Skip to content

News Arts and Science Teaching Media Library Services IEM - intern Contact
  You are not logged in Link icon Log in
You are here: Home » Infos & Services » Veranstaltungen » Events 2006 » Relating gestural control, acoustics and verbal description of guitar tones » document_view

Relating gestural control, acoustics and verbal description of guitar tones

Caroline Traube, Faculté de musique, Université de Montréal. Gastvortrag am 1.6. um 9:15 in der Bibliothek am IEM

Guitar timbre is affected by structural components of the guitar (static control parameters) and but also by the gestures applied by the performer on the instrument (dynamic control parameters). In particular, the plucking position along a guitar string has a major effect on guitar tones, by essentially modifying their degree of brightness. In this study, a digital signal interpretation of the plucking effect is derived from a comb filter model. We also investigate the relationship between subjective characteristics of sound (such as timbre) and gestural parameters. The starting point of our exploration is an inventory of verbal descriptors commonly used by professional musicians to describe the brightness, colour, shape and texture of the sounds they produce on their instruments. The voice-like quality of guitar tones may arise from the comb-filter shaped magnitude spectrum of guitar tones, whose peaks correspond approximately to the formant frequencies of identifiable vowels. These analogies at the spectral level may account for timbre descriptors that seem to refer to phonetic gestures, such as "open", "oval", "round", "thin", "closed", "nasal" and "hollow". To test this idea, we invited experimental participants to associate a consonant with the attack and a vowel with the decay of guitar tones. Our results support the idea that the perceptual dimensions of the guitar's timbral space depend in part on phonetics. In this presentation, we will also show how the comb filter model can be exploited to estimate the plucking point information from recorded guitar tones, using an iterative weighted least-square algorithm and starting from a first approximation of the relative plucking position derived from a variant of the autocorrelation function of the guitar tone acoustic signal.


Last modified 17.05.2006