Idk what the image mean.
But i understand some part so let's me help you.
CC/CCC: ch, gh, gi, kh, nh, ng, ph, th, tr, qu.
Ending consonant: c, ch, m, n, nh, ng, p, t
Here the website:
https://en.wikibooks.org/wiki/Vietnamese/Consonants
- There are 17 single consonants:
Vowels Phoneme English equivalent sounds
b /ɓ/
but
c /k/ s
ky
d /z/
zoo (Northern)
d /j/
yes (Southern & Central)
đ /ɗ/
do
g /ɣ/
go
h /h/
hat
k /k/ s
ky
l /l/
long
m /m/
me
n /n/
no
p /p/ s
port
q /k/ s
quare
r /z/
zoo (Northern)
r /ɹ/
run (Southern & Central)
s /s/
stay (Northern)
s /∫/
show (Southern & Central)
t /t/ s
top
v /v/
video
x /s/
see
- There are 11 consonant digraphs:
Vowels Phoneme English equivalent sounds
ch /c/
cha-
cha
gh /ɣ/
go
gi /z/
zoo (Northern)
gi /j/
yes (Southern & Central)
kh /x/ lo
ch
nh /ɲ/ ca
nyon
ng /ŋ/ si
ng
ph /f/
Philip
th /tʰ/ as in
top (aspirated "t"), not as in
thin
tr /c/
cha-
cha (Northern)
tr /ʈʂ/ s
traw (Southern and Central)
qu /kw/ s
quare (Northern)
qu /w/
world (Southern & Central)
- There are 8 final consonants:
Vowels Phoneme English equivalent sounds
c /k/
car
ch /k/ si
ck (cu
t in the South)
m /m/
me
n /n/
no
nh /ɲ/ ca
nyon
ng /ŋ/ si
ng
p /p/
ping
pong
t /t/
top
Sorry for late reply! Sorry for confusing, I should explain what I written in my notes!
UTAU does not support ê ă ô ư đ etc in wav file names or in oto.ini alias, so because of that I think that using X-Sampa based encoding is the best for vowels.
I found out that Vietnamese has following vowel sounds (if you don't know what each sound are, please check
here for IPA vowel chart):
X-Sampa [IPA]
a [ a ]
e [ e ]
i [ i ]
o [ o ]
u [ u ]
1 [ ɨ ]
@ [ ə ]
E [ ɛ ]
O [ ɔ ]
I recommend to record all vowel+vowel combinations like in VCV for smooth vowel sounds ("a i", "i E", "@ o" and so on)
With consonants, I think that you can use "written Vietnamese" as a base but modify it a little.
This is just my opinion but I would write [kw] sound as "kw" (not as "qu") in reclist, this is because "u" all ready symbolizes [ u ] sound.
(this is just my personal taste, tho. You can disagree with this tho)
I think that I would encode đ as D (or you can use any other symbol like &, up to you).
To write reclist in "unequivocal" way, one alphabet (a), symbol (@) or symbol combination (ch) can mean only one sound, just like for example VCCV English encoding does.
"With Northern, Southern or Central pronunciation?"
I cannot give one answer to that, you can...
1) record just Northern, Southern or Central sound only. IDK which one is the most "standard" but if you record in just one only then I recommend picking a dialect that is concerned the most "standard" and/or most spoken in Vietnamese. This option will have the least recording (Vietnamese has 9 vowels already which is a huge amount)!
2) record all dialectical sound but you'll need to
think how to encode each consonant
carefully.
Writhing reclist:
According to the info you told, Vietnamese seems not to be very complicated language to write a reclist (if minus dialect question which I can't give a direct answer). The easiest way to write CVVC reclist is writing is like this: CV+CV(+C)
For example
"baba" as ba_ba
"cacac" as ca_ca_c
For VV part, record simply v0wel+vowel e.g.
a_a
a_i
a_u
a_e
...
and so on!
The best thing what I can add is that test, test and test! Make a quick test recording VB and see how it works. If you find that some sound(s) or sound combination is/are missing, add them and so on.