One Voice or Many?: Compound Melody & Auditory Streaming (15/05/08)
Reconciling the difference between “music as composed” and “music as heard” is a challenge that engages researchers across many different disciplines. However, the task of developing technologies that can recognise and identify a musical piece, regardless of performer or performance, proves as illusive as ever. The majority of psychological investigations manipulate audio waves or utilise pure tones to investigate how a person perceives music, whereas technological research tends to devise algorithms to which music is later applied in attempts to generate electronically what the human perceptual system does with such speed and ease. In the main, these methods fall short of discovering what the differences are between music as composed, written, recorded, and music as heard. In my talk I will present recent findings which support the claim that in order to investigate what music is we must place music at the foundation of our research. My research aims to discover the extent to which we perceive musical structure in the way that music theory denotes, so as to move toward devising new theories of musical listening which place music, and not sound, at their focal point. The novel and interdisciplinary approach of combining music-analytical theory with empirical psychometric testing in order to further computational and electronic research into music technology will be discussed and preliminary results presented.