Post your conlang's phonology

[Whimemsz]

I don't, since they can't occur in codas, and there's no potential for ambiguity word-medially since onsets are required.

Oh, woops, I misread "cannot occur in codas" as "can occur in codas." Never mind then.

[txmmj]

I see what you're trying to do with the orthography, using <i> for consonant palatalization, so you don't want any of your vowel letters to begin with that letter, but...

1) There are better ways of doing that. Like using <ee oo> for /i u/. Some natlangs do that. It's a bit silly, but better than the weird combinations you're using right now
2) You could use <j> for palatalization, which would let you be much more uhh.. sensible.. with how you write vowels.
3) You seemingly go to such lengths to avoid special characters, but then use uelig for /2Y/? Really? Fucking uelig??? That letter barely even exists!

The <i> is a special letter that isn't considered a vowel in my conlang. After consonants, it makes the consonant palatalized. After a vowel, it lowers the vowel. I don't want to use <j> instead of <i> just for aesthetic reasons. As for <ee> and <oo>, that is how you would represent a long vowel. If I did that for /i/ and /u/, their lengthened version would be <eeee> and <oooo>. And as for the ue ligature, I orginally just had ae, oe, and ue for the diphthongs. And just because the ligatures conveniently exist, I decided to use them. the ue ligature not being used virtually at all doesn't bother me as my goal isn't really to look or seem as a natural language.

I used to have the palatalize consonant denoted with a dot on top, but I didn't to rid of that. Also, with the lowered vowels, I used to originally denote it through having two of the same consonant after it, then I changed it to have a grave accent on top. Personally, I sort of like the look better with the accent marks.

[Xephyr]

The <i> is a special letter that isn't considered a vowel in my conlang. After consonants, it makes the consonant palatalized.

Yes I know. That's what I said.

After a vowel, it lowers the vowel.

But even your high vowels are marked oddly! It would be better if, say, you wrote /e/ with <e> and /i/ with <ei>, because <i> at least implies raising.

As for <ee> and <oo>, that is how you would represent a long vowel. If I did that for /i/ and /u/, their lengthened version would be <eeee> and <oooo>.

Not necessarily. I've seen one grammar (can't remember what... Yurok maybe?) that uses <ee> for /i/ and <eee> for /i:/.

I used to have the palatalize consonant denoted with a dot on top, but I didn't to rid of that. Also, with the lowered vowels, I used to originally denote it through having two of the same consonant after it, then I changed it to have a grave accent on top. Personally, I sort of like the look better with the accent marks.

IMO, any of those solutions would be preferable to what you're doing now.

[Grimalkin]

Here's the phonology of my latest conlang, tentatively called Cembes:

Consonants:


Vowels:


One important thing to note is that the phoneme inventory is pan-dialectal. Cembes has three main dialects which are quite divergent from each other, yet they have the same basic set of phonemes. The symbol <ˤ> marks pharyngealisation in dialect A, glottalisation in dialect B (glottalised plosives are ejective), and pre-glottalisation in dialect C. It is in fact a remnant of an earlier tonal system - pharyngealisation/glottalisation is the descendant of what was once a rising-falling tone with secondary glottalisation.

You'll also notice that /p/ is curiously absent. The reason is that /p/ shifted to /h/ (and /pˤ/ shifted to /ʔ/)

Phonotactics:

Syllable structure: (C)V(C)(C)

> /h/ and /tɬ/ are restricted to the syllable onset.
> Permitted coda clusters are /l, r, ʁ/ + [plosive, affricate, nasal]
> Any non-glottal consonant except /j/ can be geminated.

Stress:

> The last heavy syllable takes primary stress. A syllable is heavy if it contains a long vowel, diphthong or coda consonant.
> The central vowels /ɨ/ and /ə/ repel stress. A syllable containing one these vowels is never stressed (unless it is the rightmost syllable in a word containing only these vowels)

Example words:

bagusˤ - dog (absolutive)
baguš - dogs (absolutive plural)
baguškə - at the dogs (locative plural)
dəik - new
dəigdəik - very new
kaʁn - sun
qā - to fall
qāčə - he/she fell

[txmmj]

But even your high vowels are marked oddly! It would be better if, say, you wrote /e/ with <e> and /i/ with <ei>, because <i> at least implies raising.

