The purpose of this repo is to create visually attractive waveform representations of songs. In the long term, I'd like to combine my lastfm "scrobbles" history with my iTunes library, so that it automatically generates the waveforms for the top-played songs (as rated by "play count"). For now, it takes the audio-files stored in the ./audio/ folder, and generates a visualization that is saved to the ./out/ path.

For an "automatic" selection of parameters:

1. Place your MP3 and M4A files in the ./audio/ folder.
2. Rename the files with the song title (for overlaying purposes).
3. Run the main.py file. This will load and process the files and randomly select colors from a pre-defined pool.

For finer control over the parameters, please have a look at the main.py script and modify as needed (will make improvements on the code in the near future).

• main.py: Main plotting routine.
• fonts.py: Lists all the fonts available to matplotlib with an example.
• style.py: Fonts, colors, and plot-styles definitions.
• aux.py: Functions definitions for the main routine.
• plot.py: Colors and plotting functions

The code requires the following dependencies installed:

• pydub
• matplotlib

Héctor M. Sánchez C.