Voice quality in Norwegian: effects of dialect, speaker age, sex, and speaking mode

Abstract

This study investigates the effect of Norwegian speakers’ dialectal background, age, and sex as well as speaking mode on voice quality. Recordings of read and semi-spontaneous speech were used to analyze short and long /a/ vowels produced by younger (< 40 years) and older (> 40 years) female and male speakers from four main dialectal variants. Acoustic analysis comprised both spectral tilt and harmonicity measures. The analysis revealed that the factor dialect did not have a robust impact on any voice quality measure. Speaker sex affected both types of measures more than any of the other factors, females having steeper spectral tilt and higher harmonicity values than men. Only the spectral measures differed with age, older speakers demonstrating steeper slopes than younger subjects. For spontaneous speech, somewhat steeper slopes and higher harmonicity values were observed than for read speech. Current results conflicting with previous evidence are discussed.

Voice quality in Norwegian: effects of dialect, speaker age, sex, and speaking mode