VOICE WITHOUT SPEAKER: HUMAN SPEECH SYNTHESIS IN ACOUSTIC INSTRUMENTAL CONTEXTS

Abstract

AbstractThis article provides an overview of compositional practices that invoke representations of the human voice by acoustic means, without the presence of actual human speakers, and, in identifying the limitations and deficiencies of extant practice, proposes a new method or methodological basis for achieving the same phenomenon, to different ends. Clarence Barlow's seminal synthrumentation method forms a central point of discussion; when considered alongside the work of, among others, Mayuzumi, Grisey and Ablinger, it becomes clear that, in existing synthrumental (or instrumental synthesis) practices, the innate inharmonic components of target material for reproduction have largely been discarded or overlooked. After a brief overview of the basic linguistic and perceptual features of speech, a new method designed to evoke whispered speech, with inharmonicity in mind, is demonstrated.