Osamu Fujimura
The C/D Model as a Dynamic,
Non-segmental Approach
Abstract:The Converter/Distributor (C/D) model is a theory of phonetic implementation for computing the articulatory and acoustic signals as time functions. As its input, the specification of an utterance is given both in terms of phonological representation and parameter values for the utterance/discourse situation. It uses syllables as the "segmental" units, and the content of each syllable is specified by phonological features, each of which pertains to one of the syllable components: onset, coda, or nucleus for the syllable core, or a syllable affix (p-fix or s-fix).
The prosodic structure is specified as the input to this implementation process in the form of a metrical tree for each sentence, augmented by a numerical mark optionally attached to any node of the tree according to the utterance situation. Other utterance parameter values for the discourse and speaker conditions are also specified.