Post your conlang's phonology

[TheWeaver]

Hey guys, here's a rough outline of a phonology I may use. First time posting a real phonology - I hope this works. Thoughts? Also, if '#' is not the end of a word, feel free to let me know which symbol i should use

Plosives : /t, t^w, t^h, t^j, k, k^w, k^h, k^j, q, q^w, q^h, ʔ/
/*Note: the glottal stop is a grammatical device*/

Trills/Taps : /r, ɾ/

Fricatives : /θ, ð, ʃ, ʒ, x, χ^uvular, h/
/*Note: Seems that the velar fricative 'x' and the uvular fricative 'χ' are both printed 'x', so I had to make due.*/

Approximates & Glides : /ɹ, j, w, l, ɫ/
/*Note: the approximate 'ɹ' and the trill/tap 'r'/'ɾ' are distinct*/

[t, k, q] > [ t^w, k^w, q^w] / _ [+round]
[t, k, q] > [ t^h, k^h, q^h] / _ [h]
[t, k] > [t^j, k^j] / _ [j]

[l] > [ɫ] / _ [/k, q, χ^uvular/]
/ɾɾ/ > [/r/]

Vowels:

Code: Select all

i,i:  ɨ,ɨ:/ʉ,ʉ:  u,u:
    ɪ,ɪ:     ʊ,ʊ:
ɛ,ɛ:    ɜ,ɜ:   ʌ,ʌ:
        ɐ,ɐ:

[i, i:, ɛ, ɛ:] > [ɨ, ɨ:, ɜ, ɜ:] / _ [/k, q, χ^uvular/]
[u, u:] > [ʉ, ʉ:] / _ [+front]
[/ii/, /uu/, /ɪɪ/, /ʊʊ/, /ɛɛ/, /ʌʌ/, /ɐɐ/] > [i:, u:, ɪ:, ʊ:, ɛ:, ʌ:, ɐ:]
[i, i:, u, u:] > [ɪ, ɪ:, ʊ, ʊ:] / _ # {whatever the end-of-word symbol is. Maybe '$' or '%'?}

/t, k, q/ <t, k, q>
/ɾ, ɹ/ <ŕ, r>
/θ, ð, ʃ, ʒ, x, χ^uvular, h/ <vh, v, zh, z, xh, x, h> /*Note: this romanization is legal because no fricative can touch another*/
/j, w, l/ <j, w, l>
/i, u, ɪ, ʊ, ɛ, ʌ, ɐ/ <i, ú, y, o, e, u, a/

C : Consonant, P : Plosive, F : Fricative, G : Tap/Trill/Lat./Approx, V : Vowel
(C)(V)V(P, G)

/*Note: This syllable structure is 'safe', but I may expand it to
* (C)(V)V(P)(G), and add a style of assimilation between 3-consonant clusters (i.e. Inukitut). Opinions? */

Thanks everyone!

[Drydic]

First thing I would suggest is not putting /*slashes and asterisks around your notes.*/ It's not really necessary, if you'd like some kind of formatting on them, I'd recommend italics, which are produced by using [ i ] and [ /i ] without the spaces. The above produces and , for example. Also those same slashes are generally used in linguistics for phonemes, while by what you've described below as t^w/h/j, k^w/h/j, q^w/h[/], and ɫ aren't phonemes, but allophones. It's perfectly fine for those rules to exist, they just aren't usually written in phoneme lists (even if orthographies do show them).

Hey guys, here's a rough outline of a phonology I may use. First time posting a real phonology - I hope this works. Thoughts? Also, if '#' is not the end of a word, feel free to let me know which symbol i should use

# is used for both the beginning and end of words, actually.

Fricatives : /θ, ð, ʃ, ʒ, x, χ^uvular, h/
/*Note: Seems that the velar fricative 'x' and the uvular fricative 'χ' are both printed 'x', so I had to make due.*/

They show up as different to me. Perhaps you're using a font without Unicode support?

