✨│ai-help

[pastel oak]

no i meant the top not when you start conversion

did you manually install rvc with the repo?

[pastel oak]

Hmm i mean else have you tried a higher sample length to see if the issue persists?

and are you running it while playing a game?

[thorn musk]

it is the only thing running on my pc and yes I did

[pastel oak]

it still lags out, even on 0.30 ?

[thorn musk]

somehow it fixed itself

idk

[pastel oak]

lol

[brittle wing]

I downloaded the release

[pastel oak]

can you send me the link of which you downloaded

[brittle wing]

https://github.com/blaise-tk/RVC_CLI/archive/refs/tags/1.2.1.zip

[pastel oak]

Ah, even though you said rvc cli i didnt properly lock in that you downloaded rvc-cli since i never personally used it

xd

Let me check quickly

[brittle wing]

i added ffmpeg to path so now in theory it should be able to convert webm on the fly

pydub converts it, right?

[pastel oak]

I am not familiar with rvc cli unfortunately so i cant answer that

I think you should run the command to install cuda for your rtx 3060, do you know how to do that?

[brittle wing]

cuda toolkit?

[pastel oak]

inside the rvc cli folder, click on the url path. there you can then type in "cmd" and it opens inside the folder

then you can run the command, i think it was

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117

but im not 100% sure

[brittle wing]

still says i don't have a gpu

[crude bolt]

are you using python environments and if so did you intstall it into the python env.

cuz you can easily have installed it globally and not imported it into your env

[brittle wing]

env doesn't work either

C:\Users\Administrator>"C:\Users\Administrator\Downloads\RVC_CLI-1.2.1\env\python.exe" "C:\Users\Administrator\Downloads\RVC_CLI-1.2.1\main.py" infer --input_path "C:\Users\Administrator\test.wav" --output_path "C:\Users\Administrator\rvc_cli_test.wav" --pth_path "D:\dlmodel\morshu\model.pth" --index_path "D:\dlmodel\morshu\model.index"
Python path configuration:
  PYTHONHOME = (not set)
  PYTHONPATH = (not set)
  program name = 'C:\Users\Administrator\Downloads\RVC_CLI-1.2.1\env\python.exe'
  isolated = 0
  environment = 1
  user site = 1
  import site = 1
  sys._base_executable = 'C:\\Users\\Administrator\\Downloads\\RVC_CLI-1.2.1\\env\\python.exe'
  sys.base_prefix = ''
  sys.base_exec_prefix = ''
  sys.platlibdir = 'lib'
  sys.executable = 'C:\\Users\\Administrator\\Downloads\\RVC_CLI-1.2.1\\env\\python.exe'
  sys.prefix = ''
  sys.exec_prefix = ''
  sys.path = [
    'C:\\Users\\Administrator\\Downloads\\RVC_CLI-1.2.1\\env\\python39.zip',
    '.\\DLLs',
    '.\\lib',
    'C:\\Users\\Administrator\\Downloads\\RVC_CLI-1.2.1\\env',
  ]
Fatal Python error: init_fs_encoding: failed to get the Python codec of the filesystem encoding
Python runtime state: core initialized
ModuleNotFoundError: No module named 'encodings'

Current thread 0x000064d4 (most recent call first):
<no Python frame>

[delicate kettle]

Ima try to install it but I’m not tech savvy with these type of things 😂

[proud elbow]

open cmd in the env folder, then enter python -m pip list and see if torch+cuda thing is installed

[proper shale]

Good luck 🙏

[proud elbow]

if not yet, try do above, but instead do python -m pip install ...

[brittle wing]

installing requirements

Consider using the `--user` option or check the permissions.```

[delicate kettle]

Officially don’t know wth I’m doing cause I didn’t see no run install bat file 💀

So I just used run install

[proper shale]

... that's the same file

its just that your explorer doesn't show file extensions

[delicate kettle]

Oooh

[brittle wing]

install.bat is stuck

[strange mulch]

what the fluff is epochs😭

[teal dome]

what’s going on with ilaria? will the new colab link referring to ilaria rvc mainline still provide the colab gui

i did the download and then pressed run but it disconnects before it can finish

[low shard]

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

[brittle wing]

i installed everything but now it gives me this error

C:\Users\Administrator\Downloads\RVC_CLI-1.2.1\env>python ../main.py
Traceback (most recent call last):
  File "C:\Users\Administrator\Downloads\RVC_CLI-1.2.1\main.py", line 13, in <module>
    from rvc.train.extract.preparing_files import generate_config, generate_filelist
  File "C:\Users\Administrator\Downloads\RVC_CLI-1.2.1\rvc\train\extract\preparing_files.py", line 8, in <module>
    config = Config()
  File "C:\Users\Administrator\Downloads\RVC_CLI-1.2.1\rvc\configs\config.py", line 17, in wrapper
    wrapper.instance = func(*args, **kwargs)
  File "C:\Users\Administrator\Downloads\RVC_CLI-1.2.1\rvc\configs\config.py", line 32, in __init__
    self.json_config = self.load_config_json()
  File "C:\Users\Administrator\Downloads\RVC_CLI-1.2.1\rvc\configs\config.py", line 41, in load_config_json
    with open(f"rvc/configs/{config_file}", "r") as f:
FileNotFoundError: [Errno 2] No such file or directory: 'rvc/configs/v1/32000.json'

[plush dew]

why do you need the cli specifically 😭

[brittle wing]

to make a custom web ui

applio doesn't bind to 0.0.0.0

[low shard]

Ilaria RVC Mangio (old one that was in the docs) is discontinued, won't get updated and the colab is broken, just use ilaria rvc mainline

[low shard]

wdym

[strange mulch]

wat does this do

[teal dome]

i’m on ilaria-mainline-rot13-base64-real, i pressed download and it finished installing, then i pressed run and it was… doing something? and then it said execution restarted

[low shard]

Its the guide that explains what are epochs

[strange mulch]

alr ty

[teal dome]

for context idk what mainline really is, i’m just used to ilaria with the interface simply booting up through the rvc colab link (which is now defunct as i found out today)

[wispy lodge]

Are you using precompiled applio?

[brittle wing]

yes

[rare goblet]

Ayo? @brittle wing level 6 !!!

[delicate kettle]

Why isn’t the models I’m using on Applio not working

It says error

[teal dome]

huh… well that was simple. mainline is basically just like booting the normal ilaria rvc but with an extra step

unchecking “Drive_Backup” made it work (idk why i didn’t think to uncheck it)

[brittle wing]

i can confirm the file does indeed exist

[proper shale]

check to see what command prompt said

[brittle wing]

it doesn't work if you run from env folder

i need hubert_base.pt

yeah so i guess rmvpe.pt and hubert_base.pt aren't included for some reason in RVC CLI (maybe it would make the zip too big)

[delicate kettle]

I got it working somehow just had to restart the app apparently

Also how long does it take for applio to upload your covers cause mine is taking years

[odd shale]

Maybe your audio file you're trying to use is too long or your internet connection is bad

[delicate kettle]

My internet is good lmfao maybe the audio file that I got from YouTube?

Plus don’t know how I’ll find an audio file of the song I want to use that’s less heavy

[granite quail]

Trying to launch RVC-GUI and this comes up but doesnt start. I managed to launch it yesterday by trying multiple times and after a while I got it, but now it doesnt work. Anyone know the problem?

[granite quail]

I got it now again after just re-launching it several times... but surely theres a fix to this? or is there?

[brittle wing]

RVC-GUI is outdated

use Applio instead

[granite quail]

Okey thanks!

[odd inlet]

hello, im trying to download a model but its not working, does anyone know why?

[vale osprey]

-colab

[azure marsh]

☁️ Google Colabs

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

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

[rare goblet]

Ayo? @vale osprey level 1 !!!

[thorn musk]

@pastel oak sorry to bother so often but it keeps fixing itself and breaking at random

[rare goblet]

Ayo? @thorn musk level 2 !!!

[teal dome]

ilaria doesn’t work with mega links anymore? it keeps saying mega not defined

[halcyon sable]

Hello everyone, I would like to ask about rvc realtime colab? I have problems using it, it is showing me 'syntaxerror: jsonparse: unexpected character at line 1 col 1 of the json data' all the time, every single time I hit a button in the client interface.