Actually, my original plan was to have the vowel rise rather than lower. However, since I have <e o u y> representing /j w ɥ ɰ/, I would need to use <ei oi ui yi>, which I'm trying to avoid. So that's why I had the original main vowel be the heightened version and then they lower by adding an <i> and lower slighty, but not fully, as the end of a word and unstressed. Plus, I never said I was trying to make it completely phonetic.

Not necessarily. I've seen one grammar (can't remember what... Yurok maybe?) that uses <ee> for /i/ and <eee> for /i:/.

If <e> = /e/, and <ee> = /i/, and <eee> = /i:/, what about /e:/?

IMO, any of those solutions would be preferable to what you're doing now.

Well, I originally used the <i> for palatalization then came up with the idea that if this language were real, the people who spoke it would often just write a dot above the consonant rather than constantly write the <i> out. So either way they were connected, though I decided to revert back to just using the <i>. As for the vowels, I'm still trying to work things out. I honestly didn't like the look of having them being lowered due to two consonants following it. And also I wasn't sure how I liked the grave accent mark.

In any case, thanks for your ideas and opinions though! If you have any other ideas how to keep the same basic phoneme inventory while making things make more sense or look better, let me know more!

[Click]

Aihmi

Aihmi phonology.png (25.64 KiB) Viewed 6072 times

What do you think about the phonology.Haven't figured the allophony yet,but schwa is an allophone of /ɪ/ in unstressed syllables.Do you have any suggestions?
The consonant clusters of one syllable must adhere to the sonority hierarchy:
stop
affricate
nasal
fricative
flap
glide
For example /pts/,/pm/,/ps/,/pɾ/ and /pw/ are permitted,but /tsp/,/mp/,/sp/,/ɾp/ and /wp/ are not permitted unless in stressed syllables.
The consonant clusters at same MoA and PoA are not permitted anywhere.If they occur due the inflections,they assimilate to a voiced geminate at that MoA and PoA.
Example words in the orthography:
aihmi
zur
ahsemageh
txeda
urwami
The second syllable is stressed.Monosyllabic words are unstressed.

[finlay]

you have /tdzeda/ as a possible word? uh, o..k.. what I would do is count the affricates and stops as one plosive category in the sonority hierarchy.

There's nothing particularly wrong with it apart from that – you have a fairly normal set of consonants, for one thing. Your vowels are perhaps a little bit strange – there's not any apparent rhyme or reason to them – and you have a couple of odd orthographic choices there (by which I mean, the digraphs and trigraphs with h, although I suppose it's a neat way to avoid diacritics). /ɪ/ would look better if you marked it as a central vowel, at least in terms of phonology.

[Click]

you have /tdzeda/ as a possible word? uh, o..k.. what I would do is count the affricates and stops as one plosive category in the sonority hierarchy.
There's nothing particularly wrong with it apart from that – you have a fairly normal set of consonants, for one thing. Your vowels are perhaps a little bit strange – there's not any apparent rhyme or reason to them – and you have a couple of odd orthographic choices there (by which I mean, the digraphs and trigraphs with h, although I suppose it's a neat way to avoid diacritics). /ɪ/ would look better if you marked it as a central vowel, at least in terms of phonology.

Point taken.
Revised sonority hierarchy:
stop/affricate
nasal
fricative
flap
glide
I have decided that i don't like the sound of /ɪ/ so I've replaced it with the schwa.The orthography for the schwa is still <eh>
The orthography is like that cause of sound changes-200-300 years ago /ɐ/ was allophone of /ä/ before /h/,then it became phonemic after the elision of /h/.The h is still preserved in the orthography.
How <aih> and <eh> got there:
/æh/ > /æ/ because of the elision of /h/.
/ə/ was allophone of /e/ in the unstressed syllables.It later became phonemic and the /h/ elided.So /əh/ in unstressed syllables > /ə/ which can appear also in the stressed syllables.

[finlay]

Ugh, i don't like it when people use /ä/ as a phoneme symbol. Just call it /a/, especially when you're making a chart anyway. As for the other one, is it ɐ or ɒ? Because they're different. Again, it's fine to have a rounded vowel and call it ɒ even though it's actually central, or whatever.

Also, if it's a sound-change-motivated orthography choice, why don't you have a ih, oh, and uh? Surely your older form of the language would have had those?

