Contraction point by Kosmas Giannoutakis

The adaptive listening virtuosity of the pianist will enable him to contract the feedback network system towards a possible point, where space, time and timbre merge.

The set up has a circular form, with the piano in the middle of the circle, an omni-directional microphone above the piano, 12 loudspeakers on the circumference of the circle and the audience sitting in between. The system records the live sound of the piano and plays it back using delay lines with different playback speeds (transpositions) for every loudspeaker. A non-linear feedback process takes place. All 12 channels of sound are physically mixed in the acoustic space, recorded by the microphone and played back again by all 12 loudspeakers with the 12 different transpositions. Through this process all 12 sources are mutually interconnected creating a network of complex behavior. The performer's sound gestures are extended in space and time by the 12 speakers and delay lines.

At certain moments (overall 12 times) the performer leaves the piano and and make a walk around the circle listening to every speaker. He/she has to locate the speaker with the higher transposition. When he/she comes back to the piano, he/she estimates and press the key associated with the speaker with the highest transposition. The keys and speakers are mapped in a fixed relationship through the piece (speaker 1 → C, sp. 2 → C#, sp. 3 → D, etc.). The feedback system analyze the input note and evaluates a score which contracts the system accordingly by reducing the transposition range of the 12 loudspeakers. After that the system redistributes the new transposition factors to different loudspeakers and a new round begin.

After the 12th round the system locks the transposition range and reduces the time window of the delay lines. The result is that the audience looses gradually the extended space-time perception which is transformed into timbre perception. Depending on the 12 estimations of the performer, the system will end up with a different transposition range and consequently different timbre for every performance.