[plush dew]

i thing you should use w okada on colab for realtime

[halcyon sable]

yes, it is a wokada interface

sorry to put the incomplete name, I said rvc realtime and it is actually wokada as server

the interface its just not working. and the audio is not being routed well. only the microphone is displayed as a selection in the drop list. the headphones are not. (I'm following this tutorial https://www.youtube.com/watch?v=tWd1SMRXRMw&list=PLknlHTKYxuNscCoVK-PQceoQO-TlLes8a&index=12)

[rare goblet]

Ayo? @halcyon sable level 1 !!!

[teal dome]

trying to use a weights zip and it just gives me an error of “No such file or directory: ‘./models/index/model.index’ even though the only two files in it are index and pth

[plush dew]

@halcyon sable does it give you the "unexpected char" after clicking on some things?

[halcyon sable]

te agradezco mucho voy a probarlo

[halcyon sable]

yes!

[final surge]

Im trying to start the voice changer using ngrok

but for some reason when im trying to change voice i got this error

eh i cant send pictures

"
Looks like there's a bit of a problem.

unknwon message

If you clear the information being managed by this app, it may be recoverable.

Error
unhandledrejection
no error stack
SyntaxError: JSON.parse: unexpected character at line 1 column 1 of the JSON data
no stack"

[halcyon sable]

yep, I get the same stuff (me sale lo mismo )

[final surge]

any fix?

[teal dome]

@proven hill would you happen to have an idea what could be going on here?

Traceback (most recent call last):
  File "/usr/local/lib/python3.10/dist-packages/gradio/routes.py", line 437, in run_predict
    output = await app.get_blocks().process_api(
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1346, in process_api
    result = await self.call_function(
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1074, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "/usr/local/lib/python3.10/dist-packages/anyio/to_thread.py", line 33, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 877, in run_sync_in_worker_thread
    return await future
  File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 807, in run
    result = context.run(func, *args)
  File "/content/Ilaria-RVC-Mainline/infer-web.py", line 328, in import_button_click
    return import_files(file)
  File "/content/Ilaria-RVC-Mainline/infer-web.py", line 302, in import_files
    shutil.move(os.path.join(root, file), destination)
  File "/usr/lib/python3.10/shutil.py", line 836, in move
    copy_function(src, real_dst)
  File "/usr/lib/python3.10/shutil.py", line 434, in copy2
    copyfile(src, dst, follow_symlinks=follow_symlinks)
  File "/usr/lib/python3.10/shutil.py", line 256, in copyfile
    with open(dst, 'wb') as fdst:
FileNotFoundError: [Errno 2] No such file or directory: './models/index/model.index'
2024-06-11 18:28:58 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict "HTTP/1.1 500 Internal Server Error"
2024-06-11 18:28:58 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset "HTTP/1.1 200 OK"```

[wary helm]

guys i got 2 copyright strikes cause i made AI covers of songs with the voice of myself, i added AI Cover suffix just to make sure people dont confuse me with a regular song covering channel, but now that i got a copyright abuse. Can i post my videos as "covers" to avoid getting strike. Since i use my own voice and model?

[rare goblet]

Ayo? @wary helm level 1 !!!

[plush dew]

I think that's automatic, some artists and labels don't allow to reshare their music

covers are included

[teal dome]

mega doesn't work either

    m = Mega()
NameError: name 'Mega' is not defined```

[wary helm]

I got my strikes from Universal Music Group, i read an article saying they're striking to AI covers.

[teal dome]

it might be the instrumental you're using, generally youtube's filter thing comes after those when they're used in covers

[plush dew]

On your video details you can check the copyright claim and it should say if they allow or disallow usage of their music

[wary helm]

Yeah but i'm getting the instrumental from the youtube also, wouldn't the video i downloaded be striked?

[wary helm]

the copyright claim is empty, says i can contact them for information.

[teal dome]

one would assume so but it appears if the official instrumental is used in an ai cover specifically then it's more likely to be taken down, mashups on the other hand usually stay up more often

[wary helm]

Which i dont wanna do cause im afraid they'd sue me or smth

[teal dome]

for katy perry specifically someone at capitol keeps on top of anything with "katy perry" and "ai cover" in the same video so those get taken down just because it's katy perry

lots of mashups and her instrumentals are left untouched on yt at the same time

[wary helm]

i'm sorry i don't know what does mashup mean

The translate isn't translating properly i assume

[teal dome]

it's just when someone mixes multiple songs together, it's usually simple like instrumental + acapella vocals

[wary helm]

(i know English, just unfamiliar with term mashup)

[wary helm]

oh

i got it

[teal dome]

and you would think the mashups and videos using the instrumentals would be taken down but it seems it depends more on if it's for an ai cover

or probably covers in general

[wary helm]

well i was making covers like this;

i think it can be considered as a cover

and i also don't use anyones voice, it's mine

[wary helm]

this specific audio didn't got strike btw, just an example

the original audio itself is an acapella show

[teal dome]

that part around 1:28 where there's no lead vocal for a bit, is that original audio the same as one you'd find on spotify?

[rare goblet]

Ayo? @teal dome level 4 !!!

[wary helm]

nah it's a acapella group you can find on youtube "Harvard Krokodiles"

[rare goblet]

Ayo? @wary helm level 2 !!!

[teal dome]

that's weird

might just be umg exercising their power too far

[wary helm]

as i said

[teal dome]

oi there's a lot of headphone bleed at the end where the crowd cheers owie

[wary helm]

this cover got striked.

i've used rvc's remover to seperate tracks

[lofty perch]

-paperspace

[azure marsh]

RVC v2 Tutorial for Paperspace

💡 Click on the button below to open the guide.

Credits: LollenApe

[teal dome]

it probably detects the instrumental

[wary helm]

lol

[teal dome]

when your strikes expire maybe try uploading the acapella by itself without an instrumental and fingers crossed nothing happens, just to see if there's a difference

[wary helm]

yeah but there is a video posted 6 years ago with 1m view https://www.youtube.com/watch?v=_WTlWnQFYSI

[wary helm]

i don't think i can get copyrighted for singing the lyrics... Can i?

[teal dome]

yeah that's the same thing i noticed with katy perry. it looks like it depends on the context in which the instrumental is being used

if it's being used in a ai cover or probably used in normal covers in general they probably strike it

[wary helm]

now you've said it, i realized that there arent any covers for the DNA. on the youtube with the original instrumental being used

So i guess you are right, the song itself doesnt allow covers with original instrumental.

[teal dome]

i haven't thought to look into regular covers and see if they use the original instrumentals but it makes sense since the instrumentals themselves usually stay up and most mashups are safe

[wary helm]

well do i really need to wait 90 days to clear my strikes, can't i create an another channel?

i'll search if it there are any covers before i post mines from now on

[teal dome]

idk honestly

[elfin salmon]

Thank you

[teal dome]

kind of frustrated cause i spent a bunch of time filtering an instrumental+acapella of a song so i can use it for a princess peach cover and so far i just can't get mainline ilaria to import any model

[wary helm]

still thanks man

appericate it

[teal dome]

np

apparently the only thing that works is a huggingface link from this server

[brittle wing]

whenever i try to download my model on AICoverGen WebUI it shows me this

it's been doing that since yesterday and idk what to do

it worked before

[plush dew]

why dont you use applio instead

they have pretty much the same things

[teal dome]

is it like a gui? i'm not familiar at all with it

[plush dew]

do you run it locally or on the cloud

[teal dome]

public url

[odd shale]

Use applio Colab instead.

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

[plush dew]

@teal dome

[teal dome]

alright lemme try that

will applio work with mega links btw?

connectioned errored out... guess not

@plush dew

[plush dew]

colab notebook died i think?

if not what does the cell outout say

[teal dome]

  File "/usr/local/lib/python3.10/dist-packages/starlette/middleware/exceptions.py", line 65, in __call__
    await wrap_app_handling_exceptions(self.app, conn)(scope, receive, send)
  File "/usr/local/lib/python3.10/dist-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    await app(scope, receive, sender)
  File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 756, in __call__
    await self.middleware_stack(scope, receive, send)
  File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 776, in app
    await route.handle(scope, receive, send)
  File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 297, in handle
    await self.app(scope, receive, send)
  File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 77, in app
    await wrap_app_handling_exceptions(app, request)(scope, receive, send)
  File "/usr/local/lib/python3.10/dist-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    await app(scope, receive, sender)
  File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 75, in app
    await response(scope, receive, send)
  File "/usr/local/lib/python3.10/dist-packages/starlette/responses.py", line 265, in __call__
    await wrap(partial(self.listen_for_disconnect, receive))
  File "/usr/local/lib/python3.10/dist-packages/starlette/responses.py", line 261, in wrap
    await func()
  File "/usr/local/lib/python3.10/dist-packages/starlette/responses.py", line 238, in listen_for_disconnect
    message = await receive()
  File "/usr/local/lib/python3.10/dist-packages/uvicorn/protocols/http/httptools_impl.py", line 568, in receive
    await self.message_event.wait()
  File "/usr/lib/python3.10/asyncio/locks.py", line 214, in wait
    await fut
asyncio.exceptions.CancelledError```

and underneath is

what are you bruh'ing me for it's just a question ☠️

[plush dew]

what were you trying to do?

