Romanization challenge thread

[Nortaneous]

/p t ts tʃ k q~ʁ/
/b d dz dʒ g/
/s ʃ x h/
/z ʒ/
/m~ŋ n l (ɫ) r j w/
/a e o i u/

l ɫ only contrast around morpheme boundaries and word-finally; l doesn't contrast with lj. h can only appear immediately before a vowel

/dzere kharelelis bguɫim dqumarzu da tuxortʃi. bguɫi umkharje dzerem dabqamdze da ndʒa. pentʃi taxɫe na temdze. bguɫi dari pentʃis nuxor; kiltis txor ʃa mlihi tim dʒema pʃeʒus tkaʃas./
[dzere kharelelis bguɫim dʁumarzu da tuxortʃi. bguɫi uŋkharje dzerem dabʁamdze da ndʒa. pentʃi taxɫe na temdze. bguɫi dari pentʃis nuxor; kiltis txor ʃa mlihi tim dʒema pʃeʒus tkaʃas.]

[Quantum]

Palato-velar - mid-way between palatal and velar.

Not 'pre-velar' /k̘ x̘ ʟ̘̊/?

[Bristel]

/m N n ŋ/ <mm nn n ng>
/mʲ Nʲ nʲ ŋʲ/ <mm nn n ng>
/p b t d k ɡ/ <p b t d c g>
/pʲ bʲ tʲ dʲ kʲ ɡʲ/ <p b t d c g>
/f v θ ð s x ɣ h/ <f v th dd s ch gh h>
/fʲ vʲ θʲ ðʲ sʲ xʲ ɣʲ hʲ/ <f v th dd s ch gh h>
/ṽ/ <m>
/ṽʲ/ <m>
/R r L l/ <rr r ll l>
/Rʲ rʲ Lʲ lʲ/ <rr r ll l>
/a aː e eː i iː o oː u uː/ <a ā e ē i ī o ō u ū>
/ai au eu oi iu ia ui ua/ <ai au eu oi iu ia ui ua>

palatalization: Cj except before i, C[-pal]i = <Cei>

/a moi kuː xulaNˡ i nuːn iṽriðʲ koguːale nʲiː a nʲeːṽ | ko nʲdʲiːuxrastar tʲrʲiana xotluð konʲiðʲ korastar asːa imðʲi ko rʲiaxt inʲ nariðʲinʲ ina sʲuiðʲu for laːr/

A moi Cū Chulannj i nūn imreiddj cogūale nī a niēm. Co ndīuchrastar triana chotludd coniddj corastar assa imddi co riacht inj nariddinj ina sjuiddju for lār.

Ooh, I do like this one.

[Dē Graut Bʉr]

Kitzanzik:

Code: Select all

m   n   ɳ       ŋ   ŋʷ
p   t   ʈ       k   kʷ   q   ʔ
pʰ  tʰ  ʈʰ      kʰ  kʷʰ  qʰ
p'  t'  ʈ'      k'  kʷ'  q'
    ts
    ts'
ɸ   s   ʂ       x   xʷ       h
ɸ'  s'  ʂ'      x'  xʷ'
    z           ɣ
            j       w


i ɨ ʉ u
   ə
æ     ɒ

[KathTheDragon]

Palato-velar - mid-way between palatal and velar.

Not 'pre-velar' /k̘ x̘ ʟ̘̊/?

No. Further forward than that. The speakers have mouths long enough to put a substantial space between their 'palate' and their 'velum'.

[Nortaneous]

Bashkimi alphabet:
Mbushni gne éne te madhé mé uje mbi zjarr dhe kur te filloje te vloje, hidhni patatet e posavjelura. Lini te vlojne vétem gne minut. Pastaj, terhicni ato, fshini osé lini te thahén. Ne kete menyre, sythat shkaterrohén dhé tanime férmentimi tyré eshte i pamunchem. Patatet e trajtuara ne kete menyre véndosén neper kuti druri osé thase dhé mund te konservohen afersisht dy vité ne gne vénd te éret é te frésket.

