engleski

Computational modelling of segmentation processes in unaccompanied melodies

Segmentation processes in unaccompanied melodies is approached by use of the software MusickTracker, developed by authors, thus avoiding the use of notes inherent to Western Tonal Music. MusickTracker takes a digitally recorded music signal (.wav) as its input and calculates the values of pitch, loudness and timbre indicators in course of musical sound with the timestep of about 20 ms. By applying the MusicTracker to a fragment of Shakuhachi music it is shown that it it makes possible indicating the sound changes in timbre whic could not be revealed from the notes.

music perception; segmentation; computational modelling

nije evidentirano