[teal dome]

well there goes huggingface

[teal dome]

the "drop files" part for downloading models, i tried do an .index and then a .pth but neither works

i'm curious too what the 403 error that immediately pops up after finishing running the cell is about

[plush dew]

i think you can ignore that

[teal dome]

gotchu

    message = await receive()
  File "/usr/local/lib/python3.10/dist-packages/uvicorn/protocols/http/httptools_impl.py", line 568, in receive
    await self.message_event.wait()
  File "/usr/lib/python3.10/asyncio/locks.py", line 214, in wait
    await fut
asyncio.exceptions.CancelledError```
is this particular? i see the last line often

[rare goblet]

Ayo? @teal dome level 5 !!!

[plush dew]

can you link your model?

[teal dome]

yeah

https://huggingface.co/Xhepyxopila/MarioRVCModels/resolve/main/WonderPeach.zip?download=true

[plush dew]

try rerunning the notebook

idk

[teal dome]

going insane

should i check share_tunnel on the Start Applio thing?

also, here's the cell before starting the public url

DEBUG:asyncio:Using selector: EpollSelector
DEBUG:asyncio:Using selector: EpollSelector
Error: Unable to connect to Rich Presence. Could not find Discord installed and running on this machine.
Theme Loaded: Applio
DEBUG:urllib3.connectionpool:Starting new HTTPS connection (1): huggingface.co:443
DEBUG:urllib3.connectionpool:Starting new HTTPS connection (1): huggingface.co:443
DEBUG:urllib3.connectionpool:https://huggingface.co:443 "GET /IAHispano/Applio/raw/main/pretrains.json HTTP/1.1" 200 2689
DEBUG:urllib3.connectionpool:Starting new HTTPS connection (1): huggingface.co:443
DEBUG:urllib3.connectionpool:https://huggingface.co:443 "HEAD /api/telemetry/gradio/initiated HTTP/1.1" 200 0
DEBUG:urllib3.connectionpool:https://huggingface.co:443 "GET /IAHispano/Applio/raw/main/pretrains.json HTTP/1.1" 200 2689
Running on local URL:  http://127.0.0.1:6969
Running on public URL: https://a217969ab84cf92f28.gradio.live

This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)
DEBUG:urllib3.connectionpool:Starting new HTTPS connection (1): huggingface.co:443
DEBUG:urllib3.connectionpool:https://huggingface.co:443 "HEAD /api/telemetry/gradio/launched HTTP/1.1" 200 0```

and the upper half, the middle seems to just be links

+ localtunnel@2.0.2
added 22 packages from 22 contributors in 2.205s
/content/program_ml
tokenizer_config.json: 100% 25.0/25.0 [00:00<00:00, 183kB/s]
vocab.json: 100% 899k/899k [00:00<00:00, 7.37MB/s]
merges.txt: 100% 456k/456k [00:00<00:00, 42.0MB/s]
tokenizer.json: 100% 1.36M/1.36M [00:00<00:00, 6.11MB/s]
/usr/local/lib/python3.10/dist-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
  warnings.warn(
config.json: 100% 481/481 [00:00<00:00, 3.24MB/s]
2024-06-11 19:43:46.504909: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2024-06-11 19:43:46.504959: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2024-06-11 19:43:46.506192: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2024-06-11 19:43:46.513713: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-06-11 19:43:47.595939: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT

Downloading https://huggingface.co/IAHispano/Applio/resolve/main/Resources/rmvpe.pt to rmvpe.pt...```

well that was simpler than the first time

[barren acorn]

why people hear the voice electric in discord

[proper shale]

probably because of the virtual audio cable

use vac lite by muzychenko, doesn't have any of those issues

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

[teal dome]

so far the drag and drop for models works only halfway

.pth goes through just fine, but .index errors out

[barren acorn]

what setup

[proper shale]

setup64

[grizzled tundra]

Cuz ur caple is bad

[teal dome]

thanks for the help so far @plush dew, but do you know what this means? is there something wrong with the index itself?

Traceback (most recent call last):
  File "/usr/local/lib/python3.10/dist-packages/gradio/queueing.py", line 528, in process_events
    response = await route_utils.call_process_api(
  File "/usr/local/lib/python3.10/dist-packages/gradio/route_utils.py", line 270, in call_process_api
    output = await app.get_blocks().process_api(
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1908, in process_api
    result = await self.call_function(
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1485, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "/usr/local/lib/python3.10/dist-packages/anyio/to_thread.py", line 33, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 877, in run_sync_in_worker_thread
    return await future
  File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 807, in run
    result = context.run(func, *args)
  File "/usr/local/lib/python3.10/dist-packages/gradio/utils.py", line 808, in wrapper
    response = f(*args, **kwargs)
  File "/content/program_ml/tabs/download/download.py", line 36, in save_drop_model
    file_name.split("_nprobe_1_")[1].split("_v1")[0]
IndexError: list index out of range```

[final surge]

do applio have realtime voice changer?

[rare goblet]

Ayo? @final surge level 1 !!!

[teal dome]

different model this time, the peach model i intend to use (it's not on this server, but the pth imported perfectly.) it's just... the index being weird

[barren acorn]

voice chnage for rtx160?

[empty mortar]

-colab

[azure marsh]

☁️ Google Colabs

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

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

[plush dew]

i think a problem with the model

[teal dome]

jesus are .index that important? i converted a file with just the .pth and no .index and the vocal sounds nothing like peach, it straight up sounds like the singer from the sample

i'm gonna try drag-and-dropping the index from the huggingface peach that's on here

(i realise it sounds like i'm gonna try mixing the other peach pth with the huggingface index - that's not what i meant my bad. i mean i'm just gonna see if the index imports at all)

[proper shale]

index affects accent

[spiral galleon]

I need help

[teal dome]

welp. huggingface peach's index imports fine

[teal dome]

do the names matter by any chance? most of them are just "model.index" and "model.pth"

[teal dome]

and the pth imports fine as well. gonna try another model, then

putting this here for reference; same model but from different sources. so far the weights.gg's pth works but not the index, trying the huggingface one now

[proper shale]

nope

[golden ivy]

any idea what to do with this error "JSONDecodeError: Expecting value: line 1 column 1 (char 0)"

[teal dome]

according to WinDiff the index and pth's are identical between the models, but then:
huggingface: pth works, index works
weights.gg: pth works, index doesn't work

[proper shale]

outdated colab, use newer ones in https://docs.aihub.wtf/ RVC > Cloud section, especially Applio

[golden ivy]

preciated

[teal dome]

do you make sia covers by any chance

[golden ivy]

i do not, this is the first time im making a cover

familiar name or smth?

[teal dome]

ah, my bad. i saw the name- yeah

[golden ivy]

issok, been my gamer tag for a few years now

[teal dome]

i was like wow fancy meeting you here

[golden ivy]

i hope you do get to chat with them

[rare goblet]

Ayo? @golden ivy level 1 !!!

