The Music Acoustics Group at KTH

The research in the music acoustics group concerns all aspects of Sound and Music Computing (SMC), and runs in several main streams:

• the science of music performance, including emotional and gestural aspects
• technical vocology - the science of the human voice, especially in singing and for medical applications
• instrument acoustics, especially strings
• sound in interaction, including sonification and sound design
• computer-assisted music education

The interactions between these topics are rich and diverse. An experimental approach has long been a signature of the research, often using the paradigm of analysis-by-synthesis. Additional interests include modelling of sounds from everyday objects; and systems for sound field reproduction, especially for telepresence applications.

Today, the availability of computing power attracts researchers to synthesis strategies that are based on direct mathematical descriptions of the musical instruments and the human voice, commonly referred to as numerical or "physical modelling." This interest in numerical modelling has revealed a lack of reliable and detailed experimental data in many areas. This applies in particular to the function of the control systems used in actual performances, for example breathing behaviour in singers and wind instrumentalists, or bow movement in string playing. Without thorough knowledge of such control systems, the experimenter is forced to explore an overwhelmingly rich parameter space, and the chance of real success in a reasonable time is small. The issue of how sound production is controlled by the player or the singer is therefore central to our work. This leads us into topics as diverse as articulatory synthesis of singing voice, controlled by physiological control parameters such as jaw opening and tongue shape; how the sounds of footsteps reflect the walker's mood, weight and even gender; and how a string may be set into motion by a bow to achieve different expressive intents.

Our research on musical performance by computer modelling is a long term project. Its focus is presently to characterise expressions of emotion and gestural activities in music performance. Our generative music performance grammar, with music examples illustrating the effects on synthesised performances, is accessible on this website.

Our overall aim is to refine our understanding of the voice and the musical instruments (now needed more than ever, it seems) and how they are used in communication; and to take an active part in the development of improved methods for analysis and synthesis.

The 2006/2007 activity report, in Swedish, can be downloaded here (pdf).