Candidato: Dott. Gianluca Agostini

Relatore: Dott. Marco Bertola

Abstract:

This dissertation illustrates technical, theoretical and design set of problems faced in the creation of a visual mixing device. The target of this device is to make it possible to mix a song in a tridimensional graphic environment in which every single sound is represented by a sound object. This object will have a position and determined geometrical characteristics that should be related to every parameter of a sound that characterize it as an element of a mix, allowing the user to interact with it using the movement of his own hand. The goal is to substitute or be placed side by side with traditional instruments of the musical production in the recording studio and trying to create a new mixing approach. Today, multimedia nature of technologies led to new developments in the bound- ary of workflow, but the investigation on new possibilities offered by graphical representation of the sound content hasn't been carried out extensively. This is true for new interaction methods related to the development of the natural user interfaces, too. Waveform visualization, frequency spectrum, phasescope, etc. are instruments that have been perfected with new and low cost technologies but they don't represent an innovation as much as, for example, new pattern and loops handling models: Ableton Live1 or Bitwig Studio2 are two good examples. The most interesting and pioneering contribution in this area of interest belongs to David Gibson. In the middle of nineties in his book "The Art of Mixing" he introduces its audience to a path that employs a graphical representation of a mix for a pedagogical purpose. This book documents a new representation of the sound, based on the modeling in a spherical shape of the signals belonging to the channels of a mixing console, inside a virtual room that is the stereo image. In this work, therefore, I will illustrate and analyze the "vision" of Gibson and the patents and attempts made to date in this area. From a critical approach to these attempts and on the basis of the studies on the psychophysical of the perception of sound, I will propose a model that is both synthesis and overcome of the attempts made so far and I will present a prototype of a visual mixing device. For the construction of this prototype I made use of two development environments, Max/MSP3 and Processing4, which, by their nature, focus on the immediate development of algorithms through the use of base modules, referring code optimization to future developments.