[teal dome]

good luck on your cover endevours anyway

[golden ivy]

happens a lot with my league tag as well, turns out revenantren is a streamer

[teal dome]

@plush dew any clue what's going on with this?

the index files are identical to each other but the one named "model.index" gives the same error from before ```DEBUG:multipart.multipart:Calling on_part_data with data[0:50110]
DEBUG:multipart.multipart:Calling on_part_end with no data
DEBUG:multipart.multipart:Calling on_end with no data
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/gradio/queueing.py", line 528, in process_events
response = await route_utils.call_process_api(
File "/usr/local/lib/python3.10/dist-packages/gradio/route_utils.py", line 270, in call_process_api
output = await app.get_blocks().process_api(
File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1908, in process_api
result = await self.call_function(
File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1485, in call_function
prediction = await anyio.to_thread.run_sync(
File "/usr/local/lib/python3.10/dist-packages/anyio/to_thread.py", line 33, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 877, in run_sync_in_worker_thread
return await future
File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 807, in run
result = context.run(func, *args)
File "/usr/local/lib/python3.10/dist-packages/gradio/utils.py", line 808, in wrapper
response = f(*args, **kwargs)
File "/content/program_ml/tabs/download/download.py", line 36, in save_drop_model
file_name.split("nprobe_1")[1].split("_v1")[0]
IndexError: list index out of range

[proper shale]

Perhaps it's the name. Maybe naming it as model.index screwed things over, as the program wanted an added_ index?

[teal dome]

is there a way to determine what the added_ name is supposed to be? most of if not all of the newer models from weights.gg use the same naming scheme "model.index" "model.pth"

[proper shale]

it should at least have this (3 numbers in IVF are random):

added_IVF221_Flat_nprobe_1_v2.index

maybe if you rename it to this, it might work

Not sure tho

[teal dome]

lemme try rn

[rare goblet]

Ayo? @teal dome level 6 !!!

[teal dome]

thank you for the help btw

[proper shale]

yw! i hope it works 🙏

[golden ivy]

back at it, new error "ModuleNotFoundError: No module named 'mega'", using this google collab page for reference "https://www.youtube.com/redirect?event=video_description&redir_token=QUFFLUhqbjR5ODUxa2k2dXFCVTItOE9EZ2ZCWFdMemYxUXxBQ3Jtc0tuaUh0MTAtUW5SdW9pMDZ4VVl0QnBiaDJhVkxzdzNqOUlZWnN1clBEVWRIM2huYlhLZENPTEpaa2h5OHlERkFlT1JWM05mdWxZbUZKWTRweGxHSXlQd1hxbFIwSmhFeVJCd2xqQm9rQ09zVFFHVVBUdw&q=https%3A%2F%2Fcolab.research.google.com%2Fdrive%2F1Gj6UTf2gicndUW_tVheVhTXIIYpFTYc7%3Fusp%3Dsharing&v=gtjCmJ_6d0k"

[teal dome]

yeah that worked! step in the right direction

but

what about the ones where i don't know what the added_ is supposed to be

would it matter?

[proper shale]

yeah that's outdated

[golden ivy]

dang

[proper shale]

yeah no, you just rename it like that and it'll do the thing apparently

[teal dome]

alright, lemme see what happens...

yep that worked

now the ultimate test... seeing how a conversion comes out

nope. it converts the sample but the voice is as if it was trained from the same sample instead of being peach herself

omgg

wait nevermind

it did the thing where it disappears and reappears to indicate it refreshed but it didn't. i tried again just now and now it's peach's voice

@proper shale thank you and denno a lot! holy moly

hopefully anyone else who's trying to use a weights.gg model with applio sees this and gets an idea of what to do

[proper shale]

yw!

[modern gale]

its not ai generated instrumentals or anything?? since theyre being made without using own vocals id assume a website can do it for you..? do you know which

[proper shale]

yeah you'd have to remix it by yourself because depending on AI for instrumentals can be hit or miss

i mean, you can use suno/udio to generate ai instrumentals/music but they're probably gonna be out of the original tune + other issues

[modern gale]

i wanna know how people are making them and possibly do it myself i just dont know how theyre generating instrumentals or what models theyre using

[modern gale]

ill check that out

[nocturne mural]

What happens is that the code to get the index is not prepared to receive the index without the format that it should have, maybe I will make a commit to make an exception for this case.

LikeCat

[brittle wing]

ok I noticed something

my pc lately has been restarting by itself only when I have RVC downloaded

what can I do?

I resetted like 4 times

[plush dew]

does it ever happen when playing heavier games?

[rare goblet]

Ayo? @plush dew level 8 !!!

[brittle wing]

nope

[plush dew]

thought it could be a power supply issue

[brittle wing]

thought so too

[plush dew]

does it straight up turn off or blue screen?

[nocturne mural]

https://github.com/IAHispano/Applio/commit/124c93009487832628d2a2244764fa784832d982 In case you want to update the applio code manually

well, bye

[brittle wing]

no it just restarts

I check temps

cpu is at 70 when its on

72.8 rn

[proud elbow]

u mean when running it?

[brittle wing]

no even when Im not running it sometimes

like

when I reset and download it

not even having it on

it restarts

[proud elbow]

starting to or finishing download?

[brittle wing]

after I install

[rare goblet]

Ayo? @brittle wing level 3 !!!

[plush dew]

do you mean it restarts when you have it downloaded on your pc?

[brittle wing]

yep

[plush dew]

if it's not a drivers issue i don't think it's really related to rvc

[proud elbow]

starting or finishing unzipping? and is it zip file or exe (extractor)?

[brittle wing]

someone said I need to reinstall drivers, doing it as we speak

[brittle wing]

extracted

[plush dew]

use ddu to do a clean drivers uninstall

[brittle wing]

will get back to you in a bit

[proud elbow]

extracted to an ssd or hdd?

[brittle wing]

ssd

[proud elbow]

and have u tried on another disk?

[brittle wing]

currently doing the driver fix

[proud elbow]

unlikely the gpu driver issue

[plush dew]

i'm pretty sure it is a psu/ram issue

otherwise i think it would bs

[proud elbow]

more like either the ssd or the unzipper application

[plush dew]

i thought it crashed after installing it

rather than during installation

[brittle wing]

ok drivers have been reinstalled

we'll see if it restarts now

if i open the rvc web thing can i open the website on my phone

So far, no restarts

[agile socket]

you guys genuinely made it to confusing

Im reading this thingit mentions installing libraries and depencies 20 times not once is it working

[plush dew]

what are you trying to install

[agile socket]

MY RVClooksweirdddd

or differnt from my friends

and i cant selecta output devicethat wont crash RVC

[violet heron]

It should install it for you

Or come with it

What guide are you reading?

[brittle wing]

Restarted twice in the last 10 minutes

[brittle wing]

Whyyyy

[proud elbow]

I suspect you might have ignored this

[proud elbow]

could possibly be the faulty PSU or SSD, there's no way even pcie gen 5 SSD could draw significant power enough to shut down the pc without any running game/workloads

[proud elbow]

[brittle wing]

Ok I think I found the issue
I set the max usage on the cpu to 99% and now made it 100% and now haven’t seen a restart nearly an hour now

If that fails, ill buy a new ssd

[brittle wing]

My hdd is external, idk if it’d work

[proud elbow]

then give it a try

[brittle wing]

Will do tomorrow

Currently eepy

[sly hare]

Hello! Where can i find the guide to fix the delay on the voice as well as issues with the voice capturing other people’s voices in discord. I am not a tech savvy person i apologise if this issue has been raised multiple times

[odd zinc]

-colab

[azure marsh]

☁️ Google Colabs

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

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

[rare goblet]

Ayo? @odd zinc level 1 !!!

[brittle wing]

Will the CPU usage limit be restarted again?

or permanently expired?

[rare goblet]

Ayo? @brittle wing level 1 !!!

[spice drum]

hey im using the tool, im pretty new, and all the voices sound like shit. any idea how to make them better? (ps what do you recommend for f0 method)

[proud elbow]

neither of them at all

-gui

[spice drum]

[azure marsh]

Yeah, I still use RVC Easy GUI.

[spice drum]

what does -gui mean

[spice drum]

THATS AN APP??

[proud elbow]

refer to https://docs.aihub.wtf/ for newer and better stuffs

[spice drum]

im just wondering whats the best f0 method or what one would recommend

because rn all my voices sound like they are goblins

[proud elbow]

the guide explains it all

[spice drum]

guess ill just fuck around and find out ig

[rare goblet]

Ayo? @spice drum level 2 !!!

[spice drum]

SHUT UP!!!!!!!!

[brittle wing]

does anyone know a good voice model that is just squeaks? LOL

[spice drum]

you could try grabbing a mouse of the side of the road but idk if that will suit the audio

[brittle wing]

LMAO

[spice drum]

you could get a good sample if you live in new york

[red lily]

I need help?

