The colour of music: Spectral characterisation of musical sounds filtered by a cochlear model

In music research, attention is more and more: focused on the main factors of sound perception: This holds in particular for the capability of discrim inating music sound features, notably attacks and periodic components, both of which are believed to generate timbre sensation. In this paper it is ar gued that the frequency spectrum of the stationary part of instrument sound s are coded, at some stage, within the nervous system, into a small number of parameters. Somewhat as in vision the light spectrum of a uniformly pain ted area is coded by the retina into three fundamental colours, the sound p arameters seem to represent what may be called "instrument sound colours". The analysis is carried out on music sound samples pre-filtered by a non-li near model of the human cochlea. The main reason for this approach lies in the fact that the non-linear filtering performed by the cochlea works to re duce significantly the information content of acoustic input. Some properti es of the model, particularly those related to the efficacy of non-linearit y ill tone masking and noise suppression, are evidenced by examples. The re liability of the model in sound colour characterisation, also in the case o f noise-added sounds, is investigated as well. A principal-component analys is of cochlear responses to sounds generated by 21 musical instruments seem s to indicate that the spectra of the stationary parts of instrument sounds are perceived as combinations of three fundamental components.