aj-spec / emotion-based-music-recommendation

Design, implement and test the effects of a music recommendation system that utilize user emotional data and music preference to recommend songs complimentary to their affect.

Emotion-Based Music Recommendation Docs

Getting Started - Clone and Setup

• Have Python3 and NodeJS installed.
• Acquire developer access keys to the following APIs: Twitter, Spotify, & IBM-Watson -Clone the repository into a new folder called 'ebmr', then move into the 'ebmr' folder and open the directory in your preferred editor

git clone https://github.com/tamargoadam/emotion-based-music-recommendation ebmr
cd ebmr
code

• Change the environment variables in 'sample.env' and rename the file to '.env'
• Install pipenv and then use pipenv to install the requirements for this project.

pip install pipenv
pipenv install

• Move into the client server directory and then install the front-end dependencies.

cd client
npm install

Running the Server

• Within a terminal window, navigate to the project's root directory, 'ebmr'.
• Start the backend Flask server from within pipenv's shell virtual environment.

pipenv shell
python server/api/routes.py

• Open a new terminal window, and navigate to the project's root directory.
• Move into the client folder, and then run the front end server in development mode

npm run dev

• At this point, you will now be able to access the service and create playlists from localhost:8080/

Check your Pipfile

• Within Pipfile, alter the python version to match yours.

[requires]
python_version = "3.x"

API & Framework Documentations

• Spotipy API
• Tweepy API
• IBM-Watson Tone Analyzer API
• Preact, 3kb lightweight alternative to React