush dew]

are you on colab?

[tidal iron]

yes

[plush dew]

link it

[tidal iron]

do you mean the website link?

[plush dew]

the notebook yh

the colab link

[tidal iron]

nvm it's already working sorry

[rare goblet]

Ayo? @tidal iron level 1 !!!

[plush dew]

dw

[azure marsh]

Suggestions for @tidal iron

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot 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

How to use?

🤗 Hugginface Spaces

• 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

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

[languid briar]

error step 2

[brave garnet]

RVC Colabs and Spaces

⠀

Local Forks 🖥️

⠀
Mainline RVC
Original project, suggested for advanced users,
by the RVC-Project team.

Applio
Simplified, suggested for all, by the Applio team.

RVC Studio
Simplified, suggested for all, by SayanoAI.

Mangio-RVC
Simplified, may not be supported anymore, by Mangio621.

AICoverGen
Simple yet great way to make covers, by SociallyIneptWeeb.

Replay
From the greators of weights.gg, excellent product for everyone.
⠀

Made with love by Ilaria 💖

[azure marsh]

Suggestions for @languid briar

☁️ Google Colabs

• Applio, by IA Hispano Google Colab
• RVC Disconnected, by Kit Lemonfoot 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

How to use?

🤗 Hugginface Spaces

• 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

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

[analog obsidian]

are u talking about the click?
try mouth de clicker or de-click

[real elm]

what do people usually use?

[analog obsidian]

for what?

[real elm]

im a little lost since there's many rvcs

[analog obsidian]

applio and mainline

[real elm]

which would you recommend

[analog obsidian]

applio, is easier and have multiple good stuff in it

[real elm]

i would've figured, i've only ever messed with applio, not much of mainline

what is replay?

is that any good? never seen that before so i would assume it's new

[analog obsidian]

no, its just another rvc* fork

[real elm]

ah, i see

[rare goblet]

Ayo? @real elm level 1 !!!

[small briar]

How do i train the voices

[analog obsidian]

http://docs.aihub.wtf/essentials/how-to-make-voice-models/

[analog obsidian]

for further help go to #📑│making-models that* channel is for helping new model makers

oh i see... i don't know

[brittle wing]

How long does it take to convert a ai cover in gradio usually?

[rare goblet]

Ayo? @brittle wing level 1 !!!

[broken plover]

is the way you speak to correctly do it change for each model?

some models i can speak faster and some i have to speak slower and more clearly

[opal kelp]

The RAM is yellow for me, is it normal?

[plush dew]

it becomes yello/red when the usage is high

unless it crashes it doesn't matter ig

[hardy grove]

which one is your voice changer

[proven hill]

i dont use a voice changer

but i suggest okada

[hardy grove]

did you make one?

[hardy grove]

yeah a voice changer that sounds like a 22 year old girl

or 25

like that

[proven hill]

dm me

[ashen trellis]

-rvc

[azure marsh]

Documentation

AI HUB Docs

https://docs.aihub.wtf/

🍏 Applio Docs

https://docs.applio.org/

✨ More guides

• Full AI Voice Model Training Guide (local) by Christopher Villanueva
• Model training with Mainline RVC by Raven (ravencutie21)
• AICoverGen Colab Guide by Eddy (Spanish Helper)
• Create a model with RVC disconnected (colab) by Angetyde

[broken plover]

how to i make the voices not come out sounding laggy, it only does this in discord

[proper shale]

use VAC by Muzychenko. most likely a cable issue...

https://software.muzychenko.net/freeware/vac470lite.zip

in case that link makes u feel iffy just search "virtual audio cable" and it should be one of the first options, then just download the lite version

[broken plover]

Ive fixed it by doing the stuff at the bottom of the rvc guide

[rare goblet]

Ayo? @broken plover level 1 !!!

[broken plover]

it helped with like the settings in discord and task manager stuff

ty tho

[proper shale]

ah

glad you could fix it though, i just realized i had a late reply there lol

[brittle wing]

can you launch applio on 0.0.0.0 so i can use it across my network?

[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 💖

[split frigate]

why i got no index file? even all the way down

[odd shale]

Which colab did you use?

Maybe you forgot to click on "train index" before starting training

[split frigate]

can i click it now?

[split frigate]

mangio

[odd shale]

Your model is already trained, right?

[split frigate]

i stopped training and click on it

i hope there is no issues will occur

[ashen matrix]

error in code on step 2of colab

[split frigate]

still training

[odd shale]

Wait till it finished training.

