Actually I've a good mind to use the underdot for the retroflexes, <tr> suggests more an affricate pronunciation to me, just I need to work out the encoding for the shortcut.cedh wrote:You could also write /ʁ/ as <g> and the retroflex stops as <r rh r'>, in parallel with the sibilants. Another alternative would be to write the sibilant affricates as either <ts tsh ts'> in parallel with the retroflexes, or as <c ch c'>. Both of the latter are in danger of suggesting a mistakenly shibilant pronunciation of what should be /tsʰ/ though.Frislander wrote:I've been thinking about revisions for Proto-Muyan and this is what I came up with for the phonology.
/t s ʈ k kʷ q/ <t s tr k kw q>
/tʰ t͡sʰ ʈʰ kʰ kʷʰ qʰ h/ <th sh trh kh khw qh h>
/t’ t͡s’ ʈ’ k’ kʷ’ q’ ʔ/ <t' s' tr' k' kw' q' ‘>
/n j ɺ˞ w ʁ/ <n y l w r>
/i u/ <i u>
/ə/ <e>
/æ ɒ/ <a o>
Additionally there is a pitch-accent system. Each content word has at least one accented syllable (most words have only one accent, which wis on the root, though compounding and one or two affixes can produce words with multiple accents) and that syllable may take one of three tones/phonations: /á a̤ a̰/ <á à ã>.
Syllable structure is CV(n, l, r).
Random phonology/phonemic inventory thread [2011–2018]
- Frislander
- mayan
- Posts: 2088
- Joined: 14 May 2016 18:47
- Location: The North
Re: Random phonology/phonemic inventory thread
Re: Random phonology/phonemic inventory thread
p t k k͡p q q͡p
s θ ɬ
z l
m n
s θ ɬ
z l
m n
My meta-thread: viewtopic.php?f=6&t=5760
Re: Random phonology/phonemic inventory thread
Labiouvulars are always nice.Omzinesý wrote:p t k k͡p q q͡p
s θ ɬ
z l
m n
/i y ɯ u
e ø ɤ o
æ ɑ/ + length
<ı ü i u
e ö ė o
ä a>
The vowel system itself's obviously nothing special, just trying out an orthography (could replace diaeresis with a single dot too, just have single dots on e and i because ï looks meh). Long vowels are doubled, orthographically.
A consonant system to go with it, again nothing too special:
/m n ɲ ŋ
p t c k
s ʃ h
l ʎ
ʋ j ʕ
r/
<m n ny ñ
p t ty k
s š h
l ly
v y g/3
r>
(or Cj for palatals)
Could add another series, probably not pharyngeal or uvular, perhaps voiceless sonorants. Most consonants can appear geminate.
- Frislander
- mayan
- Posts: 2088
- Joined: 14 May 2016 18:47
- Location: The North
Re: Random phonology/phonemic inventory thread
Inspired a bit by Naga languages.
/p t tʰ k/
/β s x/
/m n ŋ/
/l j/
/i e ɘ ɐ a u o/
Syllable structure is (C)(j)V, where any other consonant may be followed by the glide /j/.
Stress is non-phonemic and fixed in word-final position.
/p t tʰ k/
/β s x/
/m n ŋ/
/l j/
/i e ɘ ɐ a u o/
Syllable structure is (C)(j)V, where any other consonant may be followed by the glide /j/.
Stress is non-phonemic and fixed in word-final position.
Re: Random phonology/phonemic inventory thread
I stole Swahili's inventory and removed a few sounds, leaving this behind.
/m n ɲ ŋ/ m n ny ng'
/b̃ d̃ t͡ʃ̃ g̃/ mb nd nj ng
/ɓ ɗ ɠ/ b d g
/p t t͡ʃ k/ p t j ch k
/ṽ z̃/ mv nz
/w f v θ ð s z ʃ j h/ w f v th dh s z sh j h
/l/ l
/i ɯ/ i u
/ɤ/ o
/ɛ/ e
/ɑ/ a
The syllable structure is (C)(V).