I tried using Ilaria Mainline Rot13/Based64..

I tried both, but somehow they don't let me convert a song.. just errors.

https://colab.research.google.com/drive/16LkwvFZeudTpUOsE_6bMjOq2qkxFo8Hr?usp=sharing#scrollTo=t8gn3pmO14On

I did use this before.. but yeah, i know this one ain't gonna work.

But this one, won't even let me to:

https://colab.research.google.com/drive/1mHKTGH5e3SAyDSBss1KtiYRbDdQzwSMs#scrollTo=0t_xKpQ-IuD2

[brittle wing]

-colab

[azure marsh]

☁️ Google Colabs

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

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

[umbral mango]

how to fix the stutters in voice when using vb-audio virtual cable

[plush dew]

let us know atp :v

[brittle wing]

How do I download the model in the new Ilaria RVC mainline?

[plush dew]

there should be a download tab

[brittle wing]

Last time I opened there wasn't

I'm not on the colab now

[brittle wing]

Uh where is it

@proven hill

Nvm found it

[brittle wing]

Ok so left overnight with a live stream on and the voice changer active, it didn’t restart
It seemed the issue was that I set the max CPU power to 99% to reduce the load

Still unsure, gonna wait a few days to confirm

[sage hatch]

hey

is google colab rvc still working!

[left sparrow]

Is RVC capable of TTS, or does someone know a better tool?

I can't sing but I desperately want to make a cover for personal meme usage, and I already have a model for it, but I need to generate the vocal tracks before anything fun can be done.

[tame mica]

there's some forks with tts capabilities (e.g applio) but you can also use an 11labs audio to use it

[proven acorn]

What's the most easiest way to make ai vocals

[tame mica]

bruh

-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

[finite pond]

RVC2_HF keeps giving errors

[halcyon sable]

hello all, I would like to know the diffference between the index file

[analog obsidian]

added index is the one you can use
trained is for dev purposes, ignore it

[halcyon sable]

ok thanks

[proper shale]

don't use that, probably outdated atp

[delicate kettle]

What are the best settings to use when using an ai model to make song covers

I notice that most of the time the models have static when they sing compared to versions that others are using

[scarlet crystal]

im having this problem JSONDecodeError Traceback (most recent call last)
<ipython-input-12-5dcc48ca3d02> 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)

can anyone help me?

[azure marsh]

Suggestions for @scarlet crystal

☁️ Google Colabs

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

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

[pastel oak]

the standard settings are usually already optimized well. you can i guess reduce index influence but else its fine

[scarlet crystal]

thank you guys appreciate it

[rare goblet]

Ayo? @scarlet crystal level 1 !!!

[delicate kettle]

Cause ngl I be trying to have the ai model not be mimicking the beginning of a song

@pastel oak

[scarlet crystal]

how do i replace vocals?

[placid holly]

does Hop-Lenght actually work on RMVPE?

last time i know it for Crepe stuff

i'm trying applio

[pastel oak]

is the song you input in there clean vocal with no background noise, no doubled voice etc?

[delicate kettle]

Well I mean I looked up the vocals version of the song but somehow it still contains background noise no matter what I do

Unless there’s some way to get rid of it

[rare goblet]

Ayo? @nimble blaze level 2 !!!

[placid holly]

i didn't read the manual, : it not for RMVPE

[lone gull]

hey, quite new here, I was wondering if there's an offline tts where you can load rvc .pth models? I have one for Vits that works but it's outdated.

[soft tiger]

Im in need of a AMD version of okada or another software for Radeon gpu's

can someone help

[crude flame]

You can use emoji's fork of w-okada https://github.com/deiteris/voice-changer/releases/tag/b1945

[soft tiger]

what do i download

@crude flame

and thank you for helping

[crude flame]

the dml version

[scarlet crystal]

i have one question after ive installed this once do i need to run it every time i open the site?

and its saying this

anyyone knows wsp?

[dusk plaza]

er

[soft tiger]

its not letting me select

and i was tryna find one that runs on my pc and not a browser

[wispy lodge]

It runs on your PC, it's just that the interface is in the browser

But wdym its not letting to select? You choose the model, select your GPU and it doesn't apply?

[brittle wing]

Having issues where its only picking up my voice with passthru cable input and output dont seem to be working correctly [Image attached]
https://cdn.discordapp.com/attachments/1130960997637685299/1250529585133654127/image.png?ex=666b45ee&is=6669f46e&hm=50ab0de5c317d9f62476e5d2661e286dcc476f1d714047e2e0e52322fa897e1c&

[real sphinx]

sup guys!

Coming from this website
https://github.com/SociallyIneptWeeb/AICoverGen?tab=readme-ov-file

Can someone please guide me where can I get
'UVR-MDX-NET-Voc_FT.onnx', 'UVR_MDXNET_KARA_2.onnx', 'Reverb_HQ_By_FoxJoy.onnx'
'hubert_base.pt', 'rmvpe.pt'

as these links don't work anymore
MDX_DOWNLOAD_LINK = 'https://github.com/TRvlvr/model_repo/releases/download/all_public_uvr_models/'
RVC_DOWNLOAD_LINK = 'https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/'

[scarlet crystal]

thanks

[odd zinc]

• colab

[azure marsh]

☁️ Google Colabs

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

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

[tidal vault]

Hello, I need help deleting a model from the weights.gg website. I don't want to put the person i modeled in danger, so please get back to me when you can.

[worldly lodge]

do you not know how to?

[tidal vault]

I got it taken down

its all good now

[worldly lodge]

oh good job :)

[tidal vault]

but yea i dont actually know how

i emailed them to take it down, but i dont know how to manually

they took it down

[worldly lodge]

hmm usually after you make a model you can edit your model and there should be a button that says delete model in red

[brittle wing]

Hi guys i installed rvc and tensorboard but when i try to enter tensorboard i got this error pls help me the error is : No scalar data was found.
Probable causes:

You haven’t written any scalar data to your event files.
TensorBoard can’t find your event files.
If you’re new to using TensorBoard, and want to find out how to add data and set up your event files, check out the README and perhaps the TensorBoard tutorial.

If you think TensorBoard is configured properly, please see the section of the README devoted to missing data problems and consider filing an issue on GitHub.

i just want to make a voice model with my favoruite anime character and i have 3 min clear voice .wav file 😦 she no talking too much

[rare goblet]

Ayo? @brittle wing level 1 !!!

[dusty valley]

for the applio thing, what does it mean by index file?

[limpid flicker]

is there any video tutorial? i am very confused and not sure how to use applio

download completed but still no green check?

[blissful haven]

I have a problem discord dosen't change ly voice when using it

[limpid flicker]

?

[frosty pawn]

yall start isnt loading it up for me anymore can anyone help?

[scarlet crystal]

how does this work?

like how can i install it on my pc?

[dusty lava]

illaria down?

[odd shale]

what you do want to do on Applio Colab?

[odd shale]

Yep, use Applio Colab instead

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

[granite crag]

I haven't used this in a while but why does it take so long, its been going on to 900/1.7s

[soft tiger]

dosent let me select my gpu

[proud elbow]

"dosent let me select"? you have selected the 6900 XT and it didn't work? or it still selects CPU?

[grim vine]

when i run download model it has this error

`JSONDecodeError Traceback (most recent call last)
<ipython-input-10-771e605f715c> 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)
`

Im trying to use the weekend model

i cant send screenshot tho

[azure marsh]

Suggestions for @grim vine

☁️ Google Colabs

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

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

[viral bloom]

Is there a FAQ? The default models work fine but as soon as I try and upload some from this server they do not work

[grim vine]

bet thanks imma try that out appreciate it 🙏 🙏 🙏

[zealous shuttle]

in f0 change, when i want to change a male to female, how much should i fill?

[placid holly]

12, and the opposite for other

[zealous shuttle]

Ohh....Okay

[kind aurora]

so i am a bit old.. getting my RVC from huggingface

Nvidia RVC is the Vcclient_win_std right?

or is it Cuda

[hallow zodiac]

help pls

i have it all together, but no volum

[sick ginkgo]

will rvc work on potato pcs

[spice drum]

using gui rn whats feature retrieval rate

also does any1 know like the best / good stats

[brittle wing]

Which RVC local client supports AMD GPU’s?

