YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

RVC-v2-easy-gui-tutorial

(before the tutorial) Please donate to Luis Santillan Rejekts, the creator of RVCv2 in either of these 2 places:

https://ko-fi.com/rejekts

https://www.paypal.com/paypalme/lesantillan

45% of $100 goal

Only $45 were donated to Luis. Once he has enough to pay for one of his normal job work days, "I can spend a whole 8 hour day coding to fix any issues and add new features to my projects!". I assume these features may be audio enhancers, background noise removal, or the ability to change an entire youtube video at once. But regardless, please donate, as he didn't get much.

Tutorial

Setting up the application (Google Colab):

This is simply an easy tutorial of RVC V2, using Google Colab. You WILL need to make an account on Google.

  1. Go to this Google Colab Notebook: https://colab.research.google.com/github/PhoenixStormJr/RVC-v2-easy-GUI-glitches-fixed/blob/main/EasyGUI_Inference_Only_%F0%9F%8E%AE_10_02_2024__12.ipynb
  2. Click "File" at the top left, and "Save a copy in Drive" (This solves the timeout message.)

Download a model to use for RVC V2 (Google Colab)

  1. First, go to https://huggingface.co/models .
  2. Inside "Filter by name" enter the name of the model you'd like followed by RVC.
  3. For example, I want Mario from Super Mario. So I type "Mario RVC"
  4. A list of models came up. I clicked the first one. https://huggingface.co/Xhepyxopila/MarioRVCModels
  5. Go to Files and versions.
  6. Right click the download button next to a .zip file of the model you want. IT MUST BE A .zip FILE OR ELSE THE MODEL FLAT OUT WON'T WORK!!!
  7. Click "Copy Link Address"
  8. Go back to RVC V2 Google Colab Notebook.
  9. Paste the link under "url:"
  10. Name the model whatever you like, since I searched Mario, I'm naming mine Mario.
  11. Click the play button (sideways triangle) Note: The FIRST time it will Install RVC, but the second time it will go faster. Give it around 3-5 minutes.
  12. wait until the bottom bar says something like:

"""

Downloading model:

https://huggingface.co/...

INFO: Done

Downloaded model!

"""

Use a model for RVC V2 (Google Colab)

  1. TYPE the name of your model in "model_name" (It will automatically detect the index path and model path.)
  2. Select the method you want to use to create the audio "create_audio_method" (upload_file uploads a file and record_audio uses your mic to record audio... kinda obvious)
  3. Under "Optional: You can change the pitch here or leave it at 0." self explanitory.... changes... pitch... this is useful for boys trying to sound like girls, or girls trying to sound like boys.
  4. Click the triangle again to run the cell. It'll run and convert the audio! That's all!

Setting up the application (broken)

This is simply an easy tutorial of RVC V2, using huggingface. You WILL need to make an account on huggingface.

  1. go to this website: https://huggingface.co/spaces/Clebersla/RVC_V2_Huggingface_Version

alternately go here:

https://huggingface.co/spaces?sort=trending&search=RVC+V2

and click on one of the options called RVC V2.

  1. click the 3 dots in the top right hand corner
  2. click Duplicate this space
  3. Although "Space name" does not really matter, I suggest naming it "Your username RVC V2" or whatever really
  4. Under Space hardware, if you don't mind the incredibly slow speeds, use "CPU basic * 2vCPU * 16GB FREE". Otherwise, buy an upgraded version for faster voice cloning.
  5. Click "Duplicate Space"
  6. Wait X amount of time. (I don't know how much time, I just know it's a long time on the free version. About 10 minutes... again, buy the better version for faster run times)
  7. NOTE: ===== Application Startup at 2023-10-30 01:54:00 ===== does NOT mean it's finished... keep waiting...
  8. Once it is finished, you will see the application like normal.

If you closed your browser (broken)

  1. If you clicked the X button and closed your browser, to find the application again go back to huggingface.
  2. If you are not logged in go to https://huggingface.co/https://huggingface.co/login
  3. enter username and password

Alternately if you ARE logged in go straight to https://huggingface.co/

  1. click your username bubble at the top right
  2. click profile
  3. it's the space called "RVC V2" at the top.

Download a model to use for RVC V2 (broken)

  1. First, go to https://huggingface.co/models . It's reccommended NOT to close out of the application. If you do refer to "If you closed your browser" section
  2. Inside "Filter by name" enter the name of the model you'd like followed by RVC.
  3. For example, I want Mario from Super Mario. So I type "Mario RVC"
  4. A list of models came up. I clicked the first one. https://huggingface.co/Xhepyxopila/MarioRVCModels
  5. Go to Files and versions.
  6. Right click the download button next to a .zip file of the model you want. IT MUST BE A .zip FILE OR ELSE THE MODEL FLAT OUT WON'T WORK!!!
  7. Click "Copy Link Address"
  8. Go back to RVC V2 application. Refer to "If you closed your browser" if you closed out of it.
  9. Click "Download Model"
  10. Paste the link under "Enter the URL to the Model:"
  11. Name the model whatever you like, since I searched Mario, I'm naming mine Mario.
  12. Click "Download"
  13. wait until the bottom bar says "Success."

Use a model for RVC V2 (broken)

  1. Go back to Inference.
  2. Click "Refresh" next to "1.Choose your Model."
  3. Click the arrow pointing down next to the blank area in "1.Choose your Model."
  4. Click the model we downloaded earlier
  5. Either drag and drop an audio file from your PC/Mobile device, (yes this also works on android and apple), or record your own voice. I'm going to record, so I click the record button.
  6. Under "Optional: You can change the pitch here or leave it at 0." self explanitory.... changes... pitch... this is useful for boys trying to sound like girls, or girls trying to sound like boys.
  7. Click "convert"
  8. This will take at least a minute to convert the voice. Expect even LONGER waits for more audio, mine was only 6 seconds.
  9. If the pitch is off, simply change the pitch, and click "convert" again. My pitch was off.
  10. Click the 3 dots next to the audio and download it. OK that's it!

Original RVC v2 databse:

https://huggingface.co/Rejekts/project

Local installation on Linux (MY OWN DEBUG STUFF):

Alright, so the downloading tab is broken, I will have to make my own version...

Local installation on Windows (UNFINISHED):

Will add a tutorial here as soon as I install it on Linux

You can install this in the mean time: https://www.tryreplay.io/

Local installation on Mac (MY OWN STUFF):

Mac is impossible to figure out, I found this app for Mac computers, but I do not own a Mac computer, so have fun I guess: https://www.tryreplay.io/

Figure it out yourself, mac sucks. I don't own a mac and I can't figure out how to run it on a virtual machine. It sucks.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Collection including PhoenixStormJr/RVC-V2-easy-gui-tutorial