In this book we have approached the issue of voice and speech quality perception as a research topic, starting out with the fundamental question of:

How do listeners perceive voice and speech quality and how can these processes be modeled?

Any quantitative answers to this question require measurements. The science that deals with measurements and scaling is metrology, whereby, from a classical point of view, metrology is understood as being directed toward physical measurands exclusively. With regard to the topic of this book, such a restriction would, however, imply that acoustic speech events would be measurable, but not their perceptual counterparts, namely, the perceived speech sounds. Implicitly, this would also mean that any assessment of perceptual objects by means of human observers (subjects) would fail as a measurement, although quantitative data may result from it.

One of the reasons for the reluctance to accept dimensions of perceptive quality events as measurands is the fact that there is no basis of reference in the following sense: Measuring is a planned activity in which a quantitative comparison of a mea-surand and a reference quantity of the same dimension - a unit - is carried out. This procedure, which is obvious for physical quantities, is harder to imagine for perceptual measurands, as the relevant dimensions, and even more so, appropriate units have often not yet been identified and/or defined. Consequently, as long as we do not know the relevant dimensions of speech quality and as long as we do not have defined units...