| |
Abstract:
The pseudo-palates used with the Kay Elemetrics and Reading
electropalatographic systems differ mainly according to
theirnumber of electrodes, 96 vs. 62 respectively. Differences
are also found in the spatial organization of the electrodes,
arranged in arches for Kay and in rows for Reading, and in their
distribution over the palate, with a richer electrodes coverage
in the dental, palatal and velar regions with Kay. Finally, these
two pseudo-palates greatly differ in their price according to
their number of electrodes. The objective of this study is to
compare how variations in linguopalatal contact are reflected by
the two pseudo-palates and to evaluate the contribution of the
extra 34 electrodes of the Kay system. Three types of analyses
are conducted. The first analysis looks at the differences
between the number of contacted electrodes with each
pseudo-palate for a particular segment in order to evaluate
whether addition of contact information is more pertinent for
some segments or articulatory regions. In the second and third
analyses, the pertinence of the additional information given by
the 34 extra electrodes of Kay is tested by determining whether
this information increase the distinctive potential of the
pseudo-palate. Pseudo-palates are compared on the basis of their
discrimination of different articulatory patterns and on how they
can replicate articulatory variations involved in spatial C-to-C
and V-to-C coarticulation. Results show that with greater
electrode coverage, the Kay system reflects with greater
precision the actual contact of the tongue against the palate and
subtle articulatory variations. It provides additional
information mostly in the palatal, velar and dental regions.
However, with 62 electrodes, the Reading pseudo-palate is still
able to reflect the articulatory variations studied and to
discriminate articulatory patterns. Implications of these results
for improving the resolution of the Reading pseudo-palate are
discussed.
|