Then restart and try training the model again with other name but stop when you get your .index generated by clicking "train index"

[split frigate]

well i stopped and created an index file

[odd shale]

You already found it?

[split frigate]

and continue

[split frigate]

as u said. ididnt click on create index file

so i stopped training and create it

[odd shale]

Can you stop training and click on generate index?

[split frigate]

yeah

i dont think there is an issue right?

[ashen matrix]

any helpers

[odd shale]

Which colab are you using?

[odd shale]

Nop.

[ashen matrix]

Easy GUI (for RVC v2, with crepe) (with improved downloader)

[rare goblet]

Ayo? @ashen matrix level 1 !!!

[odd shale]

Don't use that one, it's outdated

Use Applio colab instead: https://colab.research.google.com/github/iahispano/applio/blob/master/assets/Applio.ipynb

[ashen matrix]

thx

[odd shale]

You're welcome bud.

[ashen matrix]

shit too confusing I DONE

[brittle wing]

why does rvc cli's rvc.py open in visual studio when i run rvc -h and rvc

and when i try to run the .py file i get this

Traceback (most recent call last):
  File "C:\Users\Administrator\Downloads\RVC_CLI-main\rvc.py", line 10, in <module>
    from rvc.configs.config import Config
  File "C:\Users\Administrator\Downloads\RVC_CLI-main\rvc.py", line 10, in <module>
    from rvc.configs.config import Config
ModuleNotFoundError: No module named 'rvc.configs'; 'rvc' is not a package

do i need to pip install rvc?

still didn't work

[plush dew]

have you also run "python rvc.py prerequisites"?

idk never used it

i think -h is the help command

[brittle wing]

there's like 10 blaises

[brittle wing]

still same error

[proud elbow]

[brittle wing]

@peak tusk I get this when using RVC CLI. Installed from install.bat and still nothing. But a weird thing happens when i type rvc in command prompt. It just opens rvc.py in Visual Code.

[rare goblet]

Ayo? @brittle wing level 5 !!!

[snow hazel]

Can I remove Harmonies?

[brittle wing]

btw i even used the env python.exe to run it and still the same error

[woeful shore]

wheres the guide to do real time ai voices like i speak on the mic and it does it a second after

the server changed since i last checked it and i cant find it

[brittle wing]

i think you're looking for okada

[woeful shore]

i was thinking mangio

[brittle wing]

and can RVC CLI use webm for inferencing?

[woeful shore]

mangio is real time

thats the one i have

and had

im forgoring the crepe part

[proud elbow]

you might want to look forward to the RVC realtime https://rentry.co/RVCRealtimeGuide

[woeful shore]

never used it for that it was alwasys realtime
i was just gonna ask if it had an update or if rvc went into v3

[proud elbow]

RVC realtime is improvement to what mangio has

[brittle wing]

apparently RVC CLI needs 3.9, i have 3.11

can anyone help me understand what epochs and everything with rvc is please

RVC CLI probably doesn't work because I forgor to install requirements.txt

you can install it like python.exe pip install -r requirements.txt right?

[proud elbow]

then install the 3.9

[green flax]

Can I create TTS with RVC?

[brittle wing]

what real pain looks like

is there a way to- not completely start over

theres no G or D file in existence for some stupid reason

I wish I could do this locally but for some reason training AI needs the vram of god to do

the vram of jesus

and luckily it doesn't work on amd cards, so I might as well call local training impossible

[rotund agate]

i keep getting this error

ImportError: cannot import name 'Mega' from 'mega'

[thorn musk]

sorry I just saw the message I got the other VAC and its still lagging

[brittle wing]

O

Well that's a disappointment and waste of time

[indigo relic]

Can someone explain to me how I would get a Tupac song for example and turn those lyrics into a 1950’s version of it?
Like what would I need to download to be able to do this?
I have examples of songs how I want them to turn into but I just don’t know how to do it or where to even begin looking

[opal patrol]

how i can make higher sample rate for my dataset

i am trying to upgrade my Yourself fnf voice model

and my real life voice too

[fluid latch]

you cant, itll just mess up ur model in the process

higher sample rate =/= better audio

[frail tendon]

Does anyone know why it doesn’t work on the Mac? It’s rushing all the time when you want to say something yk

[proper shale]

That's a manual task imo. You'll have to remix by yourself

[proper shale]

Can you send an example on how it sounds like?

[frail tendon]

Yes Later okay?

[proper shale]

Alright sure, just know that I don't do VCs n stuff, you can record your audio or just upload the converted voice if it's not realtime

[low shard]

Use #1159290752195633273, no other channel for promoting

[low bough]

can u guys help me make music with my lirics i got here

[rare goblet]

Ayo? @low bough level 1 !!!