Istanbul alphabet:
Mbuσni ŋe εne te maδε mε uje mbi zjaρ δe kur te fiλoje te vloje, hiδni patatet e posavjelura. Lini te vlojne vεtem ŋe minut. Πastaj, terhiqni ato, fσini osε lini te θahεn. Ne kete menyre, syθat σkateρohεn δε tanime fεrmentimi tyrε eσte i pamunçem. Πatatet e trajtuara ne kete menyre vεndosεn neper kuti druri osε θase δε mund te konservohen afersiσt dy vitε ne ŋe vεnd te ερet ε te frεsket.

cf.
Mbushni një enë të madhe me ujë mbi zjarr dhe kur të fillojë të vlojë, hidhni patatet e posavjelura. Lini të vlojnë vetëm një minut. Pastaj, tërhiqni ato, fshini ose lini të thahen. Në këtë mënyrë, sythat shkatërrohen dhe tanimë fermentimi tyre është i pamundshëm. Patatet e trajtuara në këtë mënyrë vendosen nëpër kuti druri ose thasë dhe mund të konservohen afërsisht dy vite në një vend të errët e të freskët.

added bonus: combination of archaic features
Mbuscni gnö enö tö maξe me ujö mbi zjarh ξe cur tö fiłojö tö vlojö, hiξni patatet e posavjelura. Lini tö vlojnö vetöm gnö minut. Pastaj, törhichni ato, fscini ose lini tö ξξahen. Nö cötö mönürö, süξξat sccatörohen ξe tanimö fermentimi türe ösctö i pamuntzöm. Patatet e trajtuara nö cötö mönürö vendosen nöpör cuti druri ose ξξasö ξe mund tö conservohen aförsisct dü vite nö gnö vend tö eröt e tö fresköt.

[WeepingElf]

/p t ts tʃ k q~ʁ/ p t c c' k q
/b d dz dʒ g/ b d j j' g
/s ʃ x h/ s s' x h
/z ʒ/ z z'
/m~ŋ n l (ɫ) r j w/ m n l l' r y w
/a e o i u/ a e o i u

l ɫ only contrast around morpheme boundaries and word-finally; l doesn't contrast with lj. h can only appear immediately before a vowel

/dzere kharelelis bguɫim dqumarzu da tuxortʃi. bguɫi umkharje dzerem dabqamdze da ndʒa. pentʃi taxɫe na temdze. bguɫi dari pentʃis nuxor; kiltis txor ʃa mlihi tim dʒema pʃeʒus tkaʃas./

Jere kharelelis bgul'im dqumarzu da tuxorc'i. Bgul'i umkharje jerem dabqamje dda nj'a. Penc'i taxl'e na temje. Bgul'i dari penc'is nuxor; kiltis txor s'a mlihi tim j'ema ps'ez'us tkas'as.

[Jess]

/p t ts tʃ k q~ʁ/
/b d dz dʒ g/
/s ʃ x h/
/z ʒ/
/m~ŋ n l (ɫ) r j w/
/a e o i u/

p t ts tś c q
b d dz dź g
s ś j h
z ź
m n l ll r y w
a e o i u

/dzere kharelelis bguɫim dqumarzu da tuxortʃi. bguɫi umkharje dzerem dabqamdze da ndʒa. pentʃi taxɫe na temdze. bguɫi dari pentʃis nuxor; kiltis txor ʃa mlihi tim dʒema pʃeʒus tkaʃas./
dzere charelelis bgullim dqumarzu da tujortśi. bgulli umcharye dzerem dabqamdze da ndźa. pentśi tajlle na temdze. bgulli dari penśis nujor; ciltis tjor śa mlihi tim dźema pśeźus tcaśas.

I might tweak it if I had more phonotactic information. For example, if /bk/ is an illegal word initial cluster, I might write /bguɫi/ as bculli because I think it looks nicer than bgulli. I also don't like the look of <dq, bq> and think it is interesting that the only two /q/ in the sample occur in clusters... maybe dŕumarzu and dabŕamdze would be better.

ETA:

Kitzanzik:

Code: Select all

m	n	rn   	ŋ	ŋv
p	t	rt   	k	kv  q	ʔ
ph  th  rth  	kh  kvh qh	
p'  t'  rt'  	k'  kv' q'
 	c
 	c'
f	s	rs   	x	xv   	h
f'  s'  rs' 	 x'  xv'
 	z		 	 g	
 		 	 j		 w

i ï ü u
e
a o

Would have been nice to have some phonotactics and/or a sample text; for instance, the labiovelars could be <ŋw kw kwh> etc as long as CwV and CʷV don't contrast.

I thought about making a diacritic-y version in addition to this digraphy version, but who likes combining unicode diacritics anyway...

[Nortaneous]

Clusters always agree in voicing, with /q/ unmarked for voice.

