Interactive Music Summarization based on Generative Theory of Tonal Music,
Keiji Hirata and Shu Matsuda

Abstract:
This paper presents an algorithm that checks for melodic similarity and a music summarization system called "Papipuun". First, we developed a data structure for representing polyphony based on time-span reduction in the generative theory of tonal music (GTTM) and the deductive object-oriented database (DOOD). A least upper bound operation is introduced to check the similarity of polyphonies represented in our method. Next, Papipuun performs quick listening in a manner similar to stylus skipping on a scratched record, but the skipping occurs correctly at punctuations of musical phrases, not arbitrarily. Papipuun can produce a music summary of good quality with a symbolic approach, reflecting the atmosphere of an entire piece through interaction with the user. In a preprocessing phase of Papipuun, a user analyzes an input piece by time-span reduction, using a dedicated tool called TS-Editor. For the real-time phase, the user interacts with the main system, Summarizer, to perform music summarization. Summarizer ascertains a piece structure by similarity checking. When the user identifies fragments to be skipped, Summarizer deletes them and concatenates the rest.