Ultimate RVC – Someone Else’s Voice

Yesterday I wrote about how easy it is for someone to use someone else’s voice and looks with the help of generative AI and Ultimate RVC. I must say that I’m a bit confused of how little people seems to care, or maybe they just haven’t grasped the full meaning of the possible uses for it.

Anyway, I got a request from someone asking me if I could write a beginners guide on Ultimate RVC.

Here I will show how to install the program, how to run it and a few basic settings.

Installing RVC

First you need to install Git, if you don’t already have it.

Start with downloading the latest Git installer: Link to Git

Follow the installation instructions and finish the installation.

Now when you’ve got Git installed, decide where you want to install Ultimate RVC. Navigate to the folder you want the program to be installed . In that (empty) folder right click anywhere and click open command window here.

A command prompt window should open, and it looks something like this.

Copy the following command and past it in the command prompt window:

git clone https://github.com/JackismyShephard/ultimate-rvc

click your Enter button.

cd ultimate-rvc

click your Enter button.

./urvc.bat install 

click your Enter button.

Now the installation might take some time, depending on your system. Just let it run until it says that the installation is done.

When the installation is done, test starting the program by typing the following in your command prompt:

./urvc.bat run

click your Enter button.

Here is where I ran into some issues. Files were missing despite that the installation had completed without issues. If you have issues getting the program to run, you should write the following in the command prompt:

pip install requirements.txt -r

This will install all requirements, in case something was missing, and now you should be able to start the program using the:

./urvc.bat run

It will take some time for the program to initiate and fully start up. Once the program is actually running copy this:

http://127.0.0.1:7860/

And paste it in your web browser’s address field:

Running RVC

Now when you have the program up and running, the first thing you need to do is download a language model from the list of available models.

Once you have downloaded a model you want, it will show up under models on the first page. If it doesn’t you might need to refresh the page using F5.

The next step is to find a song (or pretty much anything) on youtube that you want your voice of choice to sing/read/talk.

When you have input the web address just press the generate button at the bottom of the website. It will take maybe 2-3 minutes until the finished recording shows to the right of the generate button.

There are some advanced settings, but you will have to play with them yourself and find out what they do.

There are also ways to train your own voice model, and again, this is something you will have to figure out yourself.

DISCLAIMER

You are yourself responsible for whatever you decide to use this program for. Make sure that the results are not illegal where you live, and that you yourself aren’t using the results for illegal activities.