Ѕере к'арелелис бгулым дъумарзу да тухорчи. Бгулы умк'арье ѕерем дабъамѕе да нџа. Пенчи тахле на темѕе. Бгулы дари пенчис нухор, килтис тхор ша мли'и тим џема пшежус ткашас.

or, using ejectives for plain voiced stops to fit in the aspirates: [though come to think of it it'd be better to just use Ch digraphs)

ძერე ქარელელის ბგულიმ დჴუმარზუ და ტუხორჩი. ბგული უმქარიე ძერემ დაბჴამძე და ნჯა. ფენჭი ტახლე ნა ტემძე. ბგული დარი პენჭის ნუჰორ; კილტის ტხორ შა მლიჰი ტიმ ჯემა პშეჟუს ტკაშას.

[Linguist Wannabe]

I'm stuck on how to romanise Kämpya.

Consonants

Stops /pʰ p b tʰ t d kʰ k g/

Nasals /m̥ m n̥ n ŋ /

Fricatives /θ ð sʰ s z h/

Semivowels /w j/

Other Sonorants /ⱱ ɾ l/


Vowels

Monophthongs /a e i o u/

Diphthongs /ai au ei ou/


Phonotactics

The range of permitted syllable shapes is (C) (C) V (C).

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

/pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ are also permitted as initial consonant clusters.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.


Stress and Phonation

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).


Tone Patterns

For this, please read http://linguifex.com/K%C3%A4mpya#Tone_Patterns. I'm sorry, but I don't know how to put a table in a post on this forum.


Consonant Neutralisations

Voicing and aspiration distinctions are neutralised in syllable codas, though they are still present underlyingly e.g. /sjúˈðâ̰d/ - "city" is pronounced [sjúˈðâ̰t], but when combined with the genitive clitic /jà/ to form /sjúˈðâ̰djà/ - "of the city (alienable)", it is pronounced [sjúˈðâ̰djà].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is simply pronounced as nasalisation of the preceding vowel e.g. /áˈlòṳn/ - "everyone" is pronounced [áˈlòṳⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the dative clitic /àuŋ/ to form /áˈlòṳnàuŋ/ - "to everyone", which is pronounced [áˈlòṳnàuŋ].


Allophones of /h/

[Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/
/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel, but is always pharyngeal.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a pharyngeal fricative [ħ] e.g. /à̤h/ - "question" is pronounced [à̤ħ]. Compare /ŋà̤/ - "fish" which is pronounced [ŋà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as a pharyngeal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ]. Compare /táʔ/ - "a wish" (in a nostalgic context) which is pronounced with a glottal stop [táʔ].

/h/ never occurs after vowels with harsh phonation.

Of course, if a process such a cliticisation (e.g. with the genitive clitic /jà/ or the dative clitic /àuŋ/), means that the /h/ is no longer in coda position, then these process do not occur e.g.

/à̤h jà/ - "of the question (alienable)" is pronounced [ˈà̤çà] (/hj/ assimilates to [ç]).

/dáʔh jà/ - "of the darkness (alienable)" is pronounced [ˈdáʔçà].

/à̤h àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.
/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

I'm looking forward to seeing what people come up with

[Nortaneous]

Stops /pʰ p b tʰ t d kʰ k g/ <ph p b th t d kh k g>
Nasals /m̥ m n̥ n ŋ / <mh m nh n ng>
Fricatives /θ ð sʰ s z h/ <c dh sh s z h~x> (/h/ is <h> initially and <x> in the coda)
Semivowels /w j/ <w y>
Other Sonorants /ⱱ ɾ l/ <v r l>
Monophthongs /a e i o u/ <a e i o u>
Diphthongs /ai au ei ou/ <ai au ei ou>
Phonations are harsh /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. <aa ah a>
Tones:
H 1 2 3 <a à á>
B 1+2 3 <a á>
G 1+3 2 <a à>

gáʔɾzù káiʔkdè pʰáˈzè̤t kʰwèi̤jàuŋ nóuʔp. θúkáiʔk gáʔɾ kʰwèi̤jàuŋ pjèi̤. kʰwèi̤ θwáhlôṵn.
Garzu kaikde phazeht khweih'yaung noup. Cukaik gar khweih'yaung pyeih. Khwèih cwahlouun.

er, wait, are all three tone classes contrastive for each phonation?

