Romanization challenge thread

[Haplogy]

/m n ŋ mˤ nˤ/ <m n ng mħ nħ>
/t d tˤ dˤ tʃ dʒ k g/ <t d tħ dħ ċ ġ c g>
/s z sˤ zˤ ʃ ʒ ʁ χ h/ <s z sħ zħ x j għ cħ h>
/r rˤ l w j/ <r rħ l w y>

/a e i o u ə/ <a e i o u ë>

I don't even know anymore...

[Zontas]

I just came up with this inventory, come to think about it- maybe I should call Gleb to get a job offer:

(Working Title: Rhasman- not that the name suggests a romanization)

/m mʲ n̪ n̪ʲ ŋ ŋʲ/
/p pʲ b bʲ t̪ t̪ʲ d̪ d̪ʲ k kʲ g gʲ/
/f fʲ v s sʲ z χ χʲ ʁ̝/
/βʲ ʋ̊ ɹ̪ʲ ɹ̊ ɣʲ χ̞/
/b͡βʲ d̪͡ðʲ g͡ɣʲ/
/j w/
/l/

/ɨ ʉ/
/e̞ (œ̞) ə (ɤ̞) o̞/
/ä ɒ̈/

()= phonemes I might drop from the finished project, therefore feel feel free to not romanize them

[Pole, the]

Polish

/m n ɲ/
/p b t d kʲ gʲ k g/
/t͡s d͡z t͡ʂ d͡ʐ t͡ɕ d͡ʑ/
/f v s z ʂ ʐ ɕ ʑ xʲ x/
/r l j w/
/i ɨ u ɛ ɔ a ɛ̃ ɔ̃/

I don't think /kʲ ɡʲ xʲ/ are phonemic (at least not if we don't include /mʲ pʲ bʲ vʲ/).
Similarly /ɛ̃ ɔ̃/ can be analysed as /ɛ(N) ɔ(N)/ in most cases.

[Ambrisio]

/m mʲ n̪ n̪ʲ ŋ ŋʲ/ <m m' n n' ŋ ŋ'>
/p pʲ b bʲ t̪ t̪ʲ d̪ d̪ʲ k kʲ g gʲ/ <p p' b b' t t' d d' k k' g g'>
/f fʲ v s sʲ z χ χʲ ʁ̝/ <f f' v s s' z h h' ğ>
/βʲ ʋ̊ ɹ̪ʲ ɹ̊ ɣʲ χ̞/ <v' hv r' hr ğ' x>
/b͡βʲ d̪͡ðʲ g͡ɣʲ/ <bh' dh' gh'>
/j w/ <j w>
/l/ <l>

/ɨ ʉ/ <i u>
/e̞ (œ̞) ə (ɤ̞) o̞/ <e ö y õ o>
/ä ɒ̈/ <ä a>

I am surprised that you are dropping Ö and Õ but not X (in the orthography). I think X and H are much less distinct than Ö and Õ, so realistically I would expect the latter set to survive rather than the former.

[Nortaneous]

/m mʲ n̪ n̪ʲ ŋ ŋʲ/ <m my n ny ng ngy>
/p pʲ b bʲ t̪ t̪ʲ d̪ d̪ʲ k kʲ g gʲ/ <p py b by t ty d dy k ky g gy>
/f fʲ v s sʲ z χ χʲ ʁ̝/ <f fy v s sy z x xy ğ>
/βʲ ʋ̊ ɹ̪ʲ ɹ̊ ɣʲ χ̞/ <vy fh ry r hy h>
/b͡βʲ d̪͡ðʲ g͡ɣʲ/ <bvy dhy gğy>
/j w/ <y w>
/l/ <l>

/ɨ ʉ/ <i u>
/e̞ (œ̞) ə (ɤ̞) o̞/ <e ö ë ï o>
/ä ɒ̈/ <ä a>

Although really this should be Cyrillic.

