This thesis discusses the use of Kalman filtering for noise reduction in a 3-D gesture- based computer music controller known as the Radio Drum and for real-time tempo tracking of rhythmic and melodic musical performances. The Radio Drum noise reduction Kalman filter is designed based on previous research in the field of target tracking for radar applications and prior knowledge of a drummer’s expected gestures throughout a performance. In this case we are seeking to improve the position estimates of a drum stick in order to enhance the expressivity and control of the instrument by the performer. Our approach to tempo tracking is novel in that a multi- modal approach combining gesture sensors and audio in a late fusion stage lead to higher accuracy in the tempo estimates.

When people first started to use digital technology to generate music, they were thrilled with their new ability to create novel and different sounds; accordingly, much research effort has been directed towards rich and complex methods of sound synthesis. Unfortunately the deep physical connection that exists between a musician and an instrument has not received as much attention, and so although we have machines capable of synthesizing fantastic new sounds, we don’t have the ability to use these sounds with any immediate finesse, of developing virtuosity with our new instruments. The work presented in this thesis is an exciting step towards a more dynamic future for computer-based musical performance. The Radio Drum was designed in 1987 at AT&T labs as a prototype of a three-dimensional mouse. Max Mathews later repurposed the apparatus as a musical instrument known as the Radio Baton, which in its most modern form outputs data using the MIDI protocol. The result of this work, a new system called the Audio-Input Drum, provides high-resolution gesture capture, a simplified apparatus, and access to the extremely flexible Max/MSP/Jitter real-time software signal processing environment.