Vowels:

Code: Select all

i,i:  ɨ,ɨ:/ʉ,ʉ:  u,u:
    ɪ,ɪ:     ʊ,ʊ:
ɛ,ɛ:    ɜ,ɜ:   ʌ,ʌ:
        ɐ,ɐ:

Just to make it easier, you could just write each symbol once and say "each vowel can be long (indicated with V:)".

I'm too tired to judge the rest fairly.

[Pole, the]

Fricatives : /θ, ð, ʃ, ʒ, x, χ^uvular, h/
/*Note: Seems that the velar fricative 'x' and the uvular fricative 'χ' are both printed 'x', so I had to make due.*/

They show up as different to me. Perhaps you're using a font without Unicode support?

Or, you use a wacky unicode font, like Trebuchet.

[Drydic]

Fricatives : /θ, ð, ʃ, ʒ, x, χ^uvular, h/
/*Note: Seems that the velar fricative 'x' and the uvular fricative 'χ' are both printed 'x', so I had to make due.*/

They show up as different to me. Perhaps you're using a font without Unicode support?

Or, you use a wacky unicode font, like Trebuchet.

Ok, looked at Trebuchet, it's not unicode. it has Latin-1, Central and Eastern European, Greek, and Cyrillic. And Khi does indeed look identical to x >:|

[TheWeaver]

Fricatives : /θ, ð, ʃ, ʒ, x, χ^uvular, h/
/*Note: Seems that the velar fricative 'x' and the uvular fricative 'χ' are both printed 'x', so I had to make due.*/

They show up as different to me. Perhaps you're using a font without Unicode support?

Or, you use a wacky unicode font, like Trebuchet.

Ok, looked at Trebuchet, it's not unicode. it has Latin-1, Central and Eastern European, Greek, and Cyrillic. And Khi does indeed look identical to x >:|

Thanks guys, figured it out. The problem seems to be something concerning firefox & this site. In firefox, a checkbox in the options called 'Allow pages to choose their own fonts' is by default checked. Unchecked it, and Times New Roman pops up, as well as the correct characters! I posted in the beginner's sub forum at CBB about this phonology (yes, I have no clue how to use the account here on both sites).

[Drydic]

The font the ZBB requests browsers use is Verdana, which works fine. Also Time New Roman is ugly, check out FreeSerif, same style but much cleaner. And free to boot.

Also the CBB is an entirely separate forum, so there is no way to use the same account on both sites.

[Tropylium]

OK, this is a little thing that fell out of my head after a session of transcribing Ob-Ugric data. I probably won't be doing much more with it, but here you go anyway.

Consonants
The phonetic segment inventory is:

Code: Select all

[ p pʲ t ts~tʃ  tɕ k    kʷ ]
[ b bʲ d        dʑ g       ]
[ m mʲ n      ȵ    ŋ       ]
[ f fʲ ɬ ɬʲ s ʃ ɕ  x xʲ xʷ ]
[ v vʲ      z   ʑ  ɣ       ]
[      l lʲ     j    ɥ  w  ]

Not all contrasts here are phonemic.
• Palatalization and labialization are non-contrastive in any position other than word-initially (and even there, palatalization mostly correlates to vowel backness). The alveolo-palatals count as the palatalized allophones of the alveolars, likewise [j ɥ] as [ɣʲ wʲ]. Word-internally, consonants are realized as palatalized in front-harmonic words. Word-finally, palatalization is realized only weakly.
• The contrast between voiced fricatives and voiced stops/affricates is also non-contrastive word-internally: [v vʲ z ʑ ɣ] appear between vowels, while [b bʲ d dʑ g] appear in consonant clusters. [z ʑ] also appear preceding a consonant.
• Also the contrast between [ɬ ɬʲ] and [l lʲ] is non-contrastive word-internally, with the approximants appearing between voiced sounds, the fricativs adjacent to a voiceless sound (including in geminates).
• Syllable-final stops are unreleased.
• It would be possible to analyze initial [v] and medial [w] as the same phoneme, but I'm not going to.