/m n ɲ ŋ/ m n ny ng'
/b̃ d̃ t͡ʃ̃ g̃/ mb nd nj ng
/ɓ ɗ ɠ/ b d g
/p t t͡ʃ k/ p t j ch k
/ṽ z̃/ mv nz
/w f v θ ð s z ʃ j h/ w f v th dh s z sh j h
/l/ l
/i ɯ/ i u
/ɤ/ o
/ɛ/ e
/ɑ/ a
The syllable structure is (C)(V).
Gândölansch (Gondolan) • Feongkrwe (Feongrkean) • Tamhanddön (Tamanthon) • Θανηλοξαμαψⱶ (Thanelotic) • Yônjcerth (Yaponese) • Ba̧supan (Basupan) • Mùthoķán (Mothaucian)
-
- sinic
- Posts: 401
- Joined: 21 Jul 2012 08:01
- Location: Buffalo, NY
Re: Random phonology/phonemic inventory thread
Sometimes I think about using a "staggered" stop inventory like this:
It's a little obnoxious but it's typologically justified.
Code: Select all
kʼ
t k
b d
ɓ
- DesEsseintes
- mongolian
- Posts: 4331
- Joined: 31 Mar 2013 13:16
Re: Random phonology/phonemic inventory thread
It's a bit try-hard, but definitely nice.Porphyrogenitos wrote:Sometimes I think about using a "staggered" stop inventory like this:
It's a little obnoxious but it's typologically justified.Code: Select all
kʼ t k b d ɓ
Hmm. Let's see where this could go...
Code: Select all
t͡s’ t͡ɬ’ k’ q’
t t͡s k
b d
ɓ
- eldin raigmore
- korean
- Posts: 6352
- Joined: 14 Aug 2010 19:38
- Location: SouthEast Michigan
Re: Random phonology/phonemic inventory thread
I see what you did there.Porphyrogenitos wrote:.... typologically justified.
My minicity is http://gonabebig1day.myminicity.com/xml
- Frislander
- mayan
- Posts: 2088
- Joined: 14 May 2016 18:47
- Location: The North
Re: Random phonology/phonemic inventory thread
I'm thinking I'm gonna use this phonology for a new conlang, or maybe gonna make this my "revision" of Aikuu.
/t k/ <t k>
/β ʂ ɕ h/ <w s c h>
/n ɾ/ <n r>
/i e a ɤ/ <i e a o>
In addition vowels may appear in either high or low tones.
Syllable structure is (C)V(V)(N), where N is one of /h n/ and V(V)(N) syllables only occur word-initially. When two vowels come into contact the vowels are mostly left as-is, however /e/ shifts to /i/ when it follows another vowel in the same syllable. Furthermore when three or more vowels come into contact they are simplified: if two adjacent vowels are identical one is deleted, otherwise the first vowel(s) of the sequence are deleted.
Coda /n/ assimilates to the POA of an immediately following consonant. Furthermore vowels are rounded when adjacent to /β/ and the mid-back vowel /ɤ/ is raised when it follows another vowel (this parallels the process for /e/ described above only this does not result in a merger).
/t k/ <t k>
/β ʂ ɕ h/ <w s c h>
/n ɾ/ <n r>
/i e a ɤ/ <i e a o>
In addition vowels may appear in either high or low tones.
Syllable structure is (C)V(V)(N), where N is one of /h n/ and V(V)(N) syllables only occur word-initially. When two vowels come into contact the vowels are mostly left as-is, however /e/ shifts to /i/ when it follows another vowel in the same syllable. Furthermore when three or more vowels come into contact they are simplified: if two adjacent vowels are identical one is deleted, otherwise the first vowel(s) of the sequence are deleted.
Coda /n/ assimilates to the POA of an immediately following consonant. Furthermore vowels are rounded when adjacent to /β/ and the mid-back vowel /ɤ/ is raised when it follows another vowel (this parallels the process for /e/ described above only this does not result in a merger).
Re: Random phonology/phonemic inventory thread
I'm not sure if I've already posted this.
p t k k͡p q q͡p
b d g g͡b ɢ ɢ͡b
ʋ ɹ ɰ w ʁ ʁ͡β (This series rarely appears word-initially.)
l r
s ʃ
z ʒ
m n
Maybe a bit too symmetric.
p t k k͡p q q͡p
b d g g͡b ɢ ɢ͡b
ʋ ɹ ɰ w ʁ ʁ͡β (This series rarely appears word-initially.)
l r
s ʃ
z ʒ
m n
Maybe a bit too symmetric.