/m mʲ n̪ n̪ʲ ŋ ŋʲ/ <м мь н нь ӊ ӊь>
/p pʲ b bʲ t̪ t̪ʲ d̪ d̪ʲ k kʲ g gʲ/ <п пь б бь т ть д дь к кь г гь>
/f fʲ v s sʲ z χ χʲ ʁ̝/ <ф фь в с сь з х хь ғ>
/βʲ ʋ̊ ɹ̪ʲ ɹ̊ ɣʲ χ̞/ <вь въ рь р ғь хъ>
/b͡βʲ d̪͡ðʲ g͡ɣʲ/ <бъ дъ гъ>
/j w/ <й ў>
/l/ <л>

/ɨ ʉ/ <i у>
/e̞ (œ̞) ə (ɤ̞) o̞/ <е ө ы э о>
/ä ɒ̈/ <ә a>

Anyway it's completely awful from the standpoint of realism sooo

[sangi39]

I just came up with this inventory, come to think about it- maybe I should call Gleb to get a job offer:

(Working Title: Rhasman- not that the name suggests a romanization)

/m mʲ n̪ n̪ʲ ŋ ŋʲ/
/p pʲ b bʲ t̪ t̪ʲ d̪ d̪ʲ k kʲ g gʲ/
/f fʲ v s sʲ z χ χʲ ʁ̝/
/βʲ ʋ̊ ɹ̪ʲ ɹ̊ ɣʲ χ̞/
/b͡βʲ d̪͡ðʲ g͡ɣʲ/
/j w/
/l/

/ɨ ʉ/
/e̞ (œ̞) ə (ɤ̞) o̞/
/ä ɒ̈/

()= phonemes I might drop from the finished project, therefore feel feel free to not romanize them

I tried to fit this around something like Irish orthography:

Broad : Slender
m mʲ <m>
n̪ n̪ʲ <n>
ŋ ŋʲ <ng>
p pʲ <p>
b bʲ <b>
t̪ t̪ʲ <t>
d̪ d̪ʲ <d>
k kʲ <c>
g gʲ <g>
f fʲ <ph>
χ̞ <h>
s sʲ <s>
χ χʲ <ch>
ʋ̊ <f>
w <v>
v βʲ b͡βʲ* <bh>
ɹ̊ <r>
z ɹ̪ʲ d̪͡ðʲ* <dh>
ʁ̝ ɣʲ g͡ɣʲ* <gh>
l j <l>

*These 3 consonants are considered “over-slender”.

Vowel marking follows a similar rule to the Irish “slender with slender and broad with broad” where vowels graphs on either side of a consonant in a word indicate whether that consonant is palatalised or not.

Code: Select all

ɨ: B-ui-B S-i-B  B-uí-S S-í-S
ʉ: B-u-B  S-iu-B B-ú-S  S-iú-S
e: B-e-B  S-ie-B B-é-S  S-ié-S
ə: B-oi-B S-ei-B B-oí-S S-eí-S
o: B-o-B  S-io-B B-ó-S  S-ió-S
ä: B-ae-B S-ea-B B-aé-S S-eá-S
ɒ: B-a-B  S-ia-B B-á-S  S-iá-S

ɨ: O-úi-B  O-úí-S
ʉ: O-íu-B O-íú-S
e: O-íe-B O-íé-S
ə: O-éi-B O-éí-S
o: O-ío-B O-íó-S
ä: O-éa-B O-éá-S
ɒ: O-ía-B O-iá-S

There is no way, as of yet, to unambiguously mark an over-slender consonant without a following vowel.

I dropped the vowels in brackets straight away, but I'd go for umlauted <e> and <o> for /ɤ̞/ and /œ̞/ respectively with acute umlauts where required.

Samples would be nice just to see how this turns out

[Ambrisio]

Here's a(n easy) Gleblang:

Nasal m n nˤ
Stop p t k kˤ ʔ
Fricative s ħ ʕ
Approximant l

High i iː ĩ ĩː u uː ũ ũː
Mid-high e eː ẽ ẽː o oː õ õː
Mid-low ɛ ɛː ɛ̃ ɛ̃ː ɔ ɔː ɔ̃ ɔ̃ː
Low a aː ã ãː

syllable structure:
C1C2V where C1 is an alveolar nonpharyngealized consonant, C2 and V are a consonant and vowel.