For a practical orthography:

Code: Select all

< p t/c č k kw >
< b d     g    >
< m n     ŋ    >
< f ł s š x xw >
< v l z   g w  >

Palatalized consonants before back vowels written ‹Ci›; non-palatalized consonants before front vowels written ‹Cr›. /tɕ/ gets a symbol of its own because why not, there's already č.

Vowels

Code: Select all

i y u  iː    uː /< i ü u  ī   ū
       ea øa oː /<        ē ô ou
ɛ   ʌ  ɛː    ɔa /< e   o  â   ō
   a         ɑː /<    a       ā

(The transcription could be better, IDC.)
• Vowel harmony applies: vowels must be followed with vowels of the same backness. In suffixes, alternations between e~o, i~u, ī~ū, and also ī~ou thus appear. a ā count as neutral, however.
• â ē ô ō cannot appear in non-initial syllables.

Phonotactics
• Syllable structure: (C)CV(C). Null onsets not allowed.
• The only allowed word-final consonants are m n ŋ, p t k, f s š x. Inside a word, also b l~ł z g can appear syllable-finally.
• Labialized consonants cannot appear before close vowels.
• Palatalized consonants cannot appear before /u ɔa/ (a shift to /y øa/ has historically happened here).
• Voiced palatals/velars /g~ɣ~j/, /ŋ/, /w~ɥ/ and voiced sibilants [z~ʑ] do not appear word-initially.
• In syllable-initial clusters, (1) the consonants must agree in voicing and nasality, and (2) the first consonant must be pronounced at a more frontal POA than the second.

And for a test drive, here's a Swadesh list:
possuŋ "all"
münckas "ash"
nōpout "bark"
xretten "belly"
dīfiš "big"
vāk "bird"
sulem "to bite"
lexīš "black"
łēt "blood"
dougāŋ "bone"
łčandaŋ "breasts"
skoulom "to burn"
nutkon "claw"
boŋkan "cloud"
vekkiš "cold"
tūmom "to come"
sinem "to die"
vāwom "to drink"
kuššuš "dry"
xosmon "dog"
kwāf "ear"
nrefeŋ "earth"
cülbem "to eat"
łotout "egg"
bliŋ "eye"
mam "fat"
tōłfot "feather"
čôndian "fire"
miat "fish"
vikkeŋ "flesh"
kwokom "to fly"
fīt "foot"
poum "full"
łôŋkom "to give"
boškap "good"
vōzūš "green"
fiamaŋ "hair"
skūp "hand"
saftom "to hear"
fxen "heart"
nugon "horn"
pūkuš "hot"
vānot "I"
dovom "to kill"
trep "knee"
dem "to know"
łâzet "leaf"
łkawom "to lie"
xūmom "liver"
cexīš "long"
xâgiat "louse"
niāpkāp "man"
kwałłās "many"
düł, düłciŋ "moon"
vimbet "mountain"
tkum "mouth"
kōs "name"
brekłen "neck"
mubluš "new"
sāp "night"
čüzben "nose"
trɛmen "not"
xanāk "one"
fotkout "path"
niāvot "person"
šōzok "rain"
sāntuš "red"
diouton "root"
fūloŋ "salt"
ceŋget "sand"
voupem "to say"
kwessem "to see"
xamfaŋ "seed"
tuglom "to sit"
šam "small"
lukon "smoke"
kason "skin"
puzom "to sleep"
keldem "to stand"
faftuk "star"
ptōŋ "stone"
sīteŋ "sun"
divem "to swim"
vełket "tail"
nivāt "this"
dulot "thou"
płagon "tongue"
xuskon "tooth"
vrâkem "tree"
cesiāk "two"
vresem "to walk"
sôveŋ "water"
suŋok "we"
miandān "what"
tasnuš "white"
vuldān "who"
niābnoun "woman"
šâk "wind"
łam "year"
bīš "yellow"
(No points for noticing the verb citation form ending -em/-om or the adjectival suffix -iš/-uš. There are some other morphological things that might be noticable in here too, though.)