[Click]

The vowel is /ɐ/,not /ɒ/.Sorry for the confusion.The Segoe UI italic sucks.

[popisfizzy]

I have two phonologies I feel like showing off. The first is for Lang6, with a WIP name of Eníš brűcedźȅ. This language is deliberately-weird, and also unstable. Basically, in the setting the language was in, a number of individual, bizarre features managed to conglomerate all at once. The prestige dialect also managed to fossilize some of these features, while most dialects are phasing them out. Basically, in a hundred or so years, there will be a lot of dialects with a lot of variation, some undoubtedly being mutually incomprehensible.

Another is lang10, which doesn't even have the conceit of a WIP name. Somewhat-detailed phonology, though. Also, a little less back-story than Lang6. In my current thought, it's going to be an in-setting constructed language (divinely-inspired), so it will probably be fairly regular.

What I have taken from these is that I work myself out on making a detailed phonology, and then get bored with everything else. Huzzah~~~

[Nortaneous]

Got bored and farted out a speedlang and script. Don't have my notes with me, so I'm working from memory here, but.

/p_h p t_h t k_h k/ <ph p th t kh k>
/b_h b d_h d/ <bh b dh d>
/s S G h/ <s ś g h>
/m n l L r/ <m n l λ r>

/a e o i ai/ <a e o i y>

Phonotactics:
* only word-final consonants allowed are /m n d s h/, but word-final /m/ is realized as [N\] or nasalization
* all word-internal clusters end in an unaspirated stop or /s S G/, and most dialects drop /G/ as the second element in clusters. There's a chart of which clusters are allowed, but I don't have it with me. Voicing contrast doesn't exist in clusters.
* no initial clusters, no clusters of more than two consonants

thinking about having some complicated umlaut system for the vowels, but I can't be bothered right now

[maıráí]

[[See two posts down.]]

[Whimemsz]

That post is utterly meaningless without more detail and examples. "Harmony" is extremely broad; it works differently in different languages, and has different domains in which it operates, and different triggers. To truly adequately describe it (if it's realistic) you're going to need to give a lot lot more information. (Also "aspiration harmony" makes no sense if you're applying it to a change in place of articulation.) Your post is unclear as to which distinctions are phonemic and which are not. You don't address phonotactics and syllable structure (and how this might interact with the harmony rules). This is why just posting a list of sounds and thinking you've made up a phonology is so stupid. People need to wean themselves of this habit.

[maıráí]

I feel like I killed this thread. It's gotten no replies in ages. T_T



IPA
(romanization)

Consonants

Nasals:
m n ɲ
(m n ng)

Plosives:
p b t d k g

Fricatives:
f v s ʃ h*
(h v s sh h)

Affricates:
ts tʃ
(c ch)

Approximants:
l j

f~h are allophones. The f is very open.

Voicing harmony between p t k and b d g.
Aspiration harmony between s c and sh ch. (There are many S's that are not subject to harmony; these are "strong S words", and are typically older words.)

Consonants can be medium or long (n, nn). Long consonants mostly occur in combinations.
moon (soup) + naka (drinking water) = moonnaka (thin soup)
They are also required in names;
anemone flower field > Anemonne (ends in CV; nothing is added)
rus red > Russè (ends in C; schwa added to easy pronunciation)
aee air > Aeehhè (is only one CV; default consonant H and schwa added)

In word formation, and even across word boundaries, t+C usually becomes a long C.
mot (word) + kauuk (large) = mokkauuk (phrase)
Y mot, se jooi. (Say this word.) >> y mos:e jo:i
(Although this phrase is just as likely to be pronounced "Y moce jooi".)

Ng only occurs finally in a syllable (teng, sjong, etc.). It also combines with the vowel into a nasal vowel in casual speech.

Vowels

Front Vowels:
a~æ ø y
(ä ö y)

Back Vowels:
ɑ o u
(a o u)

Neutral Vowels:
i e ə ɚ
(i e ə/è ər/èr)

Diphthongs:
(ao au ai ae; äö äy äi äe
oa ou oi oe; öä öy öi öe
ua uo ui ue; yä yö yi ye
ea eo eu; eä eö ey; ei)

Other Combinations:
ju jy je ja jo ja~jæ jø
(ü ÿ ë ja jo jä jö)

Vowels have front/back harmony.
Vowels can be short or long. (a, a:)
Long diphthongs can be early (so:i) or late (soi:).

