mrblueblue/moodmusic

Natural Language Processing that turns messages into music and visuals.

Mood Music

###Turning the Twitter stream into music.

Mood Music uses the excellent Sentiment Library by Anrew Sliwinski and the Web Audio API to generate music and color based on the Twitter stream.

Running Locally

To get started, you must first register your own Twitter application and create a config.js with your API keys. See config.example.js for more details.

After having set up your API keys, you can grab the needed dependencies:

npm install

And then serve the application:

nvm use iojs
node app.js

Go to 127.0.0.1:3000 in your browser to see the app in action!

App Structure

├── app.js                         # Express server
├── server.js                      # Server entry point file
├── public/                        # Static files to be served
└── client/                        # Application source
    ├── index.js                   # Client entry point    
    ├── components/                # React components folder
    │   └── App.jsx                # Main React component
    ├── utils/                     # Utility funcitons
    │   └── createAudioContext.js  # Wraps component in AudioContext component
    └── config/                    # Configuration files
        ├── colors.js              # Configs for colors
        └── music.js               # Configs for notes, frequencies, etc.

React Client

The main React component is App.jsx. Instead of using mixins, the component is wrapped in a higher-order component that passes down the HTML5 Audio context and methods. This is done in the createAudioContext function.

io.js Server

Server code is written in ES6 and is run by io.js. Mainly, I just wanted to use destructuring 😃.