[gach]

Here's something I've been toying with for some time now exploring the limits of minimalism. The system is supposed to be as small as possible while still comfortably producing something balanced and believable.

Segments:

/t k ʔ s n/
/a o i/
high vs. low surface tone on each syllabic segment, underlyingly high vs. low vs. neutral

Allophony:

Sibilation:
t > s / _i
(a productive phonemic change really)

Intervocalic lenition:
k > [j] / V_i
k > [w] / V_o
In other intervocalic position /t k/ are become lenis stops but rarely voice.

s > [h] / _#

Nasal assimilation:
The nasal consonant is underlyingly alveolar but assimilates to the POA of following /t k/ [dental, velar],
n > [n̪] / _t
n > [ŋ] / _k
It can also appear as a syllabic consonant in which case it's primarily velar,
n > [ŋ] / [+syllabic]
It dissimilates, however, with the POA of preceding /t k/ so that
n > [ŋ] / t_, [+syllabic]
n > [n] / k_, [+syllabic]
The dissimilation is overruled by the assimilation rule so that /tnt/ and /knk/ are still [tnt] and [knk].

Initial consonant prosthesis:
The glottal stop is distinctive at the beginning of words and contrasts with zero onsets (vowel initial words). In quick speech initial vowels acquire prosthetic onset glides to make the distinction more apparent,
a > [ɦa] / #_
o > [wo] / #_
i > [ji] / #_

Desyllabification:
The potentially syllabic elements /a o i n/ loose their syllabicity in the order n > i > o > a when in contact with each other. Non-syllabic /o i/ become
o > [w] / [-syllabic]
i > [j] / [-syllabic]
Syllabic /n/ surfaces when the nasal isn't in contact with a vowel within the same word. /a/ must always remain syllabic. I still need to pinpoint the exact rules here, but the basic use of the rules is to break vowel sequences with medial /o/ or /i/, e.g. /aoa/ > [awa], /oia/ > [oja].

Values of the vowels:
The vowels don't have any strict conditional quality alternations and are in pretty free alternation. The distinctive features for the vowels are
/a/ [+low]
/o/ [+round, +back]
/i/ [+high]
The vowels try to keep these values contrastive while they might otherwise bend towards the adjacent vowels. For example /o/ is likely to pull nearby /a/ and /i/ further back in the vowel space.

Tones:

Each vowel has one of the three underlying tones which surface as either high or low surface tone. Underlying low is always surface low and underlying high surface high except on the final syllable of words with three or more syllables where it becomes low as well. The neutral tone is low by default but becomes high following a high surface tone. Like the underlying high tone, it's also subject to the final syllable lowering.

Rising and falling contour tones surface routinely as vowels with different tones fall into diphthongs and long vowels. Non-syllabic vowels just discard their tones. The syllabic nasal on the other hand acquires the neutral tone.

Phonotactics:

This is the main place which needs more work. The basic premise is that the language allows moderately complex consonant clusters and vowel sequences. Two consecutive vowels form long vowels and diphthongs but longer vowel sequences fall into different syllables and tend to be broken anyway by medial /o/ and /i/ becoming non-syllabic. There shouldn't ever be sequences of three or more identical vowels, but if there are they will just reduce into simple long vowels with some simple tone shuffling rule.

Clusters of at least up to three consonants at word boundaries and four within the words are allowed. The syllabic nasal counts as a vowel and splits consonant clusters. To get any concrete rules here I still need to look at the historical background. Meanwhile have a handful of random sample words, high surface tone is marked with the accent:

ʔostá
stakʔá
kóskootsn
jákna
josoi
kstá

[Nortaneous]

Here's how I'd do it:

/p t k n/
/a o i/ + high/low tone contrast
(C)V(n)

Allophony:
tí kí > s h / _V _CV́ _#
tí kí > sí hí / _CV̀
ti ki > si hi
no > m / _V _#
no > mo / _C
np nt nk > mb nd ŋ
p t k s h > b d g ɺ ɟ / _V̀
í devoices to [s̩] preceding a low-toned syllable
n > ŋ / _#
ai > e

Potentially syllabic elements lose syllabicity in order i > o > a. Tone applies to whole syllable.

