Perception of pathological voice quality is essential in clinical voice evaluation and validation of objective measures of voice. Patients and their families decide whether treatment is successful based largely on whether the patient sounds better. Similarly, clinicians make many decisions about managing speech and voice disorders based upon on perceptual judgments. However, these """"""""subjective"""""""" measures of voice quality are not highly regarded as either clinical or research tools, because of inherent problems with inter- and intrajudge reliability, they are considered to lack objectivity, and there is no accepted set of perceptual scales used by clinicians. This apparent contradiction has resulted from a lack of cogent research into the sources of variability in voice quality judgments. The proposed research addresses the fundamental questions of what vocal qualities to measure and how to measure them by examining how each component of the """"""""speech chain"""""""" contributes error to voice quality ratings. What voice qualities are perceptually real and perceptually important will be determined using multidimensional scaling and factor analytic techniques to examine the perceptual characteristics of large sets of pathological male and female voices. The influence of type, of rating task, rating context, and clinical experience on voice ratings will also be examined. Finally, the results of these studies will be used to test an """"""""anchored"""""""" voice rating protocol that is resistant to the error resulting from these effects. Evaluating pathological voices using such a protocol is analogous to measuring size with a ruler, as compared to estimation by eye or memory. The long-term goal is to maximize clinical rating reliability by identifying, investigating, and ultimately controlling sources of variability in voice quality ratings. Once this goal is achieved, standardization of voice rating procedures may be achievable. Considering the key role of voice quality perception in both clinical and research practices, the need for increased understanding and reliability, and eventual standardization in this field cannot be overstated.
Showing the most recent 10 out of 35 publications