[Click]

Stops /pʰ p b tʰ t d kʰ k g/ f p b c t d ch k g
Nasals /m̥ m n̥ n ŋ/ hv m hr n ng
Fricatives /θ ð sʰ s z h/ tg y s x z h
Semivowels /w j/ v y
Other Sonorants /ⱱ ɾ l/ vr r l
Monophthongs /a e i o u/ a e ı o u
Diphthongs /ai au ei ou/ aı au eı ou
Phonations are harsh /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. a ao aʼ
Tones:
H 1 2 3 a á à
B 1+2 3 á a
G 1+3 2 a à

Garʼzù kaıkʼde fázeot chveıyàung noupʼ. Tgúkaık garʼ chvrıyàung pyeı. Hveı tgváhlòun.
gáʔɾzù káiʔkdè pʰáˈzè̤t kʰwèi̤jàuŋ nóuʔp. θúkáiʔk gáʔɾ kʰwèi̤jàuŋ pjèi̤. kʰwèi̤ θwáhlôṵn.

[Dē Graut Bʉr]

Would have been nice to have some phonotactics and/or a sample text; for instance, the labiovelars could be <ŋw kw kwh> etc as long as CwV and CʷV don't contrast.

The syllable structure of Kitzanzik is CV(C), and /kw/ and /kʷ/ don't contrast. The language has not really been fleshed out yet, so I can't give a real sample yet, but this Kitzanzik-ish gibberish should work as well.

/xʷætsuʂ q'ɨtʰəj hɒʈu ɣukʰə. makʷ'iŋ s'uzɣiʈ ɸɨtʰɨ. ɳæxʷə ŋutsuxʷɨ həʔə x'ɒp'u k'ɨzæ mintsæw ŋʷɨqʰuzu./

[Linguist Wannabe]

Thanks for such a quick response

Stops /pʰ p b tʰ t d kʰ k g/ <ph p b th t d kh k g>
Nasals /m̥ m n̥ n ŋ / <mh m nh n ng>
Fricatives /θ ð sʰ s z h/ <c dh sh s z h~x> (/h/ is <h> initially and <x> in the coda)
Semivowels /w j/ <w y>
Other Sonorants /ⱱ ɾ l/ <v r l>
Monophthongs /a e i o u/ <a e i o u>
Diphthongs /ai au ei ou/ <ai au ei ou>
Phonations are harsh /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. <aa ah a>
Tones:
H 1 2 3 <a à á>
B 1+2 3 <a á>
G 1+3 2 <a à>

gáʔɾzù káiʔkdè pʰáˈzè̤t kʰwèi̤jàuŋ nóuʔp. θúkáiʔk gáʔɾ kʰwèi̤jàuŋ pjèi̤. kʰwèi̤ θwáhlôṵn.
Garzu kaikde phazeht khweih'yaung noup. Cukaik gar khweih'yaung pyeih. Khwèih cwahlouun.

Good idea to mark the phonations with <aa>, <ah> etc. I think you're on to something here. But the way it is so far, how do you contrast regular unstressed syllables and (stress bearing) syllables with glottalised phonation and Tone Classes 1 or 3? Both will romanise as plain <a> e.g. both /θúˈkáiʔk/ and /ˈθúʔkàik/ (if such a word ever existed) will romanise as <cukaik>.

er, wait, are all three tone classes contrastive for each phonation?

For words with breathy phonation, on the stressed syllable, both Tone Class 1 and 2 are identical (Low Breathy tone). But they differ in that, in Tone Class 1, the unstressed syllables after the stressed syllable will have Low Tone. While in Tone Class 2, they will have High Tone e.g.

Tone Class 1 /ˈsì̤tàjà/ - sister's (alienable)
vs.
Tone Class 2 /ˈsì̤tájá/ - sister's (inalienable)

However, if the stress is on the last syllable of the word, such a distinction will be obscured and the words will be the same e.g.

Tone Class 1 /áˈlòṳn/ - all / everyone / everything
and
Tone Class 2 /áˈlòṳn/ - all / every

It will only resurface again when we put a clitic after the word e.g.

Tone Class 1 /áˈlòṳnjà/ - everyone's (alienable)
vs.
Tone Class 2 /áˈlòṳnjá/ - everyone's (inalienable)


Likewise, for words with glottalised phonation, on the stressed syllable, both Tone Class 1 and 3 are identical (High Tone on the stressed syllable). But the difference here is, in Tone Class 1, the unstressed syllables before the stressed syllable will have High Tone. While in Tone Class 3, they will have Low Tone. e.g.

Tone Class 1 /tjúntʰéʔt/ - a sharp object / to sharpen
vs.
Tone Class 3 /tjùntʰéʔt/ - sharp (non restrictive) / sharply

However, if the stress is on the first syllable of the word, then this contrast will disappear in the same way.

