hecrean / isomorphic-soundscapes

losslessly convert audio mp3 into visualisations (and back again)?

soundscape

Godel Escher Bach makes the point that grooves in a vinyl records and the music that comes from them are simply different representations of the same information. They are, to use a fancy mathematical term, isomorphic. As such, we can transform one representation into the other, and back again -- without any loss of information in the process.

When we visualise music, do we want to create a lower dimensional representation of it? i.e. are we concerned about losing information in the translation process, or are we not concerned with the ability to transform the visual back into its origina (audio) state?