Harmony
All applicable consonants in a word must be voiced or voiceless OR aspirated** or unaspirated**; harmony is signaled by the first consonant, and covers the entire word and occasionally the preceding or following particles.
Likewise, all applicable vowels in a word must follow the first front or back vowel. Neutral tones are ignored; vowel harmony is not set until the first appears.

**(I forgot that it's not aspirate in the true superscript-h sense, sorry. The name is misleading; it was convenient.)

Examples:
dää (book) + sata (warehouse) = dääsädä (library)
front, voiced + back, unvoiced = front, voiced
shëë (rice) + suka (caramel) = shëëshuka (rice syrup)
"aspirate"+ "unaspirate" = "aspirate"
ha sata a building
back article, back noun
hä dää a book
front article, front noun
dini (palm) + kaas (hit) = dinigaas (slap)
neutral + front = front
giid (copy) + dääyg (large text) = giiddääyg (lots of notes)
neutral + back = back
deje (deity) + tene (hand) = dejetene (head priest)
neutral + neutral = neutral



Stress is on the first syllable, although most functional words/particles are unstressed. Tone is falling, excepting early and late diphthongs (which evolved from an abandoned tone system). These rise lightly with the first vowel, and fall with the second vowel.
Tone rises on the last phrase in informal questions; otherwise, rising tone also indicates a sort of break in speech, like that required in a copula phrase:
Puluh, sang.
puluh (rising) sang (falling) OR puluh (pause) sang
blue-MOD [COP] [DEF] water
The water is blue.


Syllable forms are:
V, VV
a, ao
aa, aoo, aao
CV, CVC, CjV, CjVC
ta, tak, tja, tjak
taa, taak, tjaa, tjaak
CVV, CVVC, CjVV
tao, taok, tjao, tjaok
taoo, taao, taook, taaok, tjaoo, tjaao
*CVCV
taka, tako, tato, toto
Most often same vowels, different consonants.


Not sure where to put:
CVCV often collapses to CCV following a vowel, only in-word.
shëëshuka > shëësh'ka
but shëë shuka > shëë shka only in casual speech.

[maıráí]

Here's a sample of the current lexicon:
(Alternate Romanization Note:
ju, jy, je = ü, ÿ, ë
w = one of many schwa symbols.)

hime eye
digi tooth
tene hand, bit
dini palm
sore limb
bolo foot, paw
tenesore arm
bolosore leg
dinisore outstretched arm ("Dinisores." Give it here. lit. To the arm.)
sju sugar
mong soup
shjee rice
nang water
naka drinking water
nakamong liquid part of soup, after cooking
nangmong water base of soup, before cooking
pete rock
tuul fire
tuulpete coal
kisa cat
nele girl
nölö boy
bääs home
gauug big
sumo cloud
kwr spear
kwr thunder
sumotuul, kwrkwr lightning
dööe hate
hei daytime
heii pretty
heei pleasant


And I'll write up something on the Valley accent soon.

[Skomakar'n]

I actually thought of Uralic as soon as I saw hime despite it being pretty different from, say, Finnish silmä already, and then I also saw kisa and tuul. On the other hand other basic words like nang and tene show no Uralic similarities and pete looks an awful lot like a Romance word. What kind of language is this?

[----]

I feel like I killed this thread. It's gotten no replies in ages. T_T

I actually have a couple basic phonologies but they're not complete enough to be presentable.

[maıráí]

I've got another phonology that I want critiques on, but it's woefully incomplete as well.

What kind of language is this?

It was initially a priori, but early on I got interested in Finnish, and so there are quite a few borrowings (kisa and tuul), as well as some patterns (CiCa, kVsV) that are favored because of it; the vowel system was originally identical to Finnish.

[Shrdlu]

While playing CIV1 and 2 I though up an language for the in-game Sherden(the barbarians that attack you from the sea occasionally).

