Transforming time series data to sound. An attempt was made.
Turns out, using data to make music is a lot harder than it sounds (no pun intended), especially without an intimate understanding of what makes a Sine wave special. Still, it does something and doesn't sound too terrible, though it doesn't show a trend in quite the same way as I had hoped.
To run:
- Uses Pyo to Pyo to make the sweet music. Annoyingly, they don't have a ready made binary for Python 3.7, so you need to compile manually.
- requirements.txt holds the rest of the dependencies.
- All a lot easier if you do it within a Virtual Environment
- Run auralise.py (don't blame me if you have to manually kill the process if it decides to keep singing at you once its finished)