Composing Musical Spaces By Means of Decorrelation of Audio Signals

Since the seminal work of John Chowning on the simulation of moving sound sources (Chowning 1971) there has been in computer music research a tremendous amount of work concerning perceptual sound space definition. Chowning's model implemented some complementary cues regarding localization, distance, and movement, among which we can recall (1) an interaction between spectral brightness and loudness; (2) a Doppler shift, well known technique for simulating speed and movement in terms of time-varying frequency rates; and (3) a control over the Azimut, the horizontal plane regarding physical separation between channels. Chowning himself referred at the last year's DAFx Conference (Chowning 2000) to the first aspect (the second being already well understood). I will deal myself this year with another aspect apparently related to the third point, concerning inter-channel temporal decorrelation of audio signals.