Random phonology/phonemic inventory thread [2011–2018]
Re: Random phonology/phonemic inventory thread
The allophony is awesomely crazy.
Kaipói, low register:
/m n̻ ŋ/ ‹m n g›
/b t̻ kx ʔ/ ‹b t k x›
/s̪ x ʕ/ ‹s h l›
/ɽ/ ‹r›
/ì í ù ú/ ‹i í u ú›
/è é ø̀ ǿ ò ó/ ‹e é ø ǿ o ó›
/à á/ ‹a á›
Kaipói, low register:
/m n̻ ŋ/ ‹m n g›
/b t̻ kx ʔ/ ‹b t k x›
/s̪ x ʕ/ ‹s h l›
/ɽ/ ‹r›
/ì í ù ú/ ‹i í u ú›
/è é ø̀ ǿ ò ó/ ‹e é ø ǿ o ó›
/à á/ ‹a á›
Re: Random phonology/phonemic inventory thread
The allophones occur in:2-4 wrote:Where?Nemesis wrote:Several alllophones can occur: ŋ r̊ l̊ ʝ ɟ͡ʝ β ɹ ɪ̈.
If the vowels are allophones, then they are not phoneMic, but phoneTic.Nemesis wrote:Some of the phonemic vowels presented may be considered alophones as ɘ and ʊ both alophones of unstressed /e/ and /u/ respectively. Dokian is an SVO language highly influenced by romance languages.
ŋ - allofone of /n/ before a velar plosive or a velar fricative.
r̊ - allofone of /ɾ/ in syllable coda position.
l̊ - allophone of /l/ in syllable coda position, especially after a diphthong.
ʝ - allophone of the diphthong /ji/ in which /j/ can become a fricative.
ɟʝ - allophone of /ɟ/, especially word internally.
β - allophone of /b/ between vowels.
ɹ - allophone of /ɾ/ before a palatal plosive.
ɪ̈ - allophone of long /i/ written ortographically as /ie/ or /ii/, which usually becomes /jɪ̈/.
The sound [ç] can be viewed as an allophone of /h/ before /i/ in most approaches, but some may consider it as an independent sound because there are words containing this sound independently from /h/, and there is even a digraph in Dokian to represent this sound <cy> maily from loan words.You say that certain sounds "may be considered allophones" of other sounds. What would speak in favour of analysing them as allophones, and what would speak in favour of analysing them as phonemes?
As for the vowels /ɘ/ and /ʊ/ some may consider them as allophones of unstressed /e/ and /o/ or /u/, but there are some minimal pairs that may lead to conffusion. And there are some dialects that make a distinction of those vowels and others that don't.
(Docian)
Re: Random phonology/phonemic inventory thread
Thank you for writing where the allophones occur.
Note that the orthography is written between <> or ‹›.
If I recall correctly, a similar thing happened with /χ/ and /ç/ in German.
You forgot where these allophones occur. In unstressed syllables or where?Nemesis wrote:ʝ - allophone of the diphthong /ji/ in which /j/ can become a fricative.
ɪ̈ - allophone of long /i/ written ortographically as /ie/ or /ii/, which usually becomes /jɪ̈/.
Note that the orthography is written between <> or ‹›.
Interesting, can you give some examples?Nemesis wrote:As for the vowels /ɘ/ and /ʊ/ some may consider them as allophones of unstressed /e/ and /o/ or /u/, but there are some minimal pairs that may lead to conffusion. And there are some dialects that make a distinction of those vowels and others that don't.
If I recall correctly, a similar thing happened with /χ/ and /ç/ in German.
Re: Random phonology/phonemic inventory thread
[a b c d e f g h i j k l m n o p q r s t u v w x y z]
<a b c d e f g h i j k l m n o p q r s t u v w x y z>
How simple
<a b c d e f g h i j k l m n o p q r s t u v w x y z>
How simple
:ces: Native
Mutually intelligibile with native language
Almost fluent
Little
Little more
Everybody can speak it!
Speedlang
Mutually intelligibile with native language
Almost fluent
Little
Little more
Everybody can speak it!
