|
I am currently engaged in a couple of research projects:
-
A Revised DFT Model for Predicting Vowel Inventories:
The Dispersion Focalization Theory (DFT), developed by researchers at the Phonetic Institute in Grenoble,
France, maintains that vowel inventories are organized according to two independent perceptual principles:
dispersion, or maximization of auditory distance between vowels, and focalization, or preference of
formant proximity within vowels, in accordance with the Center of Gravity Effect (CoGE).
In my revised DFT model, focalization is not independent from dispersion but embedded in it: Vowel inventories
are organized only according to dispersion, where dispersion means maximization of reliable distinctiveness,
and focalization enhances dispersion by boosting the dynamic reliability weights of the vowel's percept.
The model is implemented in MatLab as a simulation that identifies learnable inventories in a given acoustic space.
Detailed description of my work so far can be found in
this paper.
This has been my major research project since 2006, and is likely to become the topic of my PhD dissertation.
-
Evidence for timing unit in Hebrew:
This is an experimental study testing how phonological quantity is manifested in phonetic duration in Hebrew.
It tests the duration of the vowel /a/ in running speech in various quantity conditions (using hiatus and stress),
and provides preliminary evidence for constant linear increments in duration as a function of increment in quantity,
which resembles 'mora timing'.
-
Regressive Sibilant Coronal Harmony in Georgian:
Posterior-biased regressive Sibilant Coronal Harmony, or the tendency of alveolar (hissing) sibilants to assimilate
to following post-alvoelar (hushing) sibilants, is so robust in running speech in Georgian that it seems to be
partly grammaticized. This experimental study elicits such clusters, testing the effects of distinctive features, prosodic boundaries and
morphological status on the probability of this process.
-
Derivation of denominative verbs from CCVC nouns in Hebrew:
This experimental morpho-phonological study confronts native Hebrew speakers with serious dilemmas, when they have to
choose a novel verb to express a predicate based upon the meaning of a noun with the structure CCVC. This is because
the default verb derivation strategies are the least faithful phonologically. Phonological faithfulness, default
morphological operations and knowledge of the existing lexicon are all factors that play a role in the choosing the
appropriate strategy.
|
|