| |
Abstract:
The present paper concerns acoustic vowel reduction in Italian
and has two main purposes: the first is to know more of this
process, asking whether it is a centralization or a
coarticulation process or both; the second is to assess possible
relations between acoustic data and articulatory EPG data. To the
first purpose, acoustic duration, F1 and F2 of the vowels /a/,
/i/, and /u/ produced by two Italian subjects were analyzed under
three conditions, utterance type, stress, and syllable position
within a word. The data for each vowel could be ordered along a
hyper/hypo continuum, which indicates that F2 variations can be
in line with the hypothesis of vowel centralization, but not so
for F1, whose variations, for /i/ and /u/ appear to depend in
large part on jaw movement and can be better accounted for as
consequences of consonantal coarticulatory effects.. For the
second purpose the acoustic data on F1 and F2 and two EPG
parameters (measured for the same corpus and subjects in a
previous study) were submitted to a set of regression analyses,
whose results indicate that the variations in tongue to palate
contact can reveal relevant aspects of the vocal tract
configuration that have significant acoustic effects on both F1
and F2.
|