[Zontas]

/m mʲ n̪ n̪ʲ ŋ ŋʲ/ <m m' n n' ŋ ŋ'>
/p pʲ b bʲ t̪ t̪ʲ d̪ d̪ʲ k kʲ g gʲ/ <p p' b b' t t' d d' k k' g g'>
/f fʲ v s sʲ z χ χʲ ʁ̝/ <f f' v s s' z h h' ğ>
/βʲ ʋ̊ ɹ̪ʲ ɹ̊ ɣʲ χ̞/ <v' hv r' hr ğ' x>
/b͡βʲ d̪͡ðʲ g͡ɣʲ/ <bh' dh' gh'>
/j w/ <j w>
/l/ <l>

/ɨ ʉ/ <i u>
/e̞ (œ̞) ə (ɤ̞) o̞/ <e ö y õ o>
/ä ɒ̈/ <ä a>

I am surprised that you are dropping Ö and Õ but not X (in the orthography). I think X and H are much less distinct than Ö and Õ, so realistically I would expect the latter set to survive rather than the former.

You might not've noticed there is no /h/...

@Sangi39 It was in fact based on Old Irish.

[R.Rusanov]

I think he means "χ̞" as compared to "χ".

[Zontas]

I think he means "χ̞" as compared to "χ".

Ahh, thank you Red Rusanov.

[Vuvuzela]

Rashman in:
Latin
/m mʲ n̪ n̪ʲ ŋ ŋʲ/<m mj n nj ng ngj
/p pʲ b bʲ t̪ t̪ʲ d̪ d̪ʲ k kʲ g gʲ/ <p pj b bj t tj d dj c cj g gy>
/f fʲ v s sʲ z χ χʲ ʁ̝/<f fj v s sj z x xj r>
/βʲ ʋ̊ ɹ̪ʲ ɹ̊ ɣʲ χ̞/ <vj hw zj hz rj hr>
/b͡βʲ d̪͡ðʲ g͡ɣʲ/<bvj dzj grj>
/j w/<j w>
/l/ <l>

/ɨ ʉ/<i u>
/e̞ (œ̞) ə (ɤ̞) o̞/<ei oe e eo o>
/ä ɒ̈/<a ao>

<h> is used to break vowel digraphs, and <'> to break consonant digraphs

[Qwynegold]

I need help with this Hunglang.

m mː mːː mʲ mʲː mʲːː n nː ɲ ɲː ŋ ŋːk
M MM * M MM * N NN NY NNY N NNK
p pː pʲ pʲː ɓ ɓʲ t tː ɗ c cː ʄ k kː kʲː ɠ ʔ
P PP P PP B B T TT D TY TTY DY K KK KK G X
t͡s t͡sː t͡ɕ t͡ɕː
C CC CY CCY
f fː fːː fʲ fʲː fʲːː v vʲ s z ʑ x xː ç çː
F FF * F FF * V V Z Z ZY H HH HY HHY
ʋ ʋː ʋʲ ʋʲː ɫ ɫː ɫːː ʎ ʎː ʎːː
V VV V VV L LL * LY LLY *
i iː y yː u uː
I Í Ü Ű U Ú
ei
Ě
ø̞ ø̞ː o̞ o̞ː
Ö Ő O Ó
ɛ ɛː
E É
ɑ ɒː
A Á
Y is skipped before I, Í, Ü, Ű, Ě. V is used instead of F when preceding a consonant.

Things in red are allophones, capital letters are orthography that I've decided on so far. What I need help with is deciding how to write the overlong consonants in a way that isn't too un-Hungarian. You can redo the whole orthography for your own enjoyment if you like, but please give me suggestions for the overlong consonants that will work with the rest of the orthography I have here. I've been thinking about writing them with MX, FX, etc, because these overlong consonants stem from ʔ shit. But I don't quite like the X in this orthography. Luckily X for /ʔ/ is not going to be very common, because it only appears word finally, and if the next word begins with a consonant, that consonant will be used instead of X because the /ʔ/ turns into gemination. Oh, and I, Í, Ü, Ű, Ě causes palatalization, just so you know.