My meta-thread: viewtopic.php?f=6&t=5760
Re: Random phonology/phonemic inventory thread
Consonants
Nasals: /m mʷ n̪ ɳ ɲ ŋ ŋʷ/
Stops: /p pʷ t̪ t͡s ʈ c k kʷ ʔ/
Fricatives: /s x xʷ ħ h/
Laterals: /l ɭ ʎ/
Approximants: /ɻ j w ʕ̞/
Trills: /r/
Modal Vowels
Short: ɪ ɛ ɑ o
Long: iː eː ɑː oː
Diphthongs: ɑi̯ ɑːi̯
Creaky Vowels
Short: ɪ̰ ɛ̰ ɑ̰ o̰
Long: ḭː ḛː ɑ̰ː o̰ː
Diphthongs: ɑ̰i̯ ɑ̰ːi̯
I'm not sure if I want to add a distinct prenasalized series or not but I believe I will be allowing N+Stop clusters at onset. (I am fairly certain it the nasal+stop cluster will yield voiced plosive allophones after the nasal... hm)
Maximum syllable structure will be NCVC
Stops in coda position are aspirated and in word final position they are aspirated quite strongly.
I want to implement creaky-harmony but I will need to fiddle with the language more.
Nasals: /m mʷ n̪ ɳ ɲ ŋ ŋʷ/
Stops: /p pʷ t̪ t͡s ʈ c k kʷ ʔ/
Fricatives: /s x xʷ ħ h/
Laterals: /l ɭ ʎ/
Approximants: /ɻ j w ʕ̞/
Trills: /r/
Modal Vowels
Short: ɪ ɛ ɑ o
Long: iː eː ɑː oː
Diphthongs: ɑi̯ ɑːi̯
Creaky Vowels
Short: ɪ̰ ɛ̰ ɑ̰ o̰
Long: ḭː ḛː ɑ̰ː o̰ː
Diphthongs: ɑ̰i̯ ɑ̰ːi̯
I'm not sure if I want to add a distinct prenasalized series or not but I believe I will be allowing N+Stop clusters at onset. (I am fairly certain it the nasal+stop cluster will yield voiced plosive allophones after the nasal... hm)
Maximum syllable structure will be NCVC
Stops in coda position are aspirated and in word final position they are aspirated quite strongly.
I want to implement creaky-harmony but I will need to fiddle with the language more.
Re: Random phonology/phonemic inventory thread
/n
t ɗ k (kʷ)
β s
l
i ɯ u
e ɑ/
<n
t d k (q)
w/b s
l/r
i o u
e a>
t ɗ k (kʷ)
β s
l
i ɯ u
e ɑ/
<n
t d k (q)
w/b s
l/r
i o u
e a>
-
- sinic
- Posts: 401
- Joined: 21 Jul 2012 08:01
- Location: Buffalo, NY
Re: Random phonology/phonemic inventory thread
Very...interesting. What exactly is the place of /kʷ/? Is it only in some dialects? And what might the phonotactics be?Auvon wrote:/n
t ɗ k (kʷ)
β s
l
i ɯ u
e ɑ/
<n
t d k (q)
w/b s
l/r
i o u
e a>
Re: Random phonology/phonemic inventory thread
It's marginally phonemic, appearing mainly in a few function words and some loanwords. Many speakers will merge it with k.Porphyrogenitos wrote:Very...interesting. What exactly is the place of /kʷ/? Is it only in some dialects? And what might the phonotactics be?Auvon wrote:/n
t ɗ k (kʷ)
β s
l
i ɯ u
e ɑ/
<n
t d k (q)
w/b s
l/r
i o u
e a>
Phonotactics are something like (C)V(A) where A=/n s l/. All words are C-initial, with some method of inserting C when an underlyingly V-initial morpheme appears at the start of a word. There might also be some diphthongs.
Re: Random phonology/phonemic inventory thread
Minimalist inventory I thought up after reading about some Khoisan languages. It has 12 non-clicks and 40 clicks.