Speedlang
Re: Random phonology/phonemic inventory thread
REVOLUTIONARYBatrachus wrote:[a b c d e f g h i j k l m n o p q r s t u v w x y z]
<a b c d e f g h i j k l m n o p q r s t u v w x y z>
How simple
Maxédri
Faljüdax
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Random phonology/phonemic inventory thread
I created this and I called it "Ipish" (from IPA) or as an internal name maybe <Aipis>/aipis/[aipis]?Batrachus wrote:[a b c d e f g h i j k l m n o p q r s t u v w x y z]
<a b c d e f g h i j k l m n o p q r s t u v w x y z>
How simple
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
Re: Random phonology/phonemic inventory thread
So I was having issues with IPA (again) and I can't seem to fix it, so I'll describe the phones and give an example if it is English-like.
Note: most of my 'letters' have sound morphs, meaning one phone morphs to another phone depending on rules or at random. I'll put that in too.
Tiyen:
l - voiced labiodental (I believe they underline the dentals in IPA)
d - voiced dental
t - voiceless dental
n - voiced dental nasal
m - as in 'mother'
T - as in 'thorough'
s - as in sip = morphs to: ss - as in ship = and to: SS - as in chip (at random)
z - as in zipper
j - voiced domed postalveolar fricative
x - a 'k', but soft like 'kh'
y - (IPA does 'j' I believe) as in yoddle = morphs to: g as in gold (at random) (g is new)
v - as in violin
f - as in fall
r - as in rip
w - voiced labiovelar approximant
h - as in hero
b - as in bold = morphs to: p - as in pip (b must be followed by a consonant or 'a' or it morphs to p, which can be followed only by the consonants l and y if these are then followed by o, i, or u/uu, otherwise p is followed by a vowel other than 'a'.) (b/p is new)
u - as in under = uu - as in soon
o - as in open = oii/oy - as in boy
i - as in ill = ii - as in feel
e - as in elf = eii/ey - as in age
a - as in off = aii/ay - as in ice (ae at the end of a word also makes aii/ay as in ice)
And that's all of the phonology. What do you think? (I know it's super bad. Eek!) :D
(Uh...the letters I use are the same ones I use when I write out my language. It's quicker than grabbing IPA.)
Note: most of my 'letters' have sound morphs, meaning one phone morphs to another phone depending on rules or at random. I'll put that in too.
Tiyen:
l - voiced labiodental (I believe they underline the dentals in IPA)
d - voiced dental
t - voiceless dental
n - voiced dental nasal
m - as in 'mother'
T - as in 'thorough'
s - as in sip = morphs to: ss - as in ship = and to: SS - as in chip (at random)
z - as in zipper
j - voiced domed postalveolar fricative
x - a 'k', but soft like 'kh'
y - (IPA does 'j' I believe) as in yoddle = morphs to: g as in gold (at random) (g is new)
v - as in violin
f - as in fall
r - as in rip
w - voiced labiovelar approximant
h - as in hero
b - as in bold = morphs to: p - as in pip (b must be followed by a consonant or 'a' or it morphs to p, which can be followed only by the consonants l and y if these are then followed by o, i, or u/uu, otherwise p is followed by a vowel other than 'a'.) (b/p is new)
u - as in under = uu - as in soon
o - as in open = oii/oy - as in boy
i - as in ill = ii - as in feel
e - as in elf = eii/ey - as in age
a - as in off = aii/ay - as in ice (ae at the end of a word also makes aii/ay as in ice)
And that's all of the phonology. What do you think? (I know it's super bad. Eek!) :D
(Uh...the letters I use are the same ones I use when I write out my language. It's quicker than grabbing IPA.)
The above title is DECEPTIVE: I am a girl...and a succubus.
Re: Random phonology/phonemic inventory thread
Then use X-SAMPA. You can convert it to IPA using this.kingdemon wrote:So I was having issues with IPA (again) and I can't seem to fix it [...]
