The Sound of Data

Charts are great. Bar charts, line charts, scatter plots. They show you what’s there. But they don’t make you feel it.

Sound does something different. When you map a dataset to frequencies and rhythms, your brain processes it in a new way. Patterns that are invisible in a spreadsheet become obvious in a melody.

Sonification 101

The idea is simple. Take a data point. Map it to a musical parameter. Pitch, volume, duration, timbre. Play it. Repeat for the next data point.

Temperature data becomes a rising melody as summer approaches. Stock prices become a jittery rhythm that calms during weekends. Heart rate data becomes a drum pattern that speeds up when you exercise.

What I Learned

The hardest part isn’t the code. It’s choosing the right mapping. Map everything to pitch and it sounds like noise. Map too little and it sounds like a single note.

The sweet spot is 2-3 parameters. Pitch for the main variable. Volume for the secondary. Maybe a filter sweep for the third. Your ears can track these independently. Add more and it becomes mush.

Try It Yourself

Grab any CSV. Open Tone.js. Map column A to a synth’s frequency. Hit play. You’ll hear your data in a way you’ve never experienced it before.

The numbers will sing.