Tom Bourbon [941005.0833]
[Avery Andrews 941005.1753]
[Description of Avery's research design omitted here -- TB]
The expected result (if you're a linguist) is that subjects would
resist disturbances that merged or otherwise distorted phonemes,
but not ones that didn't (for example, slowing down the overall
speech-rate, but who knows?)
Yes! (I'm not a linguist, Avery, and I don't _know_, but I'm willing to
make a bet.)
Is anyone out there watching and reading all of the nifty PCTish research
projects on "really big topics" we are giving away for free here on csg-l
this month? October is the time for our give-the-world-a-research-program
bonanza.
Later,
Tom
[Avery Andrews 941005.1753]
Here's a vague idea for a possible sort of PCT experiment in
linguistics. Suppose we could come by a speech synthesis
program whose output parameters could be systematically varied
by the program. For example, the gap between the F1 and F2 formants
of an /i/ vowel could be made larger or shorter, or the lag between
stop-release of a consonant and voicing onset time of the following
vowel increased or decreased. The program would read a passage of
text, with some slowly varying term being added to one of these
parameterws, & the experimental subject would hold a stick, being
instructed to try to keep the speech sounding intelligible.
The expected result (if you're a linguist) is that subjects would
resist disturbances that merged or otherwise distorted phonemes,
but not ones that didn't (for example, slowing down the overall
speech-rate, but who knows?)
Avery.Andrews@anu.edu.au