Timbrenauts: Creative explorations in timbre space

Abstract:

In this research-creation project we wished to use experimental designs from timbre perception research to generate data models that inform the creation of new compositions for an atypical instrumental duo. We wished to test whether research methodologies can be adopted for use in perceptually informed orchestration/composition practice. The experimental portion generated similarity spaces for each instrument alone as well as both together, using recordings of a variety of sound production approaches on each instrument as stimuli. We hired an additional performer on an instrument which combines with Berk Schneider’s trombone to make a unique duo, for which we have commissioned short pieces from members of the ACTOR community. We asked that the composers make use of the structures of similarities and differences revealed in the experimentation phase to shape the pieces. E.g. creating sequential and concurrent auditory groupings via timbral similarities, exploring maximal timbral differences, or tracing trajectories through the timbre spaces as seed material. This project did not aim for generalizability; the sounds collected for the experiment cover the timbral expressions and extended techniques possible on the instrumental pair as played by the specific performers chosen, and with the interests of collaborating composers taken into account. We shared both our data and work process in the resultant TOR module, allowing others to make use of either. Another outcome was a paper detailing the differences between the individual timbre spaces of each instrument as opposed to the one generated for both together. Which areas of the individual instrumental timbre spaces get compressed or expanded after their stimuli sets are used as one larger stimuli set for a third similarity judgment experiment could reveal what perceptual features are important to listeners in an instrumental solo as opposed to an ensemble situation.