On the Alignment of Prosodic Events

The current study examines the relationship between intonational gestures as given by the accent commands of the Fujisaki model and the syllabic grid on the example of spontaneous American English from the Buckeye Corpus. As an initial step the data were labelled according to American English ToBI conventions. Intensity contours were extracted from the band-filtered speech signal and modelled using a second-order linear model like the accent control mechanism of the Fujisaki model. It was observed that certain accent types such as L* occasionally required the use of accent commands with negative polarity. The timing properties of the underlying accent commands correspond to the type of accent label, i.e., for instance, early for H*L and late for L*H. Accents in the vicinity of a boundary exhibited slightly higher accent command amplitudes whereas the highest syllable command amplitudes were observed for phrase-initial and medial accents. With respect to the alignment of accent commands, the onsets of syllable commands did not prove to be more precise anchoring points than the segmental boundaries of the syllable.