When humans vocalize non-speech sounds, as when imitating other sound-producing mechanisms, how effective is this vocalization? What purpose might this ability serve? The human vocal tract is adept at producing speech sounds, allowing verbal communication despite variations in these sounds due to differences in size, shape, and dynamics of this mechanism among individuals. In contrast, non-speech sounds that were not originally produced by the human vocal tract derive from even more structurally different mechanisms---those of animal calls or machinery sounds, for example. This study takes a theoretical and experimental look at this endeavor. Included is an examination of the nature of certain sound symbolisms showcasing speech as a proxy for non-speech sounds; one well-known example we touch on is jazz scat singing.; Through experiment, we determined that in imitative tasks of non-speech sounds, subjects were intuitively able to produce distinguishing articulatory movements for each sound even when lacking auditory feedback. These results motivated us to undertake an exploration of systematic uses of speech---both in written and acoustic form---as a supplemental method for labeling the sounds of musical instruments, an approach that we have termed Music Phonetization. Potential advantages of this approach include the pervasiveness and, as will be demonstrated, innate precision of the human voice, as well as the existence of a standardized template independent of cultural and historical contexts (i.e. the International Phonetic Alphabet) which can be employed for this labeling. The conclusions we will draw are also relevant for the development and evaluation of other non-speech, vocal forms of communication like those in human-machine interfaces, with members of other species, and among humans with differing language competence (e.g. adults and toddlers).
展开▼