L can't be labiodental because it is articulated with the tongue. I think you meant dental.kingdemon wrote:l - voiced labiodental
I have troubles understading this. Can you clarify, please?kingdemon wrote:b - as in bold = morphs to: p - as in pip (b must be followed by a consonant or 'a' or it morphs to p, which can be followed only by the consonants l and y if these are then followed by o, i, or u/uu, otherwise p is followed by a vowel other than 'a'.) (b/p is new)
My try at converting your phonology to IPA:
Consonants:
/m n̪/
/b t̪ d̪ ɡ~j/
/f v θ s~ʃ~t͡ʃ z ʒ x h/
/l̪ ɹ w/
~ indicates free variation ("at random")
Vowels:
/iː uː/
/ɪ/
/e eɪ̯ o oʊ̯/
/aɪ̯ ɑ ʌ/
Allophony:
/b/ is realised ("morphs to") [p] in some environments I don't understand well.
/ae/ is realised as [aɪ̯] at the end of words.
The phonology is not super bad.kingdemon wrote:And that's all of the phonology. What do you think? (I know it's super bad. Eek!) :D
Consonants are quite interesting, but I suggest changing /x/ to /k/ to make the consonant inventory more natural.
I think the vowels are influenced by English which has weird vowels thanks to the Great Vowel Shift. I suggest looking at this for some inspiration.
I hope I've helped!
Re: Random phonology/phonemic inventory thread
Proto-Tsiwáka
/m n ŋ-ɴ/ <m n ng>
/p t k q/ <p t k q>
/t͡s t͡ʃ t͡ɬ/ <ts tj tl>
/ɸ-f θ s ʁ-h/ <f th s qh>
/ʋ l-ɬ/ <v l>
/a-ɐ i-ɪ u-ɯ-ʊ/ <a i u>
/a: i: u:/ <á í ú>
I'm thinking this will be my first language using triconsonantal roots, purely because I like the pair 'Tsáwak, Tsiwáka'.
/m n ŋ-ɴ/ <m n ng>
/p t k q/ <p t k q>
/t͡s t͡ʃ t͡ɬ/ <ts tj tl>
/ɸ-f θ s ʁ-h/ <f th s qh>
/ʋ l-ɬ/ <v l>
/a-ɐ i-ɪ u-ɯ-ʊ/ <a i u>
/a: i: u:/ <á í ú>
I'm thinking this will be my first language using triconsonantal roots, purely because I like the pair 'Tsáwak, Tsiwáka'.
Re: Random phonology/phonemic inventory thread
Yes you have helped. :)2-4 wrote:Then use X-SAMPA. You can convert it to IPA using this.kingdemon wrote:So I was having issues with IPA (again) and I can't seem to fix it [...]
L can't be labiodental because it is articulated with the tongue. I think you meant dental.kingdemon wrote:l - voiced labiodental
I have troubles understading this. Can you clarify, please?kingdemon wrote:b - as in bold = morphs to: p - as in pip (b must be followed by a consonant or 'a' or it morphs to p, which can be followed only by the consonants l and y if these are then followed by o, i, or u/uu, otherwise p is followed by a vowel other than 'a'.) (b/p is new)
My try at converting your phonology to IPA:
Consonants:
/m n̪/
/b t̪ d̪ ɡ~j/
/f v θ s~ʃ~t͡ʃ z ʒ x h/
/l̪ ɹ w/
~ indicates free variation ("at random")
Vowels:
/iː uː/
/ɪ/
/e eɪ̯ o oʊ̯/
/aɪ̯ ɑ ʌ/
Allophony:
/b/ is realised ("morphs to") [p] in some environments I don't understand well.
/ae/ is realised as [aɪ̯] at the end of words.
The phonology is not super bad.kingdemon wrote:And that's all of the phonology. What do you think? (I know it's super bad. Eek!) :D
Consonants are quite interesting, but I suggest changing /x/ to /k/ to make the consonant inventory more natural.
I think the vowels are influenced by English which has weird vowels thanks to the Great Vowel Shift. I suggest looking at this for some inspiration.
I hope I've helped!
