metadata

license: afl-3.0
language:
  - en
  - ja

This model is trained by using so-vits-svc-fork

Hardware used:

CPU: AMD Ryzen 9 3900X
RAM: 64 GB
GPU: 3090 24GB

Acquiring the dataset

Software used:

ultimatevocalremovergui
Audacity

Step 1

Find videos, music, podcasts or whatever that contains the voice you want to make a model of.

Step 2

Snip out the parts of the videos/music you want to use for the dataset. The clearer the audio, the better. This means no background noise whatsoever. Each file must be a maximum of 10 seconds!
You can do this via Audacity or any other software you feel familiar with.
For a decent model, you will need about 100 samples.

Step 3

If a sample has a background noise (which it will most likely have), remove it via `ultimatevocalremovergui`.

Removing background noises

Installing the requirements

Step 1

Install ultimatevocalremovergui by following the following steps:

git clone https://github.com/Anjok07/ultimatevocalremovergui
cd ultimatevocalremovergui
nano environment.yml
Fill it with the following text:

channels:
  - defaults
dependencies:
  - python=3.10
  - tk
  - pip
  - pip:
    - -r requirements.txt

Save it by pressing ctrl+x followed by Y then press enter
conda env create -f environment.yml
conda activate ultimatevocalremovergui
python UVR.py

Step 2

The software will now startup (this might take a bit). It will look like this: First we need to download a model like so:

Click on the wrench icon next to the Start Processing button.
At the top of the new window that opens, click on the tab called Download Center
Select the radio button called Demucs
Select Demucs v4: htdemucs_ft
Click the download button underneath this combobox

Now that the model is downloaded we are going to remove the background noise from our voice sample. To do this do the following:

At the top, click the Select Input button
Select your voice sample
Now click on the Select Output button
IMPORTANT! Your output should be like this: dataset_raw/{speaker_id}/**/{wav_file}.{any_format}, example: dataset_raw/sinon/wav/sample1.wav. This folder can be anywhere on your system
Select a directory where you want the processed file to appear
Now under the text CHOOSE PROCESS METHOD select Demucs
Make sure the model is selected under the text CHOOSE DEMUCS MODEL
Click on GPU Conversion to speed up the process
Now click on Start Processing and wait until it is done
After it's done, navigate to the folder you set as output and listen to it. Does it sound ok? if it does, you are now done, if it doesn't, don't use this file in your dataset

Training the model

Here is a quick explanation on how I trained this model.

Software used:

so-vits-svc-fork (The software to morph your voice)
qpwgraph (this is used to reroute the output to another process like Discord or Telegram)

Step 1

First, install qprgraph:

paru qpwgraph

Now, clone the so-vits-svc-fork repo:

git clone https://github.com/voicepaw/so-vits-svc-fork

Then, cd into the repo:

cd so-vits-svc-fork

Now, make a conda environment:

conda create -n so-vits-svc-fork python=3.10

Now, activate the conda environment:

conda activate so-vits-svc-fork

Now, install the requirements:

python -m pip install -U pip setuptools wheel
pip install -U torch torchaudio --index-url https://download.pytorch.org/whl/cu118
pip install -U so-vits-svc-fork (This will install the package inside your conda environment, meaning you can run it anywhere on your system as long as you are in your conda environment)

Step 2

Navigate to the directory where you dataset is at, for example, if your dataset is at /mnt/Shark/Projects/Sinon-Voice/training/dataset_raw/sinon/wav/ navigate to /mnt/Shark/Projects/Sinon-Voice/training run the following commands:
svc pre-resample
svc pre-config
svc pre-hubert
svc train -t

Using the model

Step 1

Now, run the program:

svcg

On the right side in the application that just opened, make sure to set the input device to default (ALSA) and the output also to default (ALSA)

Step 2

At the top, select your model and config files. These are located in your training folder at: logs/44k/

Step 3

You can now tweak some settings, for example the pitch (I recommend a value of 12 to begin with)
Turn off Auto predict

Step 4

After tweaking the settings to your liking, press the button called Infer at the very bottom to start the voice morph

Additional info

If nothing happens, take a look at the terminal and act accordingly