The Harapita language - ei Harapitam emje.
/p[p] pv[p_>] t[t] tv[t_>] tt[t:] k[k] kk[k:]/
/m[m] mm[m:]/
/r[r\] r[r] l[l] ll[l:]/
/f[f] s[s] ss[s:] z[z] zz[z:] h x[K] xv[tK] c[G]/
/j[j] jj[j:]/

/a e i o u y å ä/ /ei ai oa ea ou/


==syllable
(V)C(V+/m mm/ +/p t k r s z j/)V
Only /p/ or an vowel can end a noun. /p/ is part of an older suffix meaning "to have" which has been switched to an freestanding construction.
Except /h/, all fricatives can not begin or end a noun
/j/ can only begin a noun
digraphs can not begin a word.

small vocab'
Mi I
Milli You
a mi myself
a milli yourself
<note /+m/ or /+ma/ before an other nasal is the genitive.>

[Nortaneous]

Think I've mostly settled on a Hanheliubl orthography, so I can post this now.

Code: Select all

pʰ p tʰ t ʈʰ ʈ tʃʰ tʃ kʰ k ʔ    p  b t  d ṭ  ḍ ch ǧ  k   g  '
m̥  m n̥  n ɳ̥  ɳ ɲ̊   ɲ  ŋ̊  ŋ      mh m nh n ṇh n ňh ň  ngh ng
f  v s  z ʂ    ʃ   ʒ     ʁ h    f  v s  z ṣ    sh zh     gh h
   w         ɻ     j               w         r    y
        l          ʎ                 l            ly

i iː ɯ ɯː u uː                  i ii ɨ ɨɨ u uu
e eː ɤ ɤː o oː                  e ee ə əə o oo
æ æː      ɑ ɑː                  a aa      ɔ ɔɔ

æɪ æːɪ ɒʊ ɒːʊ iʊ iːʊ            ai aai au aau iu iiu

Subject to change depending on what I do with the sound changes; they kind of suck shit, so.

Sample:
/ʃikæːʈʰ ɲ̥eː ef ʃeːɻeʁ mŋæː mŋæː | oʒoŋ ef feːʃæʁ hætʰæʁes mŋæː dlæː | tʰfætsiʊ̯n ef iːʊ̯veːʁ noʈɤs | kʰeːʁe miʊ̯n̥es mŋæː kʰæʔæʔæːʈ | ɲ̩eː ef jæːʃeʁ mŋæːk ʈʰ/
Shikaaṭ, ňhee ef sheeregh mngaa mngaa. Ozhong ef feeshagh hataghes mngaa dlaa. Tfadsiun ef iiuveegh noḍəs. Keeghe miunhes mngaa ka'a'aaḍ. Ňhee ef yaashegh mngaak ṭ.

And in the old orthography, this was:
Šikāṭ, ňhē ef šēreħ mŋā mŋā. Ožoŋ ef fēšaħ hataħes mŋā dlā. Tfadsiun ef īuvēħ noḍẹs. Kēħe miunhes mŋā kaʔaʔāḍ. Ňhē ef jāšeħ mŋāk ṭ.

Another thing I'm thinking of doing is:

Shikaath, nhyee ef sheeregh mngaa mngaa. Ozhong ef feeshagh hataghes mngaa dlaa. Tfadsiun ef iiuveegh nodheys. Keeghe miunhes mngaa ka'a'aad. Nhyee ef yaashegh mngaak th.

[Herr Dunkel]

I think my conlang's phonology has been perfected sufficiently enough to be presentable, so I'll present it for criticism.

Sounds
Kti has 24 phonemes, twelve monophthongs vowels and as many consonants.
Nearly every phoneme has allophonical equivalents in certain enviroments, and the language possesses a small variety of morphophonological processes that restrict certain clusters and whatnot.

Consonants

The consonant phonemes in Kti are spread out through the mouth as such:

Code: Select all

        t  d          k      ʔ
        s  z   ʃ  ʒ   x
   m      n
          r

The phonemes are stabile and unchanging, but their phonetic realisation is quite diverse:

Code: Select all

        t tʲ  d dʲ   ʈ  ɖ     kʲ    k       q     ʔ
        s sʲ  z zʲ   ʃ  ʒ     xʲ     x       χ     h
   m      n  nʲ
          r  rʲ

I will later on explain the correspondences between the phonemes and phones.