/tínòakía tíoankò nònòakíopàitì píkiòtòn/ [s̩màhá swáŋò mòmwàhóbèɺì ps̩ɟòdòŋ] <smahá soángo momoahóbeli psjodon>

[gach]

That certainly pushes it a bit further. I'm just a bit unconvinced about interpreting changes of the type C₁V > C₂ as purely allophonic rather than productive phonemic alternations. Is there really enough evidence within the system to make such an analysis preferable to just positing a few more phonemes?

Certainly in a system like this you'll end up with a lot of unaltering cases of /s/ /h/ and /mV/ which would easily loose their connection with the underlying or historical sequences behind them. Thus speakers of the language would find it easy to generalise these sounds into previously illegal positions in new words treating them as independent phonemes and the allophonic stage of the sequences wouldn't probably last for too long. The explanation I've read for the origin of the retroflex nasal and lateral /ɳ ɭ/ in Khanty is that they started as allophones of /n l/ before the retroflex affricate but generalised to be used as independent phonemes in newly coined words. So even much more transparent allophones might be able to loose their connection to the governing conditions and the underlying phonemes. Tropylium should know a bit more about these.

Yet another way to make a minimal consonant inventory would be /p t k b d (g)/ where you contrast voiceless stops with voiced stops~continuants, though this seems a bit dull for me. However, there might still be something to it since I haven't seen any natlang or reconstruction using such a system together with a small vowel inventory.

[Nortaneous]

[sí] + -ò = [só] + -à = [swá]
[sín] + -ò = [smò] + -à = [smà]
[kádò] + -ò = [kádwò] + -à = [kádòwà]
[póbà] + <i> = [pjóbà] + -í = [pjóbè]
[tóbà] + <i> = [sóbà] + -n = [sóbàŋ] + -pá = [sóbàmbá]
[kóbà] + <i> = [hóbà] + -n = [hóbàŋ] + -kò = [hóbàŋò]
[sí] + -kò = [sígò] + -í = [sígòj]
etc.

set it on an isolated island so you don't get loanwords. and so what if it's unstable, if it'll only last a hundred years then make one of those hundred the time when you describe it

[Pole, the]

Well, know what's the best about that language?
It's a germanlang.

[Nortaneous]

where'd /S x/ go

[Pole, the]

where'd /S x/ go

/S/ just has never been created, and /x/ disappeared affecting the vowel quality, e.g. ialo "elk", alpà "to help".
(‹à› denotes /ə/.)

[Nortaneous]

Here is a phonology that contains dorsal fucking affricates.

/p t s~tʃ k ʔ/ <p t c k N/A>
/b d z~dʒ/ <b d z>
/pʼ tʼ tʃʼ kʼ/ <pʼ tʼ cʼ kʼ>
/f s̺ x h/ <f s x h>
/fʼ tsʼ kʟʼ kxʼ/ <fʼ sʼ klʼ xʼ>
/m n ŋ/ <m n ŋ>
/β r j ɣ gʟ/ <v r y g gl>
/a ɛ ɔ ɪ ʊ i u/ <a e o ë ö i u>

Phonotactics:

Diachronically, most words were sesquisyllabic: they had a minor syllable, consisting of a one-consonant initial and the vowel ə, and a major syllable of the form CV(C), although some words had an extra final vowel, giving CVCV. A monosyllabization process deleted the vowel of the initial and the extra vowel of disyllabic words, after creating two extra vowels through feature spread: CaCi gave CɛC, CaCu gave CɔC, and e o became ɪ ʊ before a final i or u. dZ > j word-finally, but before monosyllabization. The sequences /ej oj ew ow/ also became ɪ ɪ ʊ ʊ. Plosives in the preinitial lenited to fricatives, and /h/ dropped.

