Skip to content

Latest commit

 

History

History
50 lines (39 loc) · 2.46 KB

README.md

File metadata and controls

50 lines (39 loc) · 2.46 KB
Raw

Spotify Music Genre

This repo uses Python with the Spotify API and OAuth2 to generate a playlist of songs based on the genre and creates a new playlist for that specific genre.

⚠️ WARNING ⚠️

This project is still a Work In Progress, which means there will be a lot of bugs and unexpected things.

Feel free to experiment with this repo and report any issues you find.

PRs are always welcomed.

Credits:

This is taken from GitHub (@cetinsamet/music-genre-classification)

All credits go to him for making this amazing repo possible.

Deploying Serverless (AWS Lambda):

You can use @proguy914629bot/smg-serverless-lambda. This repo is optimized specifically for serverless genre classification.

*Note: The Spotify Computing is still hosted locally, but the AI (PyTorch) to classify genres is hosted on the AWS Lambda Function.

A live demo:

A live demo of this project is available here. The original version of that playlist is proguy's playlist.

Setup:

The below steps assumed you already have fulfilled the following requirements:

  • You have installed Python3.10 or above with pip in your computer.
  • You have Git installed in your computer. (Most OS comes with git built in, especially Linux distros)
  • A Spotify API Client ID and Secret
  1. Clone this repository (it might take some time depending on your internet speed).
git clone https://github.com/OpenRobot/spotify-music-genre
  1. Install the required dependencies (Note. It is recommended to use a venv instead).
pip install -U -r requirements.txt
  1. Copy config.example.py to the root folder of the repository, named config.py.
  2. Insert all the values for the essential variables. This includes:
    • SPOTIFY_PLAYLIST_ID
    • SPOTIFY_CLIENT_ID
    • SPOTIFY_CLIENT_SECRET
  3. Run the script.
python3 main.py
  1. Follow the instructions in your terminal.

Downloading the default dataset:

If you are searching for the dataset used here and you would like to download it, you can find it here