The objection of the technology is to manipulate pitch and duration of a musical instrument sound
with minimizing distortion to timbral features. It also provides a method to superimpose two musical
instrument sounds. The user can define a favorite sound like sound palette, and then replace some
part of the musical performance with it.
The technology consists of three steps
First step ”Analysis”: it separates signals of a musical instrument by using an integrated model of
harmonic and inharmonic structures and extracts timbral features. It also obtains the residue.
Second step ”Manipulation”: it manipulates pitch, duration, and energy of the inharmonic structure
to create a new sound from signals of a favorite musical instrument.
Third step ”Synthesis”: it synthesizes harmonic and inharmonic signals for the new sound and add
them to the residue to get a new performance.