spotif.ai

Type in any text, and spotif.ai analyzes your phrase and creates a Spotify playlist for you based on your phrase.

Built for nwHacks 2022 - Honorary Mention (6th place)!!

This is the repo for the backend code of spotif.ai. View the frontend code here!

Technologies

• Django/Django REST Framework
• Tensorflow
• FastAPI
• Jupyter Notebook and Google Colab
• Spotify Web API

Setup

Clone this repository.

In the root of this repository, create a virtual Python environment with python3 -m venv venv.

Activate your virtual environment by running source venv/bin/activate.

If you're on windows you'll have to run source venv/Scripts/activate.

Run pip install -r requirements.txt

Create a .env file. See .env_template for details on which environment variables you need.

Note: Since loading Tensorflow models is GPU-intensive, we set up an ngrok server in a Google Colab using FastAPI. See the colab folder. The URL generated by running this file is what is used for the environment variable COLAB_SERVER_URL.

Run python manage.py migrate.

Development

Activate the server with python manage.py runserver.

Usage

spotif.ai provides the following endpoints

POST /create

{
    "headers": {
        "Content-Type": "application/json",
        "Accept": "application/json"
    },
    "body": {
        "token": "auth-token",
        "phrase": "prompt",
        "limit": 50
    }
}

POST /playlist

{
    "headers": {
        "Content-Type": "application/json",
        "Accept": "application/json"
    },
    "body": {
        "token": "auth-token",
        "playlist_name": "name",
        "playlist_description": "Auto-generated by Spotif.ai",
        "songs": "comma-separated list of song uris",
    }
}

Contributors

Akash

Evan

Kate

Prayus