So, if we are not marking tone on unstressed syllables, then (at least in theory) we need to mark each tone class contrastively, even if they are pronounced the same on the stressed syllable.

I hope this makes sense. Should I go back and reword the wiki page?

[Nortaneous]

I wrote this before you replied:

--------
/pʰ p b tʰ t d kʰ k g/ <ph p b th t d kh k g>
/m̥ m n̥ n ŋ/ <mh m nh n ng>
/θ ð sʰ s z h/ <c dh sh s z h>
/w j/ <w y>
/ⱱ ɾ l/ <v r l>
/a e i o u/ <a e i o u>
/ai au ei ou/ <ai au ei ou>
/a̰ a̤ aʔ/
a̰ + falling, low, high = <â ā ǎ>
a̤ + low, rising = <ã ä>
aʔ + high, low = <á ả>
a + high, low = <a à>

gáʔɾzù káiʔkdè pʰáˈzè̤t kʰwèi̤jàuŋ nóuʔp. θúkáiʔk gáʔɾ kʰwèi̤jàuŋ pjèi̤. kʰwèi̤ θwáhlôṵn.
Gárzù káikdè phazẽt khwẽiyàung nóup. Cukáik gár khwẽiyàung pyëi. Khwẽi cwahlôun.
Or, with pseudo-Burmese consonants:
Gárzù káikdè hpazẽt hkwẽiyàung nóup. Thukáik gár hkwẽiyàung pyëi. Hkwẽi thwahlôun.

It'd be easier to just mark phonation, stress, and tone class, but I don't know if there's a good way to do that. Let's see...
/a̰ a̤ aʔ/ <â a/à á> (breathy voice not marked in monosyllables; change this to whichever is most common obv)
/1 2 3/ <1 2 3> -- assume class 1 if can't be class 2 or 3, assume class 2 if can't be class 3

gáʔɾzù káiʔkdè pʰáˈzè̤t kʰwèi̤jàuŋ nóuʔp. θúkáiʔk gáʔɾ kʰwèi̤jàuŋ pjèi̤. kʰwèi̤ θwáhlôṵn.
Gárzu1 káikde1 phazèt1 khwèiyaung1 nóup1. Cukáik1 gár1 khwèiyaung1 pyei1. Khwei1 cwahlôun1.

er... let's add another example sentence?

ˈkʰòʔnátm̥á kúʔn ˈḛ̂nèmì
Khónatma2 kún1 ênemi1.

and

àˈlǒṳnjà
alòunya3

There aren't many combining diacritics that play well with each other, so let's switch out the circumflex for the tilde, and then assign null, circumflex, and underdot to tone classes 1, 2, and 3 respectively.

Gárzu káikde phazèt khwèiyaung nóup. Cukáik gár khwèiyaung pyei. Khwei cwahlõun.
Khốnatma kún ênemi.
alọ̀unya
--------

But the way it is so far, how do you contrast regular unstressed syllables and (stress bearing) syllables with glottalised phonation and Tone Classes 1 or 3? Both will romanise as plain <a> e.g. both /θúˈkáiʔk/ and /ˈθúʔkàik/ (if such a word ever existed) will romanise as <cukaik>.

I was hoping to avoid having diacritics on every word, but I guess that won't work. Give it a circumflex or something: cukâik vs. cûkaik. Then whichever diacritic occurs most frequently in monosyllables can be dropped there unless there are unstressed, phonationless words.

The tone class thing makes sense. It looks like there's a tradeoff that will have to be made in the romanization: you can be redundant and mark every tone but have trivial mapping from sounds to characters, you can avoid the redundancy and only mark tone class at the cost of making it impossible to go from sounds to characters without knowing the grammar of the language, or you can mark tone class improperly like I did up there, where words that aren't class 1 but sound like they could be get written as class 1.

And if you do decide to use tone-letters -- especially if you use <h> -- you'll need apostrophes or hyphens (as I did in that first post) to disambiguate between sequences like <hy> (initial /hj/ cluster, pronounced [ç]) and <h'y> (breathy vowel followed by /j/).

[Linguist Wannabe]

Wow Nortaneous, what a thoughtful post. You're totally right about the tradeoff we need to make in terms of overly redundant marking vs. a scheme that requires knowledge of the grammar vs. a scheme that doesn't indicate everything.

But the way it is so far, how do you contrast regular unstressed syllables and (stress bearing) syllables with glottalised phonation and Tone Classes 1 or 3? Both will romanise as plain <a> e.g. both /θúˈkáiʔk/ and /ˈθúʔkàik/ (if such a word ever existed) will romanise as <cukaik>.