Here's a word list:
ˈmɛ
ˈɛ
ˈtɛ
ˈnɛ
ˈtoi
ˈtɛːv
ˈiɛv
ˈkuk
ˈmʲit
ˈmʲiʔ
ˈmʲiʋːo
ˈmʲitɛ
ˈei
ˈkɑicː
ˈmõɲ
ˈtotkut
ˈmuːtɑ̃m
ˈmuː
ˈyʄ
ˈkɑʄ
ˈkovm
ˈnɛft
ˈvʲiːʔ
ˈiʔ
ˈpʲitkɛ
ˈvɛʋɛː
ˈpɑĩnɑv
ˈpɛ̃ɲ
ˈvʲyːt
ˈkɑpɛː
ˈout
ˈnɑĩnɛ
ˈmɛ
ˈvɑɓ
ˈvɑĩm
ˈeic
ˈiʔ
ˈɛʋʲei
ˈkɑv
ˈkoit
ˈciɫpː
ˈkɛːtmɛ
ˈmɛɗ
ˈvɛzci
ˈt͡soʔ
ˈtɑʋːɑ
ˈpovː
ˈciːt
ˈmɑɠ
ˈt͡sot
ˈyvkɛɗ
ˈoɠɛːtɒː
ˈɛːcitːɛː
ˈnɛst
ˈmɛɗɛɗɛː
ˈvɛːtɛː
ˈkɛːcyɛ
ˈpʲyːzciɛ
ˈcyøtɛː
ˈovmɑ
ˈvʲeicːiɛ
ˈtɛzci
ˈtusk
ˈciɛ
ˈvɒːz
ˈvõn
ˈmɛɫk
ˈkɑ̃ŋːk

Lol, there are cox and kuk (cock in Swedish) in this word list when written with my orthography.

EDIT: Vowel nasalization is also allophonic.

[Ambrisio]

Ma pakuksin, et pead just kasutama kaksiktähti ülipikkade häälikute jaoks (nt. MM, NN, ja nii edasi). See toodaks homograafeid--aga igatahes, oleme neid harjunud (nt inglise keeles lead 'juhtima' ja lead 'plii').
(I assure you that if you can understand your conlang, you can understand this. )

Did you have these translations in mind?

ˈmɛ "I"
ˈɛ "you"
ˈtɛ "he/she"
ˈnɛ "they"
ˈkuk "who"
ˈmʲit "what"
ˈmʲiʋːo "when"
ˈmʲitɛ "not"
ˈei "doesn't"
ˈkɑicː "all"
ˈmõɲ "some"
ˈmuː "other"
ˈyʄ "one"
ˈkɑʄ "two"
ˈkovm "three"
ˈnɛft "four"
ˈvʲiːʔ "five" (perhaps too close for comfort)
ˈpʲitkɛ "tall"
ˈvɛʋɛː "blood"
ˈpɛ̃ɲ "thin"
ˈnɑĩnɛ "woman"
ˈmɛ "man"
ˈvɑɓ "free"
ˈvɑĩm "spirit"
ˈkɑv "ear"
ˈciɫpː "eye"
ˈkɛːtmɛ "hand"
ˈmɛɗ "honey"
ˈvɛzci "water"

Generally, why do conlangers love Uralic languages (especially Finnic -- this looks more Finnic than Ugric anyway).

[Nortaneous]

double the preceding vowel?

edit: also, i'd have <gy> instead of <dy>, <q> instead of <x>, and ë instead of e-breve, since ë already appears in transcriptions of dialectal hungarian

me e te ne toj tév jev kuk mit miq mivvo mite ë kajtty mony totkut mútam mú ügy kagy kovm neft víq iq pitke vevé pajnav peny vűt kapé out naine me vab vajm ëty iq evë kav koit tyilpp kétme med veztyi coq tavva povv tyiit mag cot üvked ogétá étyitté nest mededé vété ketyüe pűztyie tyüöté ovma vëttyie teztyi tusk tyie váz von melk kannk

[Qwynegold]

My new phonology (as seen in the phonology thread) needs a romanisation.

Consonants:
/m~n~ŋ~ɴ/
/p b t d k g/
/pʷ bʷ tʷ dʷ kʷ gʷ/
/pˠ bˠ tˠ dˠ kˠ gˠ/
/pˠʷ bˠʷ tˠʷ dˠʷ kˠʷ gˠʷ/
/f v s z ɣ χ/
/fʷ vʷ sʷ zʷ ɣʷ~wˠ χʷ/
/fˠ vˠ sˠ zˠ/
/fˠʷ vˠʷ sˠʷ zˠʷ/
/l~ɹ lʷ~ɹʷ lˠ~ɹˠ lʷˠ~ɹʷˠ w/

Vowels:
/i y ɯ u e ø ǝ ɤ o a ɒ/

Sample Text:
tʷɒ fpmu tsˠɯ a ksˠekø? kˠɯi̯ai̯pˠʷɯi̯ pfɒ χklu yu̯χtni stsǝ pfɒy̯fkɤ. pʷo tsˠʷɤtnɯtɒ ksa psou̯ɯfpli tχyplekˠɤ kfˠʷɤ tsǝ. tnɣeχpχʷǝ spsifkʷyto a ɒu̯ tˠø ɯ?