[north temple]

my rvc is not connecting to server

[wispy lodge]

What does it say in command line window? Should be something like this when you try to switch GPU

[brittle wing]

It can be 4,5,6,7 not always twelve

[tawny sundial]

I have a problem...

[brittle wing]

What are you using?

[tawny sundial]

mangio rvc web gui

[brittle wing]

Is that a colab?

[tawny sundial]

no

[brittle wing]

Or Huggingface

It's on computer

[tawny sundial]

should be huggingface

followed docs

yeah its mangio

[proud elbow]

huggingface spaces use CPU and it's slower than an average laptop cpu running locally

[tawny sundial]

I just keep getting the AttributeError: 'NoneType' object has no attribute 'tobytes' error

Last time I used it a few months ago I used to get this error

while of course I can chalk the slowness up to computer power, I have no clue how to fix that error.

[rare goblet]

Ayo? @tawny sundial level 2 !!!

[proud elbow]

just try the local fork anyway

-local

[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

[soft tiger]

yeah

[soft tiger]

uh lemme see

[proud elbow]

and you have tried restarting or even reinstalling the whole application?

[tawny sundial]

I have the local fork

thats what I said

[proud elbow]

did you input correct audio file path?

[tawny sundial]

audio path files should be correct

I double checked them

[soft tiger]

yes

[proud elbow]

the full path like D:\audio.wav ?

[tawny sundial]

yes

yeah, put in the correct folder with the right path location

[soft tiger]

.

ok it worked after a restart

[proud elbow]

I'm not 100% sure since my mangio and mainline rvc fork (prob tweaked abit) worked even on the input's full path containing spaces and some asian characters

or alternatively you can try Applio, just put the input files in the Applio\assets\audios and the input path dropdown will show them

[soft tiger]

but i was wondering if theres a application ui version

so i dont have to have a web page open

itll be a bit better for me

[tawny sundial]

I'll try a few things here

[umbral mango]

yo someone can help me whenever i call with someone it records their voice too and says it backk

[soft tiger]

@proud elbow

[soft tiger]

would you know

[wispy lodge]

There is RVC Realtime that has local UI
https://rentry.co/RVCRealtimeGuide

And original w-okada bundles a web browser to display the same web page, so it's not better

[soft tiger]

thats the one i have already

but i want one like okada so i dont have to manually input a model every 5 seconds

and then take 2 minutes scrolling throught my files, to find it

[wispy lodge]

Then there's no such app yet. This is how original okada works for example. The UI it opens is the same web page opened in a dedicated browser

[soft tiger]

for me it opens witha browser

[soft tiger]

thats more or less of how i want it

i dont want mine on a browser

[wispy lodge]

Well, there's no such option in this version yet as it was actually trying to move away from local client 😅

[soft tiger]

oh damn

[wispy lodge]

But you can download original wokada version and run its client. It will display whatever is located at http://127.0.0.1:18888

[slim geyser]

copy and paste your models into the folder it opens when you try and select a model

[soft tiger]

it works now chat

[soft tiger]

and thats now how it works on the client i was referring to

[wispy lodge]

If you really want it to run in a separate UI app, then you can use this workaround:

1. Get the original version: https://huggingface.co/wok000/vcclient000/blob/main/MMVCServerSIO_win_onnxdirectML-cuda_v.1.5.3.18a.zip.
2. Run MMVCServerSIO.exe from modified version.
3. Close the opened tab in browser (this will not close the server).
4. Run voice-changer-native-client.exe from original version.

Should look the same as on the screenshot and work as usual

However, I might make a lightweight version of it once I get to UI overhaul...

[jaunty torrent]

can someone help me with making an ai voice model?

i tried to make one but i couldn't get pass the step 5 on gradio which is the download phase

i did everything correctly and yet it caused an error when i tried downloading my model

can someone help me please?

[split frigate]

is this model training unstable and needs to retrain.. or its okay and complete till the lowest point?

[brittle wing]

What voice changer do you guys use for RVC2 models? Im asking because none of them work on w-okada! but RVC1 does

[pastel oak]

rvc v2 models should all work on wokada, unless something messed up on your installation.. check if it says anything unusual on the command window when you start the voice changer with a v2 model, anythign about pipeline

[pastel oak]

best bet is just listen to it and see how it sounds like.. also check for d/total, mel and kl

and make sure smoothening on the left is on 0.987

[split frigate]

how can i balance between 4 diagrams

[pastel oak]

kl, mel and g are next to each other so i usually only filter for d/total and go to the last page of the diagrams and watch those

fm is also good to look at

[odd shale]

🐢

[obtuse rapids]

can i use Google Colab but no sing in

[shell token]

What happened to the RVC that could pull from youtube?

Like automatically make and instrumental and all that

[proper shale]

basically, either broke down or got restricted

https://weights.gg does that though

[shell token]

Oh neat

Can I use Models from here?

[proper shale]

technically yes because it has the models from here

when doing the cover just search for the name of the model and it'll pull it from here

[shell token]

Ok yeah just searched it up

They do have them

Thanks!

[tawny sundial]

how to make this faster?

[proper shale]

Yw! :)

[proper shale]

That is veeeery slow, how long is your file?

[tawny sundial]

3:13

[proper shale]

Huh... what's your other settings?

[tawny sundial]

all stock besides mangio-crepe as pitch extraction

[proper shale]

What's your GPU?

[tawny sundial]

Intel(R) UHD Graphics 600

[shell token]

My guy you're going to be here forever with a GPU like that

[rare goblet]

Ayo? @shell token level 1 !!!

[surreal magnet]

is 0.15 sample length for RTX3080 enough? or can it go lower ?

[proper shale]

Yeah I wouldn't really recommend running locally then

[tawny sundial]

my new pc comes in like 4 days

[rare goblet]

Ayo? @tawny sundial level 3 !!!

[tawny sundial]

but I am patient

[surreal magnet]

tried to lower sample to 0.10 but it goes crackle when playing games

[tawny sundial]

you recon pm would save me some time?

[proper shale]

for gaming it's a good idea to lower settings n stuff in game

[proper shale]

id probably just use colab atp

pm is horrible

[shell token]

Quick Question, heard about that you could use Ai voices in real time, where can I do that?

Haven't fucked around with this AI stuff in a while

[proper shale]

https://rentry.co/VoiceChangerGuide

Okada's pretty good

If you have a decent GPU you can run it locally

[shell token]

Ohhh neat

Thanks

I got a 4060, is that fine?

[surreal magnet]

i see, no wonder it's just a random occurence of crackle, thanks

[surreal magnet]

4060 is good i think, saw some people run it on 3050 just fine

[pastel oak]

🥬

[proper shale]

Yup, pretty decent

[idle cedar]

how do i make the link into mega or drive from downloading it from hugging?

a lot has changed since i used this last idk if im just stupid or what

[proper shale]

you wanna upload to Huggingface?

[odd shale]

🐢: bites the lettuce

[idle cedar]

no just here

on colab step 2

to download the model

[proper shale]

oh, i think you're using an outdated colab

that step is broken, basically

https://docs.aihub.wtf/ has the recent Colabs like Applio in the RVC > Cloud section

[idle cedar]

Ah thank you bro really appreciate it!

[rare goblet]

Ayo? @idle cedar level 1 !!!

[proper shale]

Yw! :)

[arctic parrot]

hey guys I want to use the voice changer in Valorant is that possible? If yes please tell me how

[low shard]

You need to sign in, it's required, just make a google acc

[proper shale]

yeah, just get a virtual audio cable

https://rentry.co/VoiceChangerGuide explains how to get them all set up properly

[thorn musk]

every time I start putting emotion into my voice the changer starts to freak out rlly badly

[arctic parrot]

thanksss

[thorn musk]

is there aa way to fix this?

[proper shale]

What's your settings rn?

[proper shale]

You're welcome! :)

[proper shale]

Lower extra to like 1 or something - I've seen glitches can occur with extra over a certain threshold

[thorn musk]

that seems to have done the trick ty

[thorn musk]

what other tweaks do you recommend btw to make it bettr

[proper shale]

mess with index (for better accent) and pitch settings i suppose

[teal dome]

anyone notice if the quality of conversions concerning ilaria and applio are a little worse on applio’s end? i compared the same vocal file from an ilaria conversion to an applio conversion with all the default settings on both (besides the index ratio being 0.75 for ilaria, which is the default for applio). the main difference i’m hearing is applio’s lack of ability to pronounce “shhh” in words like “situation”

it also struggles pronouncing the “eck-shun” part of “projection” and says it like… “projectsion” with a sharp sss

like, project-sun. really curious if there’s an extra setting in the former that needs to be switched on or changed in applio

[analog obsidian]

same model, same pitch extraction, and same everything?