I remember you pointing me in that direction to the X-Sampa (Which I don't understand at all, sadly :( ) and it was working when I was trying it out on a word document, but not when I was making a post. I'll try it again of course and see if it is just user error, which is oft the case. :P
And yes you're right it is a dental l. :D I was tired, but I think I had it listed as a labiodental in my phones as well, which I'll have to correct when I get home. :)
As for b/p, they're really new and I'm still working out how they are morphing inside of my language. Most are random morphs or because of a double letter (technically most doubled letters are silent with the exception of ss = sh and zz = zh) and I'll probably turn them to random as well since it is pretty confusing. :) It all made sense to me last night I swear, but looking at it this morning...I'm not so sure. :P
As for x to k, I wanted a softer sound, mostly because I like softer phones, but also because the species reiith have a very high arch of the roof of their mouth preventing their short tongue from touching anything except their teeth, making harder sounds such as k or even a true g difficult, hence why l/t/d/n are all dental.
Of course k might be able to be soft...I'm just not sure.
Thank you again for all your help. :)
The above title is DECEPTIVE: I am a girl...and a succubus.
Re: Random phonology/phonemic inventory thread
Proto-Littoran:
/m n ɲ ŋ/ ‹m n ñ ŋ›
/p t c k/ ‹p t c k›
/f s h/ ‹f s h›
/r/ ‹r›
/i u/ ‹i u›
/ə/ ‹ə›
/a/ ‹a›
Isenian, its descendant:
/m n/ ‹m n›
/p ᵐb t ⁿd k ᵑɡ/ ‹p mb t nd c ng›
/s χ/ ‹s h›
/r/ ‹r›
/i iː u uː/ ‹i í u ú›
/ə/ ‹e›
/a aː/ ‹a á›
An inventory quite similar to Northeastern Littoran languages:
/m n/
/p t ʨ ʔ/
/s z/
/i e a o u/ + 5 tones, 2+ phonations
/m n ɲ ŋ/ ‹m n ñ ŋ›
/p t c k/ ‹p t c k›
/f s h/ ‹f s h›
/r/ ‹r›
/i u/ ‹i u›
/ə/ ‹ə›
/a/ ‹a›
Isenian, its descendant:
/m n/ ‹m n›
/p ᵐb t ⁿd k ᵑɡ/ ‹p mb t nd c ng›
/s χ/ ‹s h›
/r/ ‹r›
/i iː u uː/ ‹i í u ú›
/ə/ ‹e›
/a aː/ ‹a á›
An inventory quite similar to Northeastern Littoran languages:
/m n/
/p t ʨ ʔ/
/s z/
/i e a o u/ + 5 tones, 2+ phonations
Last edited by Click on 29 Oct 2012 14:41, edited 1 time in total.
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Random phonology/phonemic inventory thread
Is there a particular reason for having your proto-/k/ romanized as <k> and your descendant-/k/ as <c>? Especially since you use <c> in your proto-romanization for /c/.2-4 wrote:Proto-Littoran:
/m n ɲ ŋ/ ‹m n ñ ŋ›
/p t c k/ ‹p t c k›
/f s h/ ‹f s h›
/r/ ‹r›
/i u/ ‹i u›
/ə/ ‹ə›
/a/ ‹a›
Isenian, its descendant:
/m n/ ‹m n›
/p ᵐb t ⁿd k ᵑɡ/ ‹p mb t nd c ng›
/s χ/ ‹s h›
/r/ ‹r›
/i iː u uː/ ‹i í u ú›
/ə/ ‹e›
/a aː/ ‹a á›
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
Re: Random phonology/phonemic inventory thread
Æsthetics. I think that ‹cápsún› ‹cutisá› ‹các› ‹ciát› etc. look much better than ‹kápsún› ‹kutisá› ‹kák› ‹kiát› etc.Creyeditor wrote:Is there a particular reason for having your proto-/k/ romanized as <k> and your descendant-/k/ as <c>? Especially since you use <c> in your proto-romanization for /c/.
