A time-series phrase correlation calculation system with acoustic signal processing for music media creation

In this paper, we present a system for analyzing the time-series impression change in the acoustic signal by a unit of music phrase to support music media creation. The aim is to realize an environment to support the composers' music creation by bringing out their potentially-advanced knowledge and skills. This system analyzes the acoustic signals by focusing on the phrase and the structure of music (melody, rhythm, and harmony). By calculating the correlation between the typical phrase in each music genre and the one that is included in the target music to be analyzed, the system enables the composer to grasp the impression change of music media by the unit of music phrase. The system leads to a new creative and efficient environment for music composition based on the advanced knowledge on the music phrase and structure.