Vowels
There are officially twelve vowel phonemes, six short and six long ones:

Code: Select all

i  i:            u  u:
ɛ  ɛ:   ɞ  ɞ:    ɔ  ɔ:
                 a  a:

These remain pretty consistent in quality, except that /i:/ and /u:/ respectively become /j:/ and /w:/ word-initially.

There are five diphthongs in Kti:

Code: Select all

  Grapheme    Phoneme

     ai         ɒy
     ui         uy
     ua         uʌ
     ia         ya
     iæ         jɞ

They are counted as one long vowel and are invariable in quality.

In addition, there are three triphthongs in Kti:

Code: Select all

  Grapheme    Phoneme

    eia         eʉɑ
    aie         ɑɨɞ 
    uiæ         uʉɞ

They are counted as an "overlong" vowel and are invariable in quality.

Allophony and Morphophonology

Palatalisation
Ktarh palatalisation is an enviromental feature.
Consonants are realised unpalatalised before /a(:)/, /u(:)/, /ɔ(:)/ and /ɞ(:)/ and consonants, while palatalisation is sometimes heard before /i(:)/ and /ɛ(:)/. This palatalisation is neutralised in word-initial positions.
Only /t d s z k x r n/ palatalise.

Assimilation
Ktarh POA assimilation is a process in which /t/ and /d/ become realised as [ʈ] and [ɖ] before any of /ʃ ʒ r/.
This rule can be written as such:
t d > ʈ ɖ / __{ʃ ʒ r}

The resulting "retroflex" consonants are laminal in quality.

Uvularisation
Uvularisation is a strictly enviromental feature, where /k/ and /x/ become /q/ and /χ/ respectively before /u:/.
It does not apply word-initially.

Anaptyxis
Anaptyxis is a regular morphophonological process in Kti, which adds vowels during morphological processes.
It is not directly predictable, as each affix has its preferred anaptyxic vowel, and some have an obligatory one, but essentially anaptyxis adds a vowel during affixation as to prevent syllable structure violation.

Phonotaxis

Each vowel in Kti has to have at least one consonant next to itself.
This structure is the basis of the "syllable" in Kti.

It is extended by:

• *There can be at most two consonants or one special cluster preceeding the vowel, unless there is another "syllable" before it
  *Each vowel can have at most one consonant following it
  *In addition to the above rule, the exceptions are clusters of /m n r/ + /ʔ/, which are allowed "syllable"-finally

The special clusters mentioned are /sʔmn/, /sʔn/, /kʔn/, /zʔmn/ and /zʔn/. They can occur before a vowel, taking up the room of two consonants.
Note that the last two clusters occur only due to morphological processes.
Triphthongs tend not to have consonants after them, but this is not an ultimatum.

There are several rules as to what can come together:

• 1."Z" and "Sh" cannot be next to any fricative but can be near other consonants
  2."T" cannot be preceeded by a nasal
  3."T" cannot be followed by any plosive except the glottal stop
  4."D" can be followed by all consonants except alveolars
  5."A" cannot be next to "Æ" except when either vowel is a part of a greater whole (such as a diph- or a triphthong).
  6.Two of the same phoneme cannot be together in roots (this causes gemination)

Note that in the above rules, graphemes are used instead of phonemes, since graphemes represent different phonemes in different situations.
Going by this, we have the following combinations:

/md/ /mk/ /mʔ/ /ms/ /mz/ /mʃ/ /mʒ/ /mx/ /mr/ /mn/ /nd/ /nk/ /nʔ/ /ns/ /nz/ /nʃ/ /nʒ/ /nx/ /nm/
/nr/ /st/ /sd/ /sk/ /sʔ/ /sʒ/ /sx/ /sm/ /sn/ /sr/ /zt/ /zd/ /zk/ /zʔ/ /zm/ /zn/ /zr/ /xt/ /xd/ /xk/ /xʔ/
/xs/ /xʒ/ /xm/ /xn/ /xr/ /rt/ /rd/ /rk/ /rʔ/ /rs/ /rʃ/ /rx/ /rm/ /rn/ /tʔ/ /ts/ /tz/ /tʃ/ /tʒ/ /tx/ /tm/ /tn/ /dʃ/
/dʒ/ /dn/ /dm/ /dr/


/ae/ /au/ /aɔ/ /aɛ/ /ua/ /uɛ/ /uɔ/ /uɞ/ /ɛi/ /iɛ/ /iu/ /iɔ/ /ɔa/ /ɔɛ/ /uɛ/ /uɔ/ /uɞ/ /ɞa/ /ɞɛ/ /ɞi/ /ɞɔ/ /ɞu/
Vowel combinations apply to long vowels as well.