I was hoping to avoid having diacritics on every word, but I guess that won't work. Give it a circumflex or something: cukâik vs. cûkaik. Then whichever diacritic occurs most frequently in monosyllables can be dropped there unless there are unstressed, phonationless words.

With your new romanisation, won't this not be an issue anymore anyway? In your version with the eight diacritics, /θúˈkáiʔk/ romanises as <cukáik>, while /ˈθúʔkàik/ romanises as <cúkàik>. And I don't think you have a problem in your next version either.

And BTW the only unstressed, phonationless "words" that are in the language are clitics that attach phonologically to surrounding words (e.g. the antipassive marker /θú/) that you have romanised as part of the following word anyway.

And if you do decide to use tone-letters -- especially if you use <h> -- you'll need apostrophes or hyphens (as I did in that first post) to disambiguate between sequences like <hy> (initial /hj/ cluster, pronounced [ç]) and <h'y> (breathy vowel followed by /j/).

I think we're going to need to do this anyway, since <ng> could be read as either /ŋ/ or a cluster.

[Jess]

Kämpya:

ph p b th t d kh k g
mh m nh n ŋ
t' d' sh s z x
w y
v r l
a e i o u
ai au ei ou

falling harsh(1), low harsh(2), high tense(3): ą̀ ą ą́
low breathy(1 and 2), rising breathy(3): ah áh
high glottalized(1 and 3), low glottalized(2): á a

If you need to differentiate tone on an unstressed syllable, grave is low and high is unmarked. That way, acute accents only occur on stressed syllables and (with the unambiguous exception of ą̀) graves only occur on unstressed syllables.

gáʔɾzù káiʔkdè pʰáˈzè̤t kʰwèi̤jàuŋ nóuʔp. θúkáiʔk gáʔɾ kʰwèi̤jàuŋ pjèi̤. kʰwèi̤ θwáhlôṵn.
gárzu káikde phazeht khweihyàuŋ nóup. t'ukáik gár khweihyàuŋ pyeih. khweih t'waxlǫ̀un.

ˈkʰòʔnátm̥á kúʔn ˈḛ̂nèmì.
khonatmha kún ę̀nemi.

/θúˈkáiʔk/ vs. /ˈθúʔkàik/: t'ukáik, t'úkaik
/ˈsì̤tàjà/ vs. /ˈsì̤tájá/: sihtàya, sihtaya
/áˈlòṳnjà/ vs. /áˈlòṳnjá/: alouhnyà, alouhnya
/tjúntʰéʔt/ vs. /tjùntʰéʔt/: tyunthét, tyùnthét

In the theoretical case that it is unclear which syllable is stressed, you could stick a hyphen before the stressed one:

/'sèʔkó/ vs. /sé'kòʔ/: seko, se-ko

[8Deer]

Sukkn, inspired a bit by Salishan

Consonants
/p t~tʃ k (ʔ)/
/s~ʃ h/
/m~əm n~ən ŋ~əŋ/
/w~u: j~i: ɰ~ə: ɦ~a:/
/l~əl/
-Alveolar obstruents become palatal initially before a vowel.
- ʔ is not phonemic, but occurs before all word-initial vowels and probably in hiatus, I dunno yet.
- /h ɦ/ are post-velar and can be uvular, glottal or pharyngeal.
- All sonorants can be syllabic. When consonantal, they are realised with slight frication.

Vowels (yes, I jacked this completely from Secwepemc)
/i~e ɛ ə a ɔ u~o/
- High vowels are lowered in closed syllables.
- /i u ə a/ have long counterparts, however these are allophones of the syllabic sonorants /j w ɰ ɦ/.

A sonorant is syllabic unless it is followed by a syllabic segment or preceded by a vowel.

