Rendering into a romanisation

I think we ought to consider using a romanisation as a way to represent the sounds of a language without too much fuss. A romanisation is a phonemic transcription, and isn't exact. A phonetic transcription using IPA or International Phonetic Associated letters would be more exact, since each IPA letter represents a distinct phoneme.

A phonemic transcription would suffice if we have a correlation between the exact phonetic IPA and the less exact phonemic romanisation.

The phonemic character is followed by the IPA in square brackets [X], and then a description of the character

Vowels:
a [a] low, unrounded, front vowel
e [ε] mid-high, unrounded, front vowel
i [i] high, unrounded, front vowel
o [ɔ] low-mid, rounded, back vowel
u [ɯ] high, unrounded, back vowel

Consonants :

b [p] voiceless unaspirated bilabial plosive fricative
p [p^h] voiceless aspirated bilabilal plosive fricative
m [m] bilabial nasal
f [f] voiceless labialdental fricative
v [ʋ] voiced labiodental approximant

d [t] voiceless unaspirated alveolar stop
t [t^h] voiceless aspirated alveolar stop
n [n] alveolar nasal
l [l] voiced alveolar lateral

g [k] voiceless unaspirated velar stop
k [k^h] voicless aspirated velar stop
h [h] unvoiced glottal approximant
ŋ [ŋ] voiced velar nasal

z [ʦ] voiceless unaspirated alveolar affricae
c [ʦʰ] voiceless aspirated alveolar affricate
s [s] voiceless alveolar fricative

y [j] voiced palatal approximant
0 [^ʔ] glottal stop for syllables beginning with a vowel

tones

X¹ [X₃₃] tone 1 (陰平聲)
X² [X₁₁] tone 2 (陽平聲)
X³ [X₃₁] tone 3 (上聲)
X⁴ [X₅₃] tone 4 (去聲)
X⁵ [X₃] tone 5 (陰入聲)
X⁶ [X₅] tone 6 (陽入聲)