Adding feedback from Reddit.
This commit is contained in:
parent
dc0ffd5986
commit
a4feb97661
1 changed files with 18 additions and 0 deletions
18
notes.txt
Normal file
18
notes.txt
Normal file
|
@ -0,0 +1,18 @@
|
||||||
|
**ryodifūne**: *noun* The final run.
|
||||||
|
|
||||||
|
**kojinōmi**: *noun* A dead person.
|
||||||
|
|
||||||
|
**wabōryo**: Fighting bola.
|
||||||
|
|
||||||
|
# From Reddit
|
||||||
|
|
||||||
|
miwāfu [mi.waː.ɸɯ̥] -- word-final position after a voiceless consonant is the most common place to find /ɯ/-devoicing!
|
||||||
|
|
||||||
|
kanéko [ka.ꜛne.ko] -- looks good!
|
||||||
|
|
||||||
|
rutejìmo [ɺɯ.te.ꜜdʑi.mo] -- also looks good! Though, out of curiosity, is /di/ a valid segment? or /ti/? In Japanese, /di/ becomes [dʑi] and /ti/ becomes [tɕ] so just curious!
|
||||||
|
|
||||||
|
waryoni pagani héru [wa.ɾjo.ni pa.ɡa.ni ꜛçe.ɾɯ] -- for this one, I didn't make any changes, but the /ɯ/ at the end of héru could again be devoiced.
|
||||||
|
oe shimusògo i fapòdi eyo rutejìmo [o.e ɕi.mɯ.ꜜso.ɡo i ɸa.ꜜpo.di e.jo ɺɯ.te.ꜜdʑi.mo] -- also looks good! There could potentially be a devoicing of /i/ in shimusògo, particularly in rapid speech, but not consistently, as the /m/'s voicing can prevent it.
|
||||||
|
|
||||||
|
pibini [pi.bi.ni] -- this one I would definitely not devoice. Devoicing in the middle of a word (not word-finally) is almost ALWAYS the interaction of a voiceless fricative and a voiceless plosive. I can't think of a single example where you would have two plosives, and definitely not when one of them is voiced. Here is an example of a Japanese word with devoicing: Shikoku (island) [ɕi̥kokɯ̥]. The /i/ here is almost universally devoiced, with the /ɯ/ optionally (though preferred) devoiced.
|
Loading…
Reference in a new issue