15th International Congress of Phonetic Sciences (ICPhS-15)
Pronunciation variation defies the capability of a strictly alphabetic notation to fully characterize, as the basic parameters governing spoken language transcend the phoneme. The time scales required to accurately describe and model pronunciation are both shorter and longer than the phonetic segment, and are inherently wedded to the syllable. The specific properties of phonetic constituents are shaped by two inter-related parameters - prosodic prominence and their ordinal position within the syllable - which ultimately reflect the entropic potential associated with lexemic discriminability. The internal structure of the syllable can be parsimoniously portrayed in terms of articulatory features (e.g., voicing, place and manner of articulation, etc.) and used to characterize the microstructure of pronunciation variation. Parsing the utterance into syllabic units of variable prominence is essential for characterizing spoken language, and of utility for developing robust speech technology.
Bibliographic reference. Greenberg, Steven (2003): "Pronunciation variation is key to understanding spoken language", In ICPhS-15, 219-222.