[teal dome]

yep

[analog obsidian]

is volume envelope set to 0 in both forks?

[teal dome]

for applio it’s set to 1, i don’t remember the old ilaria rc’s, just that i didn’t change anything in it besides the index ratio

[analog obsidian]

change both to 0 and hear the results

they both are using latest mainline as base, they're just ui edits

[teal dome]

will do

nope, pronunciation of “situation” is still weird

[analog obsidian]

then i have no idea o.O im not part of the applio team so yea idk

pronunciation is handled by index file

[teal dome]

darn,,, well thanks for the help anyway

[analog obsidian]

i mean the way of pronouncing consonants and vowels

[teal dome]

i’m trying out different settings and values little by little to see if anything sounds closer to the old ilaria file

[analog obsidian]

try decreasing index rate in applio

so your model* will mimic the source audio way of saying words

[delicate kettle]

I have a friend who uses a Mac and is trying to use real time voice changer and it’s not working for him

[teal dome]

alright i’ll try 0.33, i know the newer ilaria fix uses that value as default

[analog obsidian]

[analog obsidian]

is your friend using okada's alpha version? alpha versions are unstable now

[delicate kettle]

I’m pretty sure he is

[teal dome]

0.33 definitely made a difference, it pronounces the “shhh” better but now the voice itself is far more different sounding in general

is it known or recorded somewhere what the old ilaria colab settings were?

[analog obsidian]

this is the latest alpha version https://huggingface.co/wok000/vcclient000/blob/main/vcclient_mac_2.0.23-alpha.zip

[teal dome]

(the ui was pinkish)

[analog obsidian]

and this is the latest stable version https://huggingface.co/wok000/vcclient000/blob/main/MMVCServerSIO_mac_onnxcpu-nocuda_v.1.5.3.17b.zip

for mac

[analog obsidian]

first tell him to try this version, if this isnt working, then he can try the latest stable (1.5.3.17b)

[analog obsidian]

sorry i don't know this

try playing with the index file rate

seems to be related with that

0.75 is a bit high for a default index rate
0.5 is stable

[delicate kettle]

He’s downloading it again

[teal dome]

oooh i wonder if that’s it

something i noticed as well with the applio conversion, it seems like the compression or gain on background noise is heightened

[teal dome]

setting the volume envelope to 0 definitely brings the noise down, but the voice is still a little different

so bringing volume envelope up to a little over 0.20 gets the background noise to roughly the same volume level

comparing it with the og conversion vocals and it sounds like the og vocals are clearer or a bit more upfront/raspier

setting consonants to 0.4 while having volume envelope a little over 0.20 definitely makes the breaths before singing sound much more realistic, and closer to the ilaria conversion…

and it pronounces “situation” correctly as well i think i found the closest settings

[distant ether]

somebody please send link to version that will work on 1650

[proper shale]

realtime voice changer?

https://rentry.co/VoiceChangerGuide should work probably

else try rvc realtime https://rentry.co/RVCRealtimeGuide

for converting voices in audios (AI cover stuff), also training: https://docs.aihub.wtf/rvc/local/applio/

[distant ether]

thanks

yes for realtime

[distant ether]

is it realtime ?

[proper shale]

yeah, W-Okada and RVC Realtime are all for realtime

okada is easier to use overall

[delicate kettle]

My friend downloaded vb virtual audio and his voice is coming through on games

[rare goblet]

Ayo? @delicate kettle level 4 !!!

[proper shale]

Make sure they've got the right input/output settings:

Input in the games should be the CABLE

Output in W-Okada should be the CABLE as well

[delicate kettle]

Would it be the same if he has Mac

@proper shale

[proper shale]

Yep

[delicate kettle]

For some reason his doesn’t say cable

[proper shale]

If the game doesn't allow you to change inputs, ask em to change the default input to the cable

[proper shale]

Oh yeah

It says VB Cable

Should be the same regardless

[delicate kettle]

For some reason his voice doesn’t come through

[slim wadi]

question. why are most models only trained on 10 ish minutes of audio or less? would it not make sense to train one on 30+ minutes of one or more? does it cause lag ect?

[proper shale]

Can he send his settings (both in-game and in Okada)?

[left sparrow]

What's the best tool for using RVC..? I'm trying to replace the lyrics of a song, so preferrably with TTS capabilities.

I just started messing around with RVC and I don't know any of the settings yet. I have encountered the well known "squealing" audio which means I've done something wrong, but I'm too much of a noob to figure out what exactly that is.

For the moment I've mostly played around with "RVC-GUI-pkg" and trying to make a vocal track sound like Serj Tankian with that software.

I know a bit about Stable Diffusion but other than that my AI knowledge is a bit limited.

[proper shale]

I wanna see what's up

[proper shale]

I suppose Applio is a good tool to use.

https://docs.aihub.wtf/rvc/local/applio

[delicate kettle]

That’s his settings

@proper shale

[distant ether]

There is issue VB cable says that he is instaled but i dont see him in output devises

[rare goblet]

Ayo? @distant ether level 1 !!!

[proper shale]

I suggest increasing chunks to 192.

Also, I hope the Roblox stuff is correct input/output wise, but if it isn't I believe the first one is input, and that should be VB Cable

[proper shale]

Click on "refresh" in the Input part, it should appear

Also you did need to restart the PC for it to appear as well

[distant ether]

already restarted

[proper shale]

[distant ether]

alternative

virtual cables

[proper shale]

I'd probably remove it and use VAC by Muzychenko instead yeah

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

[delicate kettle]

Is there an alternative one for Mac that my friend can try

[proper shale]

Voice changer? Sadly no

[delicate kettle]

For some reason it just doesn’t work for him

We’re still trying to figure it out

[proper shale]

[midnight fern]

Yo how do i make ai covers with existing models on ios (link)

[proper shale]

use https://weights.gg

easy to use and makes covers pretty nicely

[midnight fern]

thx

[midnight fern]

Wait can i use the existing models from here

[proper shale]

Weights gets models from here, so just search for the specific name of the model and it should be there!

[midnight fern]

Is it ilaria

[dusty frigate]

I'm new to RVC. When making audio to train. Do I use 1 kind of pitch and tone. Or do I use vast amounts like high pitch. Low pitch. Sounding natural ofcourse

[proven hill]

mh?

[brittle wing]

When I have a wav file with a friend's voice, read on how to make a model in GPT Sovtis?
yes?
Do I need to clear the vocals from the WAV first?

[proper shale]

Weights is separate from Ilaria RVC so....

[proper shale]

Yeah you do need to clean it if it's not clean

[brittle wing]

mp3 cleared vocals?

wav cleared vocals?

[proper shale]

Get the most amounts of variations you can

[proper shale]

wav

[brittle wing]

pls help pv

I can't send photos

[proper shale]

I don't really like helping in DMs, sorry

But if you hit lvl 1, you get pic permissions

[brittle wing]

..

[brisk otter]

I've got a quick question, with a 4070, what should the chunk and extra be for the new Okada?

nvm

Got the wrong chanel

Sorry guys

[proper shale]

I suppose chunks will probably be lower, but considering it's probably still on alpha... you never know what you'll get

[proven hill]

112 / 16k

[proper shale]

it's alr

[brisk otter]

Unfortunately it looks like this, which is where the confusion comes from

[brittle wing]

oh

I want to achieve a friend's voice with wav and clean up the teeth so that the voices in the background are not only his voice to train and make a model in GPT Vits

[proper shale]

roformer will probably take care of the background noises n stuff

if it has echo/reverb use the vr arch de reverb de echo stuff

[brittle wing]

So Roformer first?

[sterile fable]

Any good tools to separate podcast audio?

[brittle wing]

@proper shale What MP3 to WAV converter do you recommed?

[proper shale]

... that's the problem

You need a WAV source to make a good model, if you can't get that use Audacity ig

[brittle wing]

So you can't do a converter?

mp3 to wav?