Some example words:
/sukkn/ [ʃokkən]
/skw/ [sku:]
/ɦn/ [ɦən]
/ɦnu/ [a:nu]
/uljŋsɰ/ [ʔuljəŋsə:]

[kanejam]

Serer

/m n ɲ ŋ/ <m n ny ng>
/p t c k q ʔ/ <p t c k q '>
/b d ɟ ɡ/ <b d j g>
/ᵐb ⁿd ᶮɟ ᵑɡ ᶰɢ/ <mb nd nj ng nq>
/ƥ ƭ ƈ/ <pp tt kk>
/ɓ ɗ ʄ/ <bb dd gg>
/f s x h/ <f s hh h>
/ɾ/ <r>
/l j w/ <l y v>
/a e i o u/ <a e i o u>
All vowels can also be long. <VV>

Sample:
/no ɲak o aⁿd fo o jeːsaⁿdaː ke warna in | naː ɓisiːdaː fitna | fa jiːf a peƭaru no ɲoːw in toː adna fajnwiːn we a lajaː o ᵑɡalaːt dan | taːmaːla mateː den fo ɲak keːn refu ke wiːn we a moʄna o ᵐbuɡ no ɲoːw | no ke moʄna | ten ref a saːχ ake a jetoːχa ke ɟofna no kiːn ⁿdax ɟam a ɟeɡ boː muː refna jeːɡ o fa ɲo neːwaⁿdel/
No nyak o and fo o yeesandaa ke varna in. Naa bbisiidaa fitna. Fa yiif a pefaru no nyoov in too adna faynviin ve a layaa o ngalaat dan. Taamaala matee den fo nyak keen refu ke viin ve a moggna o mbug no nyoov. No ke moggna. Ten ref a saahh ake a yetoohha ke jofna no kiin ndahh jam a jeg boo muu refna yeeg o fa nyo neevandel.


Ode to the dot diacritic: Edit: this is Kitzanzik btw

/m n ɳ ŋ ŋʷ/ <m n ṇ ṅ ṅv>
/p t ts ʈ k kʷ q ʔ/ <b d ċ ḍ g gv ġ q>
/pʰ tʰ ʈʰ kʰ kʷʰ qʰ/ <p t c ṭ k kv ḳ>
/p' t' ts' ʈ' k' kʷ' q'/ <p' t' c' ṭ' k' k'v ḳ'>
/ɸ s ʂ x xʷ h/ <ṗ s ṣ ḥ ḥv h>
/ɸ' s' ʂ' x' xʷ'/ <ṗ' s' ṣ' ḥ' ḥ'v>
/z ɣ/ <z ẓ>
/j w/ <j v>

/i ɨ ʉ u/ <i ı u̇ u>
/æ ə ɒ/ <e a o>

/xʷætsuʂ q'ɨtʰəj hɒʈu ɣukʰə. makʷ'iŋ s'uzɣiʈ ɸɨtʰɨ. ɳæxʷə ŋutsuxʷɨ həʔə x'ɒp'u k'ɨzæ mintsæw ŋʷɨqʰuzu./
Ḥveċus ḳ'ıtaj hoḍu ẓuka. Mak'viṅ s'uzẓiḍ ṗıtı. Ṇeḥva ṇuċuḥvï haqa ḥ'op'u k'ıze minċev ṇvıkvuzu.

[KathTheDragon]

Sukkn, inspired a bit by Salishan

Consonants
/p t~tʃ k (ʔ)/ <p t c '>
/s~ʃ h/ <s h>
/m~əm n~ən ŋ~əŋ/ <(y)m (y)n (y)ñ>
/w~u: j~i: ɰ~ə: ɦ~a:/ <ú í ý á>
/l~əl/ <(y)l>

Vowels
/i~e ɛ ə a ɔ u~o/ <i e y a o u>

Some example words:
/sukkn/ [ʃokkən] <succyn>
/skw/ [sku:] <scú>
/ɦn/ [ɦən] <áyn>
/ɦnu/ [a:nu] <ánu>
/uljŋsɰ/ [ʔuljəŋsə:] <'ulíyñsý>

[Pole, the]

Sukkn, inspired a bit by Salishan

Consonants
/p t~tʃ k (ʔ)/ <p t k x>
/s~ʃ h/ <s h>
/m~əm n~ən ŋ~əŋ/ <m/am n/an g/ag>
/w~u: j~i: ɰ~ə: ɦ~a:/ <u i w/aw r/ar>
/l~əl/ <l/al>

Vowels
/i~e ɛ ə a ɔ u~o/ <e è a à ò o>

Some example words:
/sukkn/ [ʃokkən] <sokkan>
/skw/ [sku:] <sku>
/ɦn/ [ɦən] <ran>
/ɦnu/ [a:nu] <arno>
/uljŋsɰ/ [ʔuljəŋsə:] <xoliagsaw>

[kanejam]

Sukkn, inspired a bit by Salishan

Consonants
/p t~tʃ k (ʔ)/ <b t/ch k >
/s~ʃ h/ <s/sh kh>
/m~əm n~ən ŋ~əŋ/ <m n g>
/w~u: j~i: ɰ~ə: ɦ~a:/ <w j y h>
/l~əl/ <l>

