SEGMENT AUTOMATIC EXTRACTING SYSTEM FOR EXTRACTING SEGMENT IN MUSICAL COMPOSITION, SEGMENT AUTOMATIC EXTRACTING METHOD, AND SEGMENT AUTOMATIC EXTRACTING PROGRAM

摘要

A segment automatic extracting system capable of providing various applications with an impressive segment of a musical composition as metadata on the musical composition by automatically extracting a portion of the musical composition which is considered as having a high possibility that the portion is widely known by general users irrespective of the number of appearances in the musical composition. A segment automatic extracting method and a segment automatic extracting program are also disclosed. A first acoustic signature creating section (101) creates an acoustic signature representing the feature value of the acoustic signal of each of contents. An important segment extracting section (102) creates an acoustic segment signature representing the frequent appearing feature value found by searching all the created acoustic signatures. A second acoustic signature creating section (111) creates an acoustic signature from the musical composition signal. A common segment extracting section (112) judges whether or not each acoustic segment signature agrees with a part of the acoustic signature of the musical composition signal and outputs time information by which the acoustic signature of the musical signal agreeing with the acoustic signature can be specified.