For marking labialization without velarization you can use ◌̫ (U+032B) while ʷ usually means both.

[Qwynegold]

Ambrisio's Ziilbel
m̥ m n̥ n ŋ̥ ŋ ɴ̥ ɴ
hm m hn n hń ń hǹ ǹ
pʰ p tʰ t kʰ k qʰ q
p b t d k g q c
ɸ ɸː s
f ff s
l lː
l ll

i ɨ u
i e u
a
a

[Grunnen]

how about:

/p pʰ t tʰ k kʰ ʔ/
/ts tsʰ tʃ tʃʰ/
/tɬ/
/f s ʃ x χ h/
/ɬ/
/l ʁ/
/m n ŋ/

/i ə u/ ('high vowels')
/e a o/ ('low vowels')

syllable structure: CV(C)

Sample sentence:

tsatʰa laʁomai̯ni̯a.

[Qwynegold]

Ma pakuksin, et pead just kasutama kaksiktähti ülipikkade häälikute jaoks (nt. MM, NN, ja nii edasi). See toodaks homograafeid--aga igatahes, oleme neid harjunud (nt inglise keeles lead 'juhtima' ja lead 'plii').
(I assure you that if you can understand your conlang, you can understand this. )

Mä olin nähny sun englanninkielisen vastauksen ennen kuin muutit sen eestiksi, mutta en muista niin hyvin mitä siinä luki. ...että ... vaan ... jatkoksi ... MM, NN ja niin edelleen). Se tuo homografit ... olemme harjoitellut näitä ... englanniksi lead - johtaa ja lead - lyijy.

Did you have these translations in mind?

ˈmɛ "I"
ˈɛ "you"
ˈtɛ "he/she" you (pl.)
ˈnɛ "they"
ˈkuk "who"
ˈmʲit "what"
ˈmʲiʋːo "when"
ˈmʲitɛ "not" how
ˈei "doesn't"
ˈkɑicː "all"
ˈmõɲ "some" many
ˈmuː "other"
ˈyʄ "one"
ˈkɑʄ "two"
ˈkovm "three"
ˈnɛft "four"
ˈvʲiːʔ "five" (perhaps too close for comfort)
ˈpʲitkɛ "tall"
ˈvɛʋɛː "blood" broad
ˈpɛ̃ɲ "thin" small
ˈnɑĩnɛ "woman"
ˈmɛ "man"
ˈvɑɓ "free" child
ˈvɑĩm "spirit" Where did you get spirit from? wife
ˈkɑv "ear" fish
ˈciɫpː "eye" louse
ˈkɛːtmɛ "hand" snake
ˈmɛɗ "honey" forest
ˈvɛzci "water" leaf

Generally, why do conlangers love Uralic languages (especially Finnic -- this looks more Finnic than Ugric anyway).

Good guesses!

I don't love Finnic languages that much, it's just easy to work with a language you know well.

[Qwynegold]

double the preceding vowel?

edit: also, i'd have <gy> instead of <dy>, <q> instead of <x>, and ë instead of e-breve, since ë already appears in transcriptions of dialectal hungarian

me e te ne toj tév jev kuk mit miq mivvo mite ë kajtty mony totkut mútam mú ügy kagy kovm neft víq iq pitke vevé pajnav peny vűt kapé out naine me vab vajm ëty iq evë kav koit tyilpp kétme med veztyi coq tavva povv tyiit mag cot üvked ogétá étyitté nest mededé vété ketyüe pűztyie tyüöté ovma vëttyie teztyi tusk tyie váz von melk kannk

Hmm... there are situations where there will be things like áa for /ɑː.ɑ/. There wasn't a single case in my word list, but I think it will be somewhat common. If the overlong consonants are marked like you suggest, I wonder if it would look confusing.

Yeah, I thought about <gy>, but it's hard to justify that conworld historically. Anyhow, it doesn't have to be super close to Hungarian. I chose e-caron because they have that in Czech.

