This work arises from the fine initiative of the Greek government in setting up the Institute for Language and Speech Processing (ILSP), to support the growth of language technology in Greece.

The paper introduces the computer-based audio tool MEL-IRIS, which carries out the musical analysis from input in the form of musical instrument digital interface (MIDI), wave (WAV), or MP3 files. The use of the RChromatic Index here is not to be confused with its more familiar use in the mathematical theory of the coloring of graphs. The authors use the phrase in an attempt to classify music from many different ethnic backgrounds; the nature of each melodic cluster is represented by a sequence of colors as we progress through a musical item. This can be related to the attempts to define the style of an author in terms of a statistical analysis of word usage, with the added advantage of the presentation of results in attractive colors.

I believe that this work shares the limitations of all current attempts to discover the meaning of words or music through the statistical analysis of the final form of the creation. When we read or listen to a piece, we bring a vast knowledge of other works to help us perceive its structure. As we read or listen, the work triggers reactions unique to each individual. How can we model such a process?