Glottal stops in the preinitial dropped and turned following consonants to ejectives, and plosives (or gʟ) in the preinitial turned to ejectives before a glottal stop -- so the protoforms *ʔəsu and *səʔu merged as sʼu. r > gl, which then > g intervocalically or after a syllable containing gl. Voiced stops here used to be prenasalized, and ngg turned to ng as in Japanese; protolang voiced stops fricated or turned to r.

Two example verb conjugations: (IEish but fuck you I'm lazy) Past tense was formed by reduplication, minus final vowel in CV roots but vowel was inserted before plural -w (~ -u) affix. Monophthongization of mid vowel + approximant series must have happened after vowel dropping.

Code: Select all

  sg    pl     sg       pl       |   sg      pl         sg           pl
1 xʼam  xʼom   xʼaŋxam  xʼaŋxom  | 1 ʔə-xam  ʔə-xam-u   ʔə-xam~xam   ʔə-xam~xam-u
2 ŋxam  ŋxom   ŋxaŋxam  ŋxaŋxom  | 2 nə-xam  nə-xam-u   nə-xam~xam   nə-xam~xam-u
3 xam   xom    xaŋxam   xaŋxom   | 3    xam     xam-u      xam~xam      xam~xam-u
                                 |                      
  sg    pl     sg       pl       |   sg      pl         sg           pl
1 klʼe  klʼö   klʼeg    klʼegö   | 1 ʔə-re   ʔə-re-w    ʔə-re~re     ʔə-re~re-w
2 ŋgle  ŋglö   ŋgleg    ŋglegö   | 2 nə-re   nə-re-w    nə-re~re     nə-re~re-w
3 gle   glö    gleg     glegö    | 3    re      re-w       re~re        re~re-w
                                 |                        
  sg    pl     sg       pl       |   sg      pl         sg           pl
1 tʼu   tʼu    tʼur     tʼur     | 1 ʔə-du   ʔə-du-w    ʔə-du~du     ʔə-du~du-w
2 ndu   ndu    ndur     ndur     | 2 nə-du   nə-du-w    nə-du~du     nə-du~du-w
3 ru    ru     rur      rur      | 3    du      du-w       du~du        du~du-w
                                 | 
  sg    pl     sg       pl       |   sg      pl         sg           pl 
1 kʼi   kʼiz   kʼizgi   kʼizgiz  | 1 ʔə-giz  ʔə-giz-u   ʔə-giz~giz   ʔə-giz~giz-u
2 ŋgi   ŋgiz   ŋgizgi   ŋgizgiz  | 2 nə-giz  nə-giz-u   nə-giz~giz   nə-giz~giz-u
3 gi    giz    gizgi    gizgiz   | 3    giz     giz-u      giz~giz      giz~giz-u

Proto forms are on the right. Presumably prenasalized g redeveloped as a plosive. Whatever. I'm not gonna develop this thing even though it's probably the best phonology I've made so far, and it only came from being annoyed at the noobs and their dorsal affricates, fuck you all

[Nortaneous]

ɭȷıɩɹuuɯnӏɯlȷuul
[ɭjiʂvɯtajk]

/m n/ <m n>
/ʔ/ <ӏ>
/l ɭ ɻ r j v/ <l ɭ ɹ r ȷ uu>
/i ɯ u z̩~s/ <ı ɯ u ɩ>

/mʔ nʔ jʔ wʔ/ [p t c k]
/ɪl ɯl ul/ [ɛ a ɔ]
/m̩ n̩ l̩ ɻ̩ r̩/ [õ ẽ y ø ʌ]
/sɻ nɻ ɻʔ/ [ʂ ɳ ʈ]

Written without spaces because spaces are for wimps.

