Toward an effective use of timbre in data sonification

The spectro-temporal structure of a sound determines its timbre, and carries musically interesting information such as instrument type and performance expressions. Using timbre in data sonification can be viewed as an inverse transform of this process: Expressing data with timbre is equivalent to designing the spectro-temporal structure of a sound. Taking that into account, timbre is most effectively used in sonification by projecting time-series data onto the spectro-temporal structure of a sound. The temporal structure of the data often differs from the archetypal spectro-temporal structure of traditional instrumental sounds. But this discrepancy contributes to novel musical expression, based on novel timbre design. From this perspective, some sonification works are presented, such as ones produced by sonification of dynamic motion of genetically-modified worms and dynamic transitions of brain-wave data. Based on these examples, methods for effective and expressive use of timbre in data sonification will be presented.