[Aspects of voice quality quantification]

The present work is directed towards providing a complement to the current use of sustained sounds as the basis for the physical measurement of voice quality parameters. For many clinical observations, there is a clear lack of correspondence between the measurement results, for example of shimmer, jitter and harmonic to noise ratio, and the perceived quality of the speaking voice. Our work is motivated by the need to overcome this familiar problem. It has four main objectives: to use running speech as the basis for one set of measurements in the clinical protocol; to link physical measures based on the use of this data with both production and perception; to employ voice production data as the foundation for auditorily significant criteria; the presentation of quantitative comparisons using sustained sound and running speech data, derived for a range of pathological voice conditions, within the same clinical session. Our pilot observations are based on the use of clinical acoustic and laryngographic data from patients of 4 different clinical centres.