A system for expressive spectro-spatial sound synthesis

The GLOOO system is a holistic digital musical instrument for spatial sound synthesis, considering a control device, a real time synthesis algorithm with a related modeling procedure and the integration of sound reproduction systems. Statistical Spectral Modeling is introduced as a novel method for transferring a sample library of musical instrument sounds into a data model for the application in expressive synthesis systems. A custom sample library of anechoic violin sounds has been produced to generate the model used within this project. Centerpiece of the instrument is a multi-channel synthesis software, implementing the Statistical Spectral Synthesis algorithm in C++. All spectral components of the synthesis software are dynamically routed to multiple outputs and sent to a third party spatial rendering software. This allows the realization of Spectro-Spatial Sound Synthesis - the integrated spectral synthesis of sound with dynamic spatial distribution properties, controlled through a set of shared control parameters. Expressive control over synthesis and spatialization is enabled through a custom-designed haptic developed in an interdisciplinary team. Within a user study, participants had the opportunity to create individual mappings between interface and spatial synthesis, using a visual programming environment. Control and synthesis parameters of the system as well as the mapping strategies of different users were investigated. Results show a preference for specific sensor units of the control device and for specific connections between control and synthesis parameters. Case examples reveal different mapping types, including the high confidence mapper, the exploring mapper and the confused mapper. The full system can be considered an environment for further studies on human-computer interaction in musical contexts.

Author dc:creator

• Coler, Henrik von

Licence dc:rights.uri

• https://creativecommons.org/licenses/by-nc-sa/4.0/

Language dc:language.iso

en

Identifier URI

http://dx.doi.org/10.14279/depositonce-11350

OAI identifier oai:identifier

oai:depositonce.tu-berlin.de:11303/12531