rnıɩɹlnӏɩuuɯɩuuuɭȷılɩrȷӏıluuӏılӏrɩɭȷɹɩuuɭɯɩȷӏılɭɩı
[ʌnisølts vɯs vuɭjɛs ʌcɛkɛʔ ʌsɭjøs vɭɯscɛɭsi]
r-nıɩɹnӏɩ uu-ɯɩ uu-uɭȷılɩ r-ȷӏı-luuӏ-ılӏ r-ɩɭȷɹɩ uu-ɭɯɩȷӏılɭɩı
DEF-fox ANIM-quick ANIM-brown DEF-over-jump-PST DEF-dog ANIM-lazy

[2+3 clusivity]

That's pretty scary Nort, haha.

[Pole, the]

ɭȷıɩɹuuɯnӏɯlȷuul
[ɭjiʂvɯtajk]

/m n/ <m n>
/ʔ/ <ӏ>
/l ɭ ɻ r j v/ <l ɭ ɹ r ȷ uu>
/i ɯ u z̩~s/ <ı ɯ u ɩ>

/mʔ nʔ jʔ wʔ/ [p t c k]
/ɪl ɯl ul/ [ɛ a ɔ]
/m̩ n̩ l̩ ɻ̩ r̩/ [õ ẽ y ø ʌ]
/sɻ nɻ ɻʔ/ [ʂ ɳ ʈ]

Written without spaces because spaces are for wimps.

rnıɩɹlnӏɩuuɯɩuuuɭȷılɩrȷӏıluuӏılӏrɩɭȷɹɩuuɭɯɩȷӏılɭɩı
[ʌnisølts vɯs vuɭjɛs ʌcɛkɛʔ ʌsɭjøs vɭɯscɛɭsi]
r-nıɩɹnӏɩ uu-ɯɩ uu-uɭȷılɩ r-ȷӏı-luuӏ-ılӏ r-ɩɭȷɹɩ uu-ɭɯɩȷӏılɭɩı
DEF-fox ANIM-quick ANIM-brown DEF-over-jump-PST DEF-dog ANIM-lazy

I see you reinvented Armenian.

[Chagen]

Proto-Celestial:

/m n ŋ/
/p pʰ b t tʰ d k kʰ g q qʰ/
/s h/
/l~ɾ w j/

/a i u/
/a: i: u:/
/aj uj/

(C)(w,j)V(F/N)
F means "fricative", N means "nasal".

All consonants can be geminated between vowels except /w/.

This is a proto-lang, so almost no allophony besides /qi qu/ being pronounced [qe qo].

It's simple now, but the planned daughterlangs will have stuff like rounded front vowels from CwV, while others will turn Cj to Cʲ and then those'll lenite into palatal affricates...most of them will drop the uvulars or make them velar, giving /e o/ phonemic status (like the dropping of Laryngeals in PIE that colored vowels), though one dialect (the Jigoku one) will keep the uvulars. The Jigoku dialect has nasalized vowels too.

[cromulant]

This is a proto-lang, so almost no allophony

Reality doesn't work that way.

[Chagen]

I'm well aware of that. However, I feel that dealing with extensive allophony while also deriving daughterlangs would be a headache to keep a track of.

Also how many real-life reconstructed proto-langs have extensive allophony? Allophony in the proto-lang tends to be obscured by normal sound changes anyway.

Anyway this is part of a verb's conjugation run through some sound changes to derive the Shamayim daughterlang:

qaydas → ezas
qaydika → ejika
qayduqa → ezoka
qaydasay → ezase
qayduu → ezuu
qaydikay → ejike
qayduqay → ezoke

qaydu → ezu
qaydaru → ezaru
qaydawa → ezawa
qaydatim → ezatim
qaydiqu → ezeko
qaydarum → ezarum
qaydaway → ezawe

qaydimas → ejimas
qaydimika → ejimika
qaydimuqa → ejimoka
qaydimasay → ejimase
qaydimuu → ejimuu
qaydimkay → ejimke
qaydimuqay → ejimoke

qaydabas → ezavas
qaydabika → ezavika
qaydabuqa → ezavoka
qaydabasay → ezavase
qaydabuu → ezavuu
qaydabikay → ezavike
qaydabuqay → ezavoke