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Random phonology/phonemic inventory thread
Okay, I thought there could have been a crazy sound change or something like that, resulting in the elision of /k/<k>, with fomer <c>/c/ taking its place as new <c>/k/.2-4 wrote:Æsthetics. I think that ‹cápsún› ‹cutisá› ‹các› ‹ciát› etc. look much better than ‹kápsún› ‹kutisá› ‹kák› ‹kiát› etc.Creyeditor wrote:Is there a particular reason for having your proto-/k/ romanized as <k> and your descendant-/k/ as <c>? Especially since you use <c> in your proto-romanization for /c/.
Well, but esthetics are something individual, something personal, I guess.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
Re: Random phonology/phonemic inventory thread
No crazy sound changes. They are reserved for the northeastern branch of the Littoran family. /c/ and /k/ simply merged into /k/.Creyeditor wrote:Okay, I thought there could have been a crazy sound change or something like that, resulting in the elision of /k/<k>, with fomer <c>/c/ taking its place as new <c>/k/.2-4 wrote:Æsthetics. I think that ‹cápsún› ‹cutisá› ‹các› ‹ciát› etc. look much better than ‹kápsún› ‹kutisá› ‹kák› ‹kiát› etc.Creyeditor wrote:Is there a particular reason for having your proto-/k/ romanized as <k> and your descendant-/k/ as <c>? Especially since you use <c> in your proto-romanization for /c/.
Well, but esthetics are something individual, something personal, I guess.
The Northeastern branch is crazy: /k/ changes to /q/, then to [ʔ] [ʕ] [χ] in various environments and these three sounds damage the vowels and give them tone. /c/ also changes to /ʧ/ while /p/ covers the gap in velar stops by changing to [k] before the unrounded vowels.
Some Northeastern languages change [ʔ] [ʕ] to [ŋ]. See rhinoglottophilia on the Wikipedia.
I'll work on the sound changes from Proto-Littoran to some Northeastern language after Tuesday.
Re: Random phonology/phonemic inventory thread
You know guys - I'll start a new job soon, which mean I can't be a full-time conlanger anymore Nevertheless, I wanted to post this little phonology/phoneme inventory I've been thinking of for some time (an hour or so, maybe).
Consonants/initials:
kʰʷ ~ xʷ <khw>
kʷ <kw>
ɠʷ ~ gʷ <gw>
kʰ ~ x <kh>
k <k>
ɠ ~ g <g>
tʰ ~ s <th>
t <t>
ɗ ~ d <d>
ŋ̊ʷ <nghw>
ŋʷ <ngw>
ŋ̊ <ngh>
ŋ <ng>
n̥ <nh>
n <n>
k͡r̥ʰʷ ~ x͡r̥ʷ <khrw>
k͡rʷ <krw>
ɠ͡rʷ ~ g͡rʷ <grw>
k͡r̥ʰ ~ x͡r̥ <khr>
k͡r <kr>
ɠ͡r ~ g͡r <gr>
t͡ɬʰ <thl>
t͡ɬ ~ t͡ɮ <tl>
ɗ͡ɮ ~ d͡ɮ <dl>
r̥ʷ <rhw>
rʷ <rw>
r̥ <rh>
r <r>
ɬ ~ l̥ <lh>
l <l>
ʍ <wh>
w <w>
h <h>
Vowels/finals:
i~ɨ <i>
e <e>
a <a>
ɞ <o>
o <u>
Tones:
Rising <é> etc.
Falling <è>
Level/flat/neutral <e>
The exact realisation of tone may depend on the onset/initial in the syllable - voiced consonant yield a lower start of the tone.