[Nortaneous]

/p pʰ t tʰ k kʰ ʔ/ <p b t d k g 1>
/ts tsʰ tʃ tʃʰ/ <z c j q> ~ <dz ts j ch>
/tɬ/ <tl>
/f s ʃ x χ h/ <f s sh x xh h>
/ɬ/ <lh>
/l ʁ/ <l r>
/m n ŋ/ <m n ng>

/i ə u/ <i ə u>
/e a o/ <e a o>

tsatʰa laʁomai̯ni̯a
zata laromainia

[Ambrisio]

Mä olin nähny sun englanninkielisen vastauksen ennen kuin muutit sen eestiksi, mutta en muista niin hyvin mitä siinä luki. ...että ... vaan ... jatkoksi ... MM, NN ja niin edelleen). Se tuo homografit ... olemme harjoitellut näitä ... englanniksi lead - johtaa ja lead - lyijy.

Kyllä, mä tein sen tahallaan.

Good guesses!

I thought (mainly because of the numbers) that the language was closer to Finnic than Ugric. It turns out I was wring in a systematic way (there appears to be a sound shift l -> v, as in lapsi -> vɑɓ, kala -> kɑv, lehti -> vɛzci). Maybe you could look into the Komi lesson thread from L&L Museum for an actual Uralic language that uses l -> v.

I got "spirit" from Estonian vaim. "Wife" was going to be my second guess, however.

Isn't Tolkien's Quenya based on Finnish? (Although it isn't this close.)

[Grunnen]

/p pʰ t tʰ k kʰ ʔ/ <p b t d k g 1>
/ts tsʰ tʃ tʃʰ/ <z c j q> ~ <dz ts j ch>
/tɬ/ <tl>
/f s ʃ x χ h/ <f s sh x xh h>
/ɬ/ <lh>
/l ʁ/ <l r>
/m n ŋ/ <m n ng>

/i ə u/ <i ə u>
/e a o/ <e a o>

tsatʰa laʁomai̯ni̯a
zata laromainia

Interesting. Btw, where does the 1 come from, or was that supposed to be a <q>?

Looking at that I suddenly thought of this:
/p pʰ t tʰ k kʰ ʔ/ <b p d t g k q>
/ts tsʰ tʃ tʃʰ/ <ds ts dz tz>
/tɬ/ <dl>
/f s ʃ x χ h/ <f s z x xh h>
/ɬ/ <l>
/l ʁ/ <r rh>
/m n ŋ/ <m n ng>

/i ə u/ <i ə u>
/e a o/ <e a o>

tsatʰa laʁomai̯ni̯a
dsata rarhomainia

[Nortaneous]

Interesting. Btw, where does the 1 come from, or was that supposed to be a <q>?

No, I meant <1>. The Latin orthography for Chechen does it, as a holdover from Cyrillic, where they substituted in the number 1 for the palochka, the similar-looking Cyrillic letter for the glottal stop, which had bad font support. It's less ugly than <7> or the glottal stop letter and I don't like any of the other possible ways to write it at all.

[Grunnen]

Interesting. Btw, where does the 1 come from, or was that supposed to be a <q>?

No, I meant <1>. The Latin orthography for Chechen does it, as a holdover from Cyrillic, where they substituted in the number 1 for the palochka, the similar-looking Cyrillic letter for the glottal stop, which had bad font support. It's less ugly than <7> or the glottal stop letter and I don't like any of the other possible ways to write it at all.

Ah that's funny, I didn't know that. I agree there isn't a really good or straightforward way of representing the glottal stop in the latin alphabet. But I must say I don't like digits too much when they're used as letters. Which is off course because I'm not used to it.

[communistplot]

/p pʰ t tʰ k kʰ ʔ/ <b p d t g k q>
/ts tsʰ tʃ tʃʰ/ <tz ts tź tś>
/tɬ/ <tł>
/f s ʃ x χ h/ <f s ś x ħ h>
/ɬ/ <ł>
/l ʁ/ <l r>
/m n ŋ/ <m n ň>

/i ə u/ <i ĕ u>
/e a o/ <e a o>

Semi-vocalic [i̯] is transcribed <j>.

[tsatʰa laʁomai̯ni̯a]
tzata laromajnja