[Chagen]

Oh look it's a Heocg phonology:

/m n/

/p b bʰ t d dʰ k g gʰ q/
<p b bh t d dh k g gh>

/f~v s~z θ~ð x/
<f s ð h>

/ts dz tʃ dʒ/
<c z cg gc>

/w r l j/
<w r l j>

/a aː i iː ɛ e eː o oː u uː y yː/
<a ā i ī eo e ē o ō u ū y ȳ>

I'm trying to make the fucked up child of Old English and Sanskrit here.

Oddly enough, Heocg is spoken in my conworld's equivalent to Malaysia.

[Nortaneous]

Speaking of Old English, here's the unnamed Germanic language's inventory:

/pʰ b tʰ d tʃʰ dʒ kʰ g/ <p b t d tj dj k g>
/f v θ ð ɬ s z ʃ ʒ x h/ <f v th th ll s z si~sch si~zi ch h>
/m n ŋ/ <m n ng>
/l r j w/ <l r i~y w>
(/m̥ n̥ r̥/ <mh nh rh>)
/a aː ɛ eː (øː) ɔ oː i iː ɨ ɨː u uː/ <a aa e ee (Vr) o oo i y ie ye u w>
/ə/ <e>
/ai̯ au̯ ɛi̯ ɔu̯ iu̯/ <ai au ei ou yu>

Of course, there are some orthographical oddities: the first-person pronoun is always written y even though it's pronounced [itʃ] before a vowel, <tj dj> are written <ti di> before a vowel, <i u> in diphthongs alternate with <y w>, English and Welsh loanwords usually aren't respelled, many sound changes aren't represented...

[Hallow XIII]

Because of an ultimately probably futile effort to be chronologically correct in my conworld, I'm currently on hold from making conlangs with history etc. So I've been theme-langing.

Auran is spoken by wind-beings [...] It is a whistling language [...] like the wind

Dear Jesus what have I done.

p t t͡s t͡s͎ᶣ 〜 t͡ɕ͎ʷ t͡ɕ k kʷ c cᶣ ʔ
pʰ tʰ t͡sʰ t͡s͎ᶣʰ 〜 t͡ɕ͎ʷʰ t͡ɕʰ kʰ kʷʰ c͡çʰ cɕ͎ᶣʰ
s s͎ᶣ 〜 ɕ͎ʷ ɕ
sʰ s͎ᶣʰ 〜 ɕ͎ʷʰ ɕʰ
ʍ w ɥ h f ɧᶣ

EDIT: Höhö, I forgot the vowels.

/a ə i ɨ y o/, which merge in various patterns following some consonants. For instance, after the whistling/labiopalatal series, the only possible high vowels are y and ɥi - given the presence of a phonemic ɥ, this means that the high vowels are functionally non-contrastive in this environment. And so on.

[Chagen]

Oh look it's a Heocg phonology:

*stuff*

Let me modify this and add some allophony...

/p b bʰ t d dʰ k g gʰ q/
<p b bh t d dh k g gh q>

/f~v s~z θ~ð x/
<f s ð h>

/ts dz tʃ dʒ/
<c z cg gc>

/w r l j/
<w r l j>

/a aː i iː ɛ æ e eː o oː u uː y yː/
<a ā i ī eo ea e ē o ō u ū y ȳ>

I added just one vowel phoneme but here's some allophony:

/xr/ > [ʀ]
/xl/ > [ʟ̝̊]
/sts/ > [ʃ]
/sdz/ > [ʒ]
/f s θ/ > [v z ð] V_V
/w/ > [v] _#

Allophony does not apply across word boundaries so a hypothetical staf eosc /staf ɛtz/ would not be pronounced *[stav ɛʃ]