Consonants

/p b t d ʈ c ɟ k (xk) ʔ/ <p b t d ṭ c j k xk q>
/f v s̺ z̺ s̻ z̻ x ɣ h/ <f v ṣ ẓ s z x r h>
/m̥ m n̥ n ɲ̊ ɲ ŋ̊ ŋ/ <mh m nh n ñh ñ gh g>
/ɬ l/ <lh l>

The laminal sibilants s z are commonly realized as [θ ð], especially before back vowels, and can be realized as [ç ʝ~j] before front vowels. Some speakers always realize z as [j]. /ɬ/ is more narrowly a fricated retroflex lateral flap [ɺ̝̠̊], the voiceless parallel to the retroflex lateral flap /l/ [ɺ̠]: in the articulation of both, the tongue is brought up to the back of the alveolar ridge and flapped forwards, hitting the back of the bottom teeth.

Voiceless nasals are realized as longer than voiced nasals, and the voicelessness is only present for part of the realization: amha, for example, would be pronounced [aʰm̥ma].

The velar fricatives may also be realized as uvular.

Vowels

/a a: e e: ɤ ɤ: o o: i i: ɯ ɯ: u u/ <a a: e e: ẹ ẹ: o o: i i: ɯ ɯ: u u:>
/æɪ ɒʊ ɪʊ/ <ai au iu>

/ɪʊ/ becomes [eʊ] after velars, and is written <eu> in that environment.

Pitch accent

There is a pitch accent, derived from Ingomoe stress and written with an acute: syllables with the accent take a tonal downstep.

Phonotactics

Initial consonant clusters are CC, and must be of equal or rising sonority. (I think.) No final clusters.