FACE and GOAT sets

I've been transcribing these as monophthongs, [e] and [o] respectively, but I don't think that's accurate for open syllables. It sounds to me like I have very narrow closing diphthongs in these sets in open syllables, like 'say' [see_r] and 'go' [g8_-o].
I'm interested in how these sets behave in final unstressed syllables, like in 'Monday' and 'yellow'. For me, they have an alternation depending on the stress of the following syllable. In absolute final position or before another unstressed syllable, they have their full value, but when the following syllable is stressed, the diphthong is abbreviated. (The beginning of /o/ seems to be unround
"Monday is the ..." ["mVn.dee_r Iz D@]...
"Yellow is my... ["jE.l8_-o Iz mAI]
"Monday morning" ["mVn.dE "mOr/nIN]
"Yellow sheets" [jE.l@\ "Sits]
Some Ottawa Valley speakers, especially older ones, seem to reduce them to [i] and [@] respectively; where else does this occur? Does it happen in Ireland? (that's where the Valley was mostly settled from)

>> Some Ottawa Valley speakers, especially older ones, seem to reduce them to [i] and [@] respectively; where else does this occur? <<

I believe that they exist in many other NAE dialects. Probably the South especially.

I don't think I have any vowel alternation in those contexts.

"Monday is the..." ["mVndeI Iz D@]
"Yellow is my..." ["jEl7U Iz maI]
"Monday morning..." ["mVndeI "mO@`nIN]
"Yellow sheets..." ["jEl7U "Si:ts]

I don't think I ever have anything [E] or [@\] there.

<<Some Ottawa Valley speakers, especially older ones, seem to reduce them to [i] and [@] respectively; where else does this occur?>>

Around here, I think people of my late grandfather's generation often tended to reduce final unstressed /o/ to [@]. For example, I heard my grandfather pronounce "window" as ["wInd@]. But I think that that reduction may be nearly extinct here now. (I think reduction of final unstressed /o/ to [@] or even [@`] occurs in some Southern US accents, and reduction to [@] is also a feature of some very broad London accents.)

Also I think reduction of final unstressed /e/ to [I] occurred in traditional RP. For example, traditional RP tended to have things like ["t_hju:zdI] "Tuesday" and ["O:5wIz] "always". I think in modern RP those have been supplanted by [eI].

I've heard similar pronunciations (with [I] or [i]) sporadically here in the US, as well; I think they tend to be perceived as archaicisms.

Correction: <I don't think I have anything like [E] or [@\] there.>

It seems that the reduced form only occurs before consonants
"yellow animal" [jEl8_-o "E@.nI.mUo]
And I think my reduced /o/ might actually be rounded [8]. Maybe it varies depending on context. I'll have to look into that.
I remember as a kid being confused by the name of the book "Old Yeller". Eventually I figured out that it was from 'yellow'. I wonder if it was intended to be non-rhotic ["jE.l@] or whether it represents a dialect in which unstressed /o/ -> [@] and [@] -> [@`].