[proper shale]

Separating per speaker? https://huggingface.co/spaces/Delik/pyannote-speaker-diarization-3.1 should help

[proper shale]

You can, it won't make quality better though

[brittle wing]

best converter mp4 to mp3?

because I have in mp4 and I want to convert to mp3

[sterile fable]

thanks ya if its a video call podcost with 1 host and 1 guest

[rare goblet]

Ayo? @sterile fable level 1 !!!

[proper shale]

Hm... Audacity should probably do that as well, I guess

[brittle wing]

how?

[proper shale]

Just put it in Audacity, export as a mp3?

Else I'd probably just use an online converter ig

[sterile fable]

you here

[brittle wing]

https://online-audio-convert.com/en/mp4-to-mp3/

recommend? @proper shale

[proper shale]

yeah use that ig

[brittle wing]

MJ

[brittle wing]

@proper shale pls help

how to convert wav ?

mp3

yes?

[glad zealot]

It's an rvc model, which should work on vcc

[scarlet wedge]

how to solve this error?

[proper shale]

Ilaria shall solve this soon.

[proper shale]

yep

[glad zealot]

Oh it still on that error

[scarlet wedge]

thanks!

[glad zealot]

What are you planning on using it for?

You can try alternatives

[scarlet wedge]

a guy in the Brazilian AI HUB server is with this error and i don't know how to solve it.

I Already have Ilaria RVC Duplicated in my huggingface account.

[glad zealot]

No like are you using it to make a model, a cover, tts?

[scarlet wedge]

i use it to make Inference. And i already know how to make a Voice model, thanks!

[glad zealot]

If the huggingface doesn't work you can always try the colab

[scarlet wedge]

yeah

[paper dock]

should i keep training

500 epoch

g/total

sounds nice but could still be more trained idk if its just me

[scarlet wedge]

I Have a question. Can i make a good voice model of a character who spoke very little and that I would only get a few seconds of dataset? like the birds of angry birds? i already know how to make a good voice model and see the tensorboard to use the correct amount of epochs, but i don't know how to make of game characters like this.

[proper shale]

just train on low batch size

should do it

[proper shale]

mel still going, keep training

[scarlet wedge]

How much batch size do I use for how long? Do you have a document showing this?

[proper shale]

2 for veeery short datasets, like below 30s-1m

4 for anything below 5 mins

8 for 5 mins and above

[scarlet wedge]

thanks!

[brittle wing]

good?

@proper shale

.flac

[proper shale]

yeah sure

[brittle wing]

run?

[brittle wing]

.

[proper shale]

yeah sure

[brittle wing]

good

run"

?

[scarlet wedge]

@proper shale
i was searching the wiki of angry birds games and i found only audios with the .mpeg file format. Should i convert it to .wav to use it to create the model or i use it at .mpeg format anyways?

[rare goblet]

Ayo? @scarlet wedge level 3 !!!

[scarlet wedge]

thanks

[proper shale]

Yeah try that

[proper shale]

Just proceed onto the next steps, I'm gonna sleep soon so yea I'm not gonna answer all the time

[brittle wing]

I want to achieve a model of a friend that will be used for TTS and singing

[scarlet wedge]

wav or mpeg? xd

[proper shale]

WAV

[scarlet wedge]

ok, thanks

[brittle wing]

well the options are checked?

Do I leave them unchecked?

@proper shale

[proper shale]

run as is

if you don't know what it is don't mess with it...

have you read the disconnected guide btw

[brittle wing]

What to set here?

save_frequency 15?

[proper shale]

_ _

i strongly recommend doing so

[brittle wing]

I've read nothing helps, I'm just writing that it's better to leave 15 if I don't know

[proper shale]

Yeah honestly

You can probably set total epochs to 100, to ensure that you'll be able to export model once it's done

Then you can return later

[brittle wing]

So give 100 in epochs?

The total amount of epochs for the model. But since we'll use TensorBoard, use an arbitrarily large number like 2000.

[rare goblet]

Ayo? @brittle wing level 12 !!!

[brittle wing]

oh yea

[viscid moss]

just scalars

[brittle wing]

I can only custom scalars

aa

[viscid moss]

search for g/total

[viscid moss]

unmark "ignore outliers....."

and can u send a complete pic of the tensor

[brittle wing]

where?

ok

[viscid moss]

set smoothing to 0,95

[viscid moss]

that's fine

[brittle wing]

already

500 end?

epoch?

or not?

[lone gull]

whats the best way to create config.json files for all these rvc models so i can use it in vits?

[brittle wing]

@viscid moss

already
500 end?
epoch?
or not?

[viscid moss]

did it reach 500?

[brittle wing]

no

[viscid moss]

on this graphs, search for g/total and show me

[brittle wing]

are you using mainline mangio

[viscid moss]

seems good, maximize them

[viscid moss]

Mainline RVC

[brittle wing]

in colab?

[viscid moss]

U can't use rvc models on svc

[viscid moss]

yep

[brittle wing]

colab deletes all my progress after I missed the 1 hour afk timer...

300 epoch training wasted

[viscid moss]

5k seems to be a point

[brittle wing]

what?

XD

[brittle wing]

5k in the graph

on x axis

joke?

WHATR

500

good

@viscid moss

5k

[viscid moss]

-colab

[azure marsh]

☁️ Google Colabs

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

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

[brittle wing]

clear

clean*

[brittle wing]

I'm using applio it's shit, disabled after 1 hour and I lost 300 epoch progress

in colab

[viscid moss]

fine, now u need to run the next cell.

[brittle wing]

.

clean

They tell me to clean something

[viscid moss]

what the heck

[viscid moss]

hmm ignore that

[viscid moss]

.

[brittle wing]

already

[viscid moss]

check ur g drive

[brittle wing]

what next?

are you trolling @brittle wing

[rare goblet]

Ayo? @brittle wing level 3 !!!

[viscid moss]

Hmmm i can't see the model

[brittle wing]

Where did it write it down?

[viscid moss]

dowload this file:

that's was the best point

[brittle wing]

already

download

[viscid moss]

fine, let's wait

this file too

downloat it

[brittle wing]

added?

How to prevent colab from shutting down every hour? It takes 2 hours to do 200 epochs

I need to do 500

off internet

time training

?

because it crashes your GPU

limit GPU

[robust sun]

New rvc working link

[brittle wing]

Colab has a limit

GPU

@viscid moss

Bro I have 300 compute units left, I should be fine to run A100 for like 30 hours

[viscid moss]

added

[brittle wing]

already

what next?

[viscid moss]

so now u have both file in ur pc rigth?

[viscid moss]

function ClickConnect() {
var iconElement = document.getElementById("toggle-header-button");
if (iconElement) {
var clickEvent = new MouseEvent ("click", {
bubbles: true, cancelable: true,
view: window
});
iconElement.dispatchEvent (clickEvent);
}
}
setInterval(ClickConnect, 6000);

[brittle wing]

desktop?

[viscid moss]

this

[viscid moss]

yep

[viscid moss]

3 hours is the limit anyways

[brittle wing]

huh

no

is this some ublock origin code

[viscid moss]

https://docs.google.com/document/d/1Pr-AZndodmWgsbOeuHQU4IrgbatFgYc1ChOq_ZAf_5s/edit

Check this guide

[brittle wing]

already

[viscid moss]

Almost

[viscid moss]

nice

[viscid moss]

make a .zip with that both files inside

[brittle wing]

create a zip file and insert those files?

yes?

[viscid moss]

yep

[viscid moss]

What program will you use to use the model?

[viscid moss]

fine

[brittle wing]

TTS and sing

[viscid moss]

So... locally?

or colab?

[brittle wing]

yes

colab

[robust sun]

Getting error

[brittle wing]

developer fixing

soon work

[robust sun]

Okay when its updated dear

[brittle wing]

i dont know

ask Nick088

[viscid moss]

pls don't ping people

[brittle wing]

ok sorry

[viscid moss]

u need to upload the model to HF

[viscid moss]

https://docs.aihub.wtf/essentials/voice-models/#uploading-to-hugging-face

[brittle wing]

And how to make a private repository?

acces with me

[viscid moss]

For now use my fork of Ilaria RVC:
https://huggingface.co/spaces/Eddycrack864/Ilaria_RVC

[viscid moss]

u can't. If u wanna use the model on colab u can't make it private

[brittle wing]

And on which platform can I use?

private

locally?

yes>

?

[viscid moss]

installing some local app for rvc

[brittle wing]

okay