There have been many systems that automatically create a melody. However, when the created melody is not satisfactory, it is difficult for musically untrained people to manually edit it on a conventional MIDI sequencer. Therefore, we propose a melody editing method based on a melodic outline, which represents the overall shape of a melody. Given a melody, its melodic outline is obtained by applying the Fourier transform to the melody's pitch trajectory and extracting low-order Fourier coefficients. After the outline is redrawn, it is transformed into a note sequence by the inverse procedure of the extraction and a hidden Markov model. Experimental results showed that (1) for novice participants, our system was easier than the conventional piano-roll interface, (2) generated melodies were satisfactory for both novice and intermediate participants, and (3) novice participants' ideas about what melody they want became clearer as they experienced melody editing everyday.
Hidden Markov model, Fourier transform, Melodic outline, Computer-aided music composition, Melody editing
How to CiteTsuchiya, Y. & Kitahara, T. (2019) “A Non-notewise Melody Editing Method for Supporting Musically Untrained People's Music Composition”, Journal of Creative Music Systems. 3(1). doi: https://doi.org/10.5920/jcms.624