A Joint Acoustic-articulatory Study of Nasal Spectral Reduction in Read versus Spontaneous Speaking Styles

Data on nasal articulation obtained through real-time magnetic resonance imaging (MRI) are jointly used with acoustic analyses of the speech signal to analyze nasal production differences in read and spontaneous speech, especially focusing on the details of acoustic spectral reduction. In this exploratory study, vowel-nasal-vowel (VNV) segments from one speaker were examined and measures corresponding to the speed of the velum and spectral center-of-gravity of the nasal were extracted. It is observed that lower velum speeds in spontaneous nasal production could result in more vowel nasalization and hence a ‘lowering’ of the center-of-gravity of the acoustic spectrum. Such an analysis has implications for understanding speech planning and for informing design of automatic speech analysis algorithms.