There is no voicing assimilation in Kti.

Stress

Prosodic stress is very syllable-dependant and there are relatively complex rules that are governed by syllable position and structure.

Stress is pretty regular in that it is generally found in the penultimate syllable unless something else happened. If the word is monosyllabilic, the stress is on the sole syllable.
A syllable is counted as "light" if it has a short vowel, and "heavy" if it has a long vowel, diphthong or triphthong.
If a stressed light syllable is in between an unstressed light and an unstressed heavy syllable, the stress shifts to the heavy syllable.
If both a heavy penultimate and a heavy antepenultimate exist, the stress shifts to the antepenultimate.
If a heavy syllable is surrounded by two light syllables, one of which is stressed, the stress will always be on the first.

Example text:

Zok, naunem sonakasem mohitakeha, ænē ordamai s'nosur' shumer. Zhotīrem nē okuanem oktarum; toh karem soreteuenam. K'nazhiton-mū'irem shusi, tī-mū'irem denkæse. Æsorēētrataiton, shōnem...

Really, I was walking down the street, then a sound was heard behind me. I turned myself then saw a man; he appeared evil. He periodically looked at me, periodically looked around. With him drawing a knife, I screamed...

[zɔk | 'naunɛm sɔna'kasɛm mɔxʲita'nɛka | 'ɞnɛ ɔrda'mɒy 'sʔnɔsurʔ 'ʃumɛr ‖ ʒɔ'ti:rɛm nɛ: ɔ'kuʌnɛm ɔ'ktarum | tɔx 'karɛm sɔrɛtɛu'ɛnam ‖ kʔna'ʒiton‿mu:'ʔirem 'ʃusi | 'ti:‿mu:'ʔirɛm dɛn'kɞse ‖ ɞsɔr'ɛ::.ɛtra'tɒytɔn | 'ʃɔ:nɛm]

[roninbodhisattva]

An inventory I came up with today in class. It's supposed to be reminiscent of languages in Northern California.

Consonants

Code: Select all

p     t            tʃ      k      kʷ
p'    t'           tʃ'     k'     k'ʷ    ʔ
      s       ɬ    (ʃ)                   h
      ɹ, ɾ    l     j      ɣ      w    
m     n                    ŋ         

Vowels

Code: Select all

i     u     e     o     a     ə, ɚ
iʔ    uʔ    eʔ    oʔ    aʔ    əʔ
ie    uo    ea    eo
ia    ua
i:    u:                a:

Still working on syllable structure but it will probably be something akin to CV(G)(C(C)), where G is a glide/resonant.

[Ouagadougou]

Thalase-ne (Thalassian)

Consonants:

Code: Select all

                        Labial          Dental-Alveolar          Palatal          Velar          Glottal
Stops                   p pʰ b          t tʰ d                                    k kʰ g
Fricatives              f               s                                                        h
Taps                                    ɾ
Approximants                            l                        j
Nasals                  m               n                                        (ŋ)*
Coarticulated           w (ps)**        ts                                        kʷ ks

* Combinations of /n/+Velar, or Velar+Velar, could have resulted in [ŋ].
** Clusters of the form Labial Stop+/s/ have been encountered on even the earliest fragments, and [ps] would add symmetry to the stop system.
Vowels*:

Code: Select all

        Front     Central     Back
High    i ɪ
Mid     e ɛ      (ə)**        o
Low               a           ɑ

* Since Thalassian is an extinct language, the vowel qualities remain somewhat of a mystery. The vowels are usually transcribed i, ī, e, ē, a, ā, o, but whether the distinction was of length alone is impossible to infer.
** The short and long /o/ merged in Thalassian early on, but there is reason to believe that unstressed /o/ may have been a schwa, as it is regularly deleted in certain inscriptions.

[Thry]

Nice! Did you know Thalassian is already the name of a language? It's not a real conlang, but it's the name Blizzard gives to the dialect of Darnassian that Blood Elves speak in Quel'Thalas in World of Warcraft. Good name, imo.