Third International Conference on Spoken Language Processing (ICSLP 94)

Yokohama, Japan
September 18-22, 1994

Syllable Timing Computation in the C/D Model

Osamu Fujimura

Department of Speech & Hearing Science, The Ohio State University, Columbus, OH, USA

The C/D model, a syllable-based phonetic implementation theory, is reviewed, and a revised algorithm for computing the temporal structure of speech in specified utterance situations is described. The theory assumes that all prosodic information is given in the form of a pulse train representing syllables and boundaries, where the height of each pulse is a scalar measure of the strength of the syllable or boundary. Temporal characteristics of the syllables are also computed through this magnitude pattern. Syllable edges are assigned time values according to the syllable concatenation principle, and articulatory gestures are evoked for consonantal elements by subordinate demisyllabic pulses that inherit the pulse height. Implications regarding correlations in magnitudes among tautosyllabic articulatory manifestations are discussed.

Full Paper

Bibliographic reference.  Fujimura, Osamu (1994): "Syllable timing computation in the c/d model", In ICSLP-1994, 519-522.