DJ-Tweets

• 

  landing page

• 

  player ready

• 

  visualizer view

Inspiration

We wanted to build a fun app that would connect people with the Magenta.js library.

What it does

This site pulls a Twitter user's tweets, analyzes them with sentiment analysis, and then generates melodies based on the dominant emotion of each tweet. Once we have generated a short melody for each tweet, we pass the combination of those to MusicRNN to continue the sequence. Once that has loaded, we combine the two note sequences and play it in the browser, providing a visualizer, the tweet text, and an emoji representing the analyzed emotion.

How I built it

Magenta.js, MusicRNN, Tone.js, Tonal.js, p5.js, javascript, node.js, express.js, Twitter API, ParallelDots sentiment analysis API, VS Code... built among friends :)

Challenges I ran into

Learning the Magenta.js library, incorporating all the different libraries and getting them to work together

Accomplishments that I'm proud of

We got it to work more or less

What I learned

New libraries

What's next for DJ-Tweets

Work out some of the bugs

Built With

• express.js
• javascript
• magenta.js
• musicrnn
• node.js
• p5.js
• sentiment-analysis
• tonal.js
• tone.js
• twitter