SongSplit

SongSplit is an App that allows you to seperate the vocals from the beat of a song thanks to a pretrained models developed by Deezer (Spleeter), this model is written in Python and uses Tensorflow.

Drag and drop your audio file then download your outputs, a live demo availaible here : http://songsplit.herokuapp.com/

Please download in media folder a demo audio file called "audio_example.mp3".
Slow Motion Dream by Steven M Bryant (c) copyright 2011 Licensed under a Creative Commons Attribution (3.0) license. http://dig.ccmixter.org/files/stevieb357/34740 Ft: CSoul,Alex Beroza & Robert Siekawitch

Login and purchase (Fake functionnality with Stripe : https://stripe.com/docs/testing) before using.

## Setup ``` git clone https://github.com/ansa-aboudou/songsplit cd songsplit pip install -r requirements.txt python app.py ``` ## Disclaimer You may meet some memory issue particularly when using the app on heroku, so please try to upload light files. We could improve this project by converting the model into tensorflowjs. ## Credits
These repositories helped me build this project, thank you guys :) :
- https://github.com/deezer/spleeter
- https://github.com/anfederico/Flaskex