[plain fulcrum]

Hey guys, I'm trying to make a plankton voice model and I was wondering, how long of an audio do I need for the BEST results? I don't care how long it takes to train, I want the best possible result I can get, is 1 hour enough? At what point does it become over training?

[rare goblet]

Ayo? @plain fulcrum level 2 !!!

[pastel oak]

are you saying you got VB-CABLE? If yes, get VAC from the link i sent you

Else, can you explain more on "lag"?

some people explain a different problem when they say its lagging

[thorn musk]

Uhh the voice kinda repeats itself over and over

[rare goblet]

Ayo? @thorn musk level 1 !!!

[thorn musk]

Idk how to explain it properly

[pastel oak]

does it echo in a higher pitch?

or actual lagging out of syllables

[thorn musk]

No no like literally just jitters

Like he-hel-lo

Constantly

[pastel oak]

can you tell me what file you run when you start RVC

[thorn musk]

The one that ends with dml

I'm on AMD

[pastel oak]

ok can you send a screenshot of your RVC gui as well as the command window on the top after you loaded it up

[thorn musk]

I'm not on my pc atm I'll send it as soon as I get on

[pastel oak]

oke

[opal patrol]

i mean i wanna remake it from 0 with same samples which will just have better quality

[frail tendon]

pm?

[rare goblet]

Ayo? @frail tendon level 1 !!!

[proper shale]

you can send it right here as you've got lvl 1 now

[frail tendon]

okay

[sleek lodge]

what software do i use to use the downloaded voice file i have?

[frail tendon]

Does it work for Mac?

[proper shale]

I'm pretty sure Okada does work for Apple Silicon Macs (M1 and above)

[frail tendon]

i have m3

[proper shale]

https://docs.aihub.wtf/ RVC > Local section

https://rentry.co/VoiceChangerGuide for realtime

[rare goblet]

Ayo? @frail tendon level 2 !!!

[frail tendon]

this sound come

@proper shale

[green flax]

I want to create a TTS in Japanese using this site (https://github.com/skshadan/TTS-RVC-API). Can you help me identify which part of the config file I need to modify?

[brave garnet]

RVC Colabs and Spaces

⠀

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.
⠀

Made with love by Ilaria 💖

[proper shale]

share ur settings

[frail tendon]

how?

[proper shale]

screenshot

[proper shale]

use a uploaded model, and use rmvpe as your f0

should work!

[frail tendon]

where can I find f0?

[proper shale]

f0 det, there

[frail tendon]

okay

[frail tendon]

is also available for German-speaking?

or only english?

[proper shale]

i suppose so, but your pronunciation n stuff may vary depending on the model

[frail tendon]

hmm oaky

[storm echo]

I want to create an rvc model but the gg colab link doesn't work, what should I do?

[proper shale]

Which Colab? And what error are you facing?

Cause maybe you're using an outdated Colab... or maybe it's some annoying error in RVC Disconnected...

[storm echo]

It is Rejekts' gg colab link (I got the instructions on youtube), and it seems colab has blocked the link to gradio ;-;

[proper shale]

That's reeeeally old sadly

Here's a guide + link to RVC Disconnected, an up-to-date Colab.

https://docs.aihub.wtf/rvc/cloud/rvc-disconnected/

[delicate kettle]

Can I install any of these locally on my pc

Does anyone know?

[proper shale]

honestly, its better to download local forks instead of colab notebooks

https://docs.aihub.wtf/ RVC > Local section got you covered

Applio or Mainline r the main ones

[thorn musk]

here u go

[snow hazel]

How can I remove harmonies in songs?

[brittle wing]

still same error

[plush dew]

uvr 5 or mvsep i think

[proud elbow]

uninstalling 3.11 then installing 3.9 should solve that shit (dont forget the env path as well)

[brittle wing]

i just changed my path environment variable

i even installed requirements.txt

[proud elbow]

try 3.9.8

[brittle wing]

still the same error

[snow hazel]

I don’t know how to do it in UVR

[brittle wing]

and i do have the rvc folder

[plush dew]

use mvsep then

upload your file and select bsroformer if you want to separate vocals/instrumentals

[snow hazel]

Does that remove harmonies?

[rare goblet]

Ayo? @snow hazel level 10 !!!

[brittle wing]

oh i had to download the release

and it's not rvc.py, it's main.py

[mellow monolith]

yo does a sukuna rvc exist?

[brittle wing]

why does rvc cli say no supported nvidia gpu found

i have an rtx 3060