We present a speech rhythm analysis workflow based on two different but complementary methodologies: one at a segmental/syllabic level, and another one at a suprasegmental/prosodic level. This bimodal methodology is devised to study the effects of regular and irregular rhythmic priming — i.e. rhythmic stimuli presented to speakers before elocution — on the speech rhythm of people with Parkinson’s disease.
Preliminary results show that both segmental and suprasegmental rhythm metrics vary in different priming conditions. We propose that this methodology can be used to have a better grasp of speech rhythm by peering into the potential interaction or correlation between vocalic-consonantal interval distributions and pitch movement variations. If models are demonstrated to be effective, future implementations of the workflow can be envisaged in other areas of speech analysis.