There have been many attempts in computational linguistics to study phonetic patterns in English prose (mostly as a way of studying language itself). SOUNDEX is an important and well understood algorithm used in the study of phonetics that is intended to normalize similar sounding words and phonemes.

Our idea is to use this method to visualize the "sonic coloring" of a text – we hope, using actual colors.

Possible use cases would mostly be focused on comparative study of the sonic dimensions of literary texts – across genre, gender, subject matter, time, nationality, and several other vectors of interest.

The primary investigators on this project are Stephen Ramsay and Brian Pytlik-Zillig at the University of Nebraska, but the original idea was suggested by Martin Mueller.

Catherine comment: Tanya has a similar need. I think the idea is that the text could be "translated" in other forms (part of speech, phonemes, prosody, etc,) on which the same set of tools could be used (e.g. find repetitions is any of those forms.