Vowels
/i~e ɛ ə a ɔ u~o/ <i/e è ë a ò u/o>

Some example words:
/sukkn/ [ʃokkən] <shokkn>
/skw/ [sku:] <skw>
/ɦn/ [ɦən] <hn>
/ɦnu/ [a:nu] <hnu>
/uljŋsɰ/ [ʔuljəŋsə:] <uljgsy>

[Pogostick Man]

Because its Pinyin is, to put it bluntly, absolutely horrible, let's see y'all romanize Yi (presumably Nuosu, via Wikipedia):

/m̥ m n̥ n nʲ ŋ/
/m͜b n͜d ŋ͡g/
/p pʰ b t tʰ d k kʰ g/
/t͜s t͜sʰ d͜z nd͜z t͜ɕ t͜ɕʰ d͜ʑ nd͜ʑ t͡ʂ t͡ʂʰ d͡ʐ nd͡ʐ/
/f v s z ɕ ʑ ʂ ʐ x ɣ h/
/ɬ/
/l/

/a ɛ i ɔ o ɯ u u̠ ɿ ɿ̱̠/
/˨˩ ˧ ˧˦ ˥/

Syllable structure appears to be (C)V.

[Rhetorica]

Because its Pinyin is, to put it bluntly, absolutely horrible, let's see y'all romanize Yi (presumably Nuosu, via Wikipedia)

/m̥ m n̥ n nʲ ŋ/ mh m n nh nj ng
/m͜b n͜d ŋ͡g/ mb nd ngg
/p pʰ b t tʰ d k kʰ g/ p ph b t th d k kh g
/t͜s t͜sʰ d͜z nd͜z t͜ɕ t͜ɕʰ d͜ʑ nd͜ʑ t͡ʂ t͡ʂʰ d͡ʐ nd͡ʐ/ ts tsh dz ndz tcy tcjh dzj ndzj tc tch dzh ndzh
/f v s z ɕ ʑ ʂ ʐ x ɣ h/ f v s z cj zj c z x g h
/ɬ/ lh
/l/ l

/a ɛ i ɔ o ɯ u u̠ ɿ ɿ̱̠/ a e i oa o w u ur y yr
/˨˩ ˧ ˧˦ ˥/ à a á â

apparently /ɿ/ =/ ͡ɯ/. No idea what u̠ or ɿ̱̠ are, so I just copied the Pinyin values for all three.

It looks like most of the terrible Pinyin decisions are in how to deal with the really messy consonant clusters. Not too surprised... but how do you decide /d͜ʑ/= <rr>?!

[Nortaneous]

/p t~tʃ k (ʔ)/ <p t k 0>
/s~ʃ h/ <s h>
/m~əm n~ən ŋ~əŋ/ <m~em n~en ŋ~eŋ>
/w~u: j~i: ɰ~ə: ɦ~a:/ <w~uw y~iy g~eg c~ac>
/l~əl/ <l~el>

Vowels
/i~e ɛ ə a ɔ u~o/ <i é e a o u>

Some example words:
/sukkn/ [ʃokkən] <sukken>
/skw/ [sku:] <skuw>
/ɦn/ [ɦən] <cen>
/ɦnu/ [a:nu] <acnu>
/uljŋsɰ/ [ʔuljəŋsə:] <ulyeŋseg>

----

/m̥ m n̥ n nʲ ŋ/ <hm m hn n ny ng>
/m͜b n͜d ŋ͡g/ <mb nd nk>
/p pʰ b t tʰ d k kʰ g/ <p ph b t th d k kh g>
/t͜s t͜sʰ d͜z nd͜z t͜ɕ t͜ɕʰ d͜ʑ nd͜ʑ t͡ʂ t͡ʂʰ d͡ʐ nd͡ʐ/ <c ch dz nz q qh j nj tr rh r nr>
/f v s z ɕ ʑ ʂ ʐ x ɣ h/ <f v s z x y sh zh xh w h>
/ɬ/ <hl>
/l/ <l>

/a ɛ i ɔ o ɯ u u̠ ɿ ɿ̱̠/ <a ei i oa o e u ur y yr>
/˨˩ ˧ ˧˦ ˥/ <v 0 h t>

Mbo ma mu veih juh jo, nzy du i qih jy yyh mu jo sad. Mbo woz ngov nke si niv bov hei nyi jo, dih av bov hmav zyd hniv mov mu jo tad xi.