Aicapella-CNN-Based-Voice-Isolator

A Convolutional Neural Network Based audio vocal separator.

This is an extremely simple tool for separating vocals from its music, completely localized for web operation.

UNet Architecture

Running in Notebook /Colabs

Run in Google Colab or any other notebook from the given .ipynb at /src/

  minorAICAPELLAgpu.ipynb

For testing,set up model and proper directory path in source code

  testing.ipynb

Run Locally

Clone the project

  git clone git@github.com:shresthasubham/Aicapella-CNN-Based-Voice-Isolator.git

Go to the project directory

  cd Aicapella-CNN-Based-Voice-Isolator

Create a virtual environment

 python -m venv venv

Activate the environment. On Windows, the command is

%cd%/venv/scripts/activate

and on Linux and Mac, the command is

 source ./venv/bin/activate

Install dependencies

  pip install -r requirements.txt

Go to the frontend directory

  cd src/front

Execute

  python app.py

and wait for the local browser window to open and Enjoy.

Screenshots

Documentation

Documentation

Drive link

Drive

References

• Singing Voice Separation With Deep U-NET Convolutinal Networks
• Youtube-Valerio Velardo