[k] is only weakly aspirated, if at all, because it starts an unstressed, word-medial syllable.
[ɪ̈] is my close schwa/unstressed short <i>. It's more front than [ɨ], but less so than [ɪ]. The distinction between [ɪ̈] and [ə] (and other reduced vowels) is only present in syllables with secondary stress and in some loanwords, otherwise it's purely allophonic (I don't distinguish "Rosa's" and "roses" except in very careful speech).
I'm pretty sure there's secondary stress (at least higher pitch) on the [ɾi] syllable. The overall contour is something like HLHHL.
You take it a step further and actually voice your /k/ there? Heh. (While I do sometimes make it so that /g/ and /k/, or /b/ and /p/, are indistinguishable in these positions aside from vowel length, it is always in favor of merging to voiceless lenis realizations rather than to voiced ones.)
Yup, I voice it. There seems to be a consistent sound change here of /p k/ to in the same environment intervocalic flapping occurs.
Heh, if I sound change it to Mekoshan I get [ɥexpʰie].
Last edited by TaylorS on Mon Jun 20, 2011 10:58 pm, edited 1 time in total.
Nancy Blackett wrote:For what it's worth, /wI.ki."pi.djæ/.
/æ/ in an unstressed syllable? Which British dialect is this?
Probably something influenced by (or a dialect of) Scots (as Scottish English in reality these days tends to form a gradual continuum off into Scots proper, or rather, what had once been historically called Scots now is more likely to be called Scottish English even if it has changed far less than the name change would imply). Scots dialects are kind of funny about their unstressed final vowels, commonly written -ae (which vary wildly from dialect to dialect rather than typically being reduced towards schwas as they generally do in English).
You take it a step further and actually voice your /k/ there? Heh. (While I do sometimes make it so that /g/ and /k/, or /b/ and /p/, are indistinguishable in these positions aside from vowel length, it is always in favor of merging to voiceless lenis realizations rather than to voiced ones.)
Yup, I voice it. There seems to be a consistent sound change here of /p k/ to in the same environment intervocalic flapping occurs.
You do maintain a contrast in these cases by contrasting /p/ from /b/ as versus [β] and /k/ from /ɡ/ as [ɡ] versus [ɣ], without any vowel length contrast, from what I remember you as having said, right?
(It is interesting how rather similar phonological changes have rather different results here when combined with other phonological changes and tendencies, as contrasted with my tendency towards merging both sets as [b̥] and [ɡ̊] respectively but distinguishing them instead by vowel length.)