Harmonic interaction for monophonic instruments through musical phrase to scale recognition

Abstract

This thesis introduces a novel approach for the augmentation of acoustic instruments by providing musicians playing monophonic instruments the ability to produce and control the harmonic outcome of their performance. This approach is integrated into an interactive music system that tracks the notes played by the instrument, analyzes an improvised melodic phrase, and identifies the harmonic environment in which the phrase is played. This information is then used as the input of a sound generating module which generates harmonic textures in accordance with the identified scale. At the heart of the system is an algorithm designed to identify a scale from the played musical phrase. The computation relies on established music theory and is based on musical parameters retrieved from the performed melodic phrase. A database of audio recordings comprised of improvised phrases played by several saxophonists is used to test the algorithm. The results of the evaluation process indicate that the algorithm is reliable, and it can consistently recognize the scale of an improvised melody conveyed by a live musician. This discovery led to the exploration of the affordance to influence accompanying harmony by a monophonic line and integrating the phrase-to-scale match algorithm within an interactive system for music-making. By interacting and playing with the system using a repurposed controller mounted on the saxophone, performance strategies and practical ways are offered to play, modify, and further develop the system.