Syllables:
Syllables are made up of exactly one consonant/initial, and one vowel/final:
khrwè [kʰʷr̥ʷe˥˩]
nghá [ŋ̊a˧˥]
grò [ɠrɞ˧˩]
dlí [dɮi˩˧]
rhwu [r̥ʷo˥]
ngwi [ŋʷɨ˩]
Consonants/initials:
kʰʷ ~ xʷ <khw>
kʷ <kw>
ɠʷ ~ gʷ <gw>
kʰ ~ x <kh>
k <k>
ɠ ~ g <g>
tʰ ~ s <th>
t <t>
ɗ ~ d <d>
ŋ̊ʷ <nghw>
ŋʷ <ngw>
ŋ̊ <ngh>
ŋ <ng>
n̥ <nh>
n <n>
k͡r̥ʰʷ ~ x͡r̥ʷ <khrw>
k͡rʷ <krw>
ɠ͡rʷ ~ g͡rʷ <grw>
k͡r̥ʰ ~ x͡r̥ <khr>
k͡r <kr>
ɠ͡r ~ g͡r <gr>
t͡ɬʰ <thl>
t͡ɬ ~ t͡ɮ <tl>
ɗ͡ɮ ~ d͡ɮ <dl>
r̥ʷ <rhw>
rʷ <rw>
r̥ <rh>
r <r>
ɬ ~ l̥ <lh>
l <l>
ʍ <wh>
w <w>
h <h>
Vowels/finals:
i~ɨ <i>
e <e>
a <a>
ɞ <o>
o <u>
Tones:
Rising <é> etc.
Falling <è>
Level/flat/neutral <e>
The exact realisation of tone may depend on the onset/initial in the syllable - voiced consonant yield a lower start of the tone.
Syllables:
Syllables are made up of exactly one consonant/initial, and one vowel/final:
khrwè [kʰʷr̥ʷe˥˩]
nghá [ŋ̊a˧˥]
grò [ɠrɞ˧˩]
dlí [dɮi˩˧]
rhwu [r̥ʷo˥]
ngwi [ŋʷɨ˩]
- eldin raigmore
- korean
- Posts: 6352
- Joined: 14 Aug 2010 19:38
- Location: SouthEast Michigan
Re: Random phonology/phonemic inventory thread
Yay!Xing wrote:You know guys - I'll start a new job soon,
Aww.Xing wrote: which mean I can't be a full-time conlanger anymore
I'll miss you, and I bet I speak for all of us.
My minicity is http://gonabebig1day.myminicity.com/xml
Re: Random phonology/phonemic inventory thread
What is your new job going to be?Xing wrote:You know guys - I'll start a new job soon, which mean I can't be a full-time conlanger anymore
Re: Random phonology/phonemic inventory thread
It's at the municipal archives.2-4 wrote:What is your new job going to be?Xing wrote:You know guys - I'll start a new job soon, which mean I can't be a full-time conlanger anymore
Re: Random phonology/phonemic inventory thread
feeling very cautious with hlosestis because i've never done well with a borrowing (or posteriori) language, so wrote something up on the side
currently unnamed
consonants
stops /p b t̪ d̪ k g ʔ/ <p b t d k g '>
fricatives /ɸ β s z ʃ ʒ x ʁ/ <f v s z š ž h ř>
affricates /tʃ dʒ/ <c j>
approximants /w j/ <w y>
nasals /m n ŋ/ <m n ŋ>
trill /r ʀ/ <r ř>
lateral / l / <l>
vowels
/i i:/ <i í >
/a a:/ <a á>
/u u:/ <u ú>
<ř> is /ʀ/ at the beginning and when following a voiced consonant, and /ʁ/ after an unvoiced consonant.
syllable shape is (C)V(:)(C(:))
i usually prefer to use <j> for /j/ but i also rarely use <dʒ>, and i didn't want any digraphs, so i wasn't sure what i could use. I think it's ok though since i'm using c for /tʃ/ so it looks a better match.
currently unnamed
consonants
stops /p b t̪ d̪ k g ʔ/ <p b t d k g '>
fricatives /ɸ β s z ʃ ʒ x ʁ/ <f v s z š ž h ř>
affricates /tʃ dʒ/ <c j>
approximants /w j/ <w y>
nasals /m n ŋ/ <m n ŋ>
trill /r ʀ/ <r ř>
lateral / l / <l>
vowels
/i i:/ <i í >
/a a:/ <a á>
/u u:/ <u ú>
<ř> is /ʀ/ at the beginning and when following a voiced consonant, and /ʁ/ after an unvoiced consonant.
syllable shape is (C)V(:)(C(:))
i usually prefer to use <j> for /j/ but i also rarely use <dʒ>, and i didn't want any digraphs, so i wasn't sure what i could use. I think it's ok though since i'm using c for /tʃ/ so it looks a better match.