/p~β~w t~ɾ k~ɣ~j q~ʁ ʔ/
/tsʰ~s kxʰ~x~h qχʰ~χ/
/k' q'/
/m n~ŋ~ɴ/
Voiceless unaspirated stops are voiced to approximants word medially, while affricates fricativize word medially. The /x/ is realized as /h/ before non-high vowels. The /β/ is realized as /w/ when followed by non-back vowels, and the /ɣ/ is realized as /j/ before non-front vowels.
/ʘ ǀ ǃ ǂ ǁ/
/ʘʰ ǀʰ ǃʰ ǂʰ ǁʰ/
/ʘ' ǀ' ǃ' ǂ' ǁ'/
/ⁿʘ ⁿǀ ⁿǃ ⁿǂ ⁿǁ/
/ⁿʘ' ⁿǀ' ⁿǃ' ⁿǂ' ⁿǁ'/
/qʘ qǀ qǃ qǂ qǁ/
/qʘʰ qǀʰ qǃʰ qǂʰ qǁʰ/
/qʘ' qǀ' qǃ' qǂ' qǁ'/
/a e i o u/
/ai au ei eu oi ou/
Vowels can come in three phonations, normal, breathy, and glottalized. Glottalized may become low/high tone.
Root/Syllable structure is something like CV(n)((C)V(n)), and almost all roots are monosyllabic and begin with a click or glottal stop. Clicks can only appear word initially.
/p~β~w t~ɾ k~ɣ~j q~ʁ ʔ/
/tsʰ~s kxʰ~x~h qχʰ~χ/
/k' q'/
/m n~ŋ~ɴ/
Voiceless unaspirated stops are voiced to approximants word medially, while affricates fricativize word medially. The /x/ is realized as /h/ before non-high vowels. The /β/ is realized as /w/ when followed by non-back vowels, and the /ɣ/ is realized as /j/ before non-front vowels.
/ʘ ǀ ǃ ǂ ǁ/
/ʘʰ ǀʰ ǃʰ ǂʰ ǁʰ/
/ʘ' ǀ' ǃ' ǂ' ǁ'/
/ⁿʘ ⁿǀ ⁿǃ ⁿǂ ⁿǁ/
/ⁿʘ' ⁿǀ' ⁿǃ' ⁿǂ' ⁿǁ'/
/qʘ qǀ qǃ qǂ qǁ/
/qʘʰ qǀʰ qǃʰ qǂʰ qǁʰ/
/qʘ' qǀ' qǃ' qǂ' qǁ'/
/a e i o u/
/ai au ei eu oi ou/
Vowels can come in three phonations, normal, breathy, and glottalized. Glottalized may become low/high tone.
Root/Syllable structure is something like CV(n)((C)V(n)), and almost all roots are monosyllabic and begin with a click or glottal stop. Clicks can only appear word initially.
Re: Random phonology/phonemic inventory thread
/m n
pʰ p tʰ t kʰ k ʔ
tθʰ tθ tsʰ ts
θ s h
l j
r
i iː ɨ u uː
e ẽ a ã aː ãː/ in high and low tones
<m n
p b t d k g q
ċ ż c z
ṡ z h
l y
r
i ii ë u u
e ę a ą aa ąą>
High tone with acute:
<í íí ë́/ê ú úú
é ę́ á ą́ áá ą́ą́>
pʰ p tʰ t kʰ k ʔ
tθʰ tθ tsʰ ts
θ s h
l j
r
i iː ɨ u uː
e ẽ a ã aː ãː/ in high and low tones
<m n
p b t d k g q
ċ ż c z
ṡ z h
l y
r
i ii ë u u
e ę a ą aa ąą>
High tone with acute:
<í íí ë́/ê ú úú
é ę́ á ą́ áá ą́ą́>
Re: Random phonology/phonemic inventory thread
/i i: e e: a a: o o:/<i ii e ee a aa o oo>
/p t̪ t͡ʃ k q ʔ ɓ t̪' t͡ʃ' k' f s̪ z̪ s̪’ ɬ ɬ' ʃ ʒ ʃ' x ɣ ʀ h m n̪ ɲ l j/<p t c k q ' b th ch kh f s z sh hl lh x j xh gh g r h m n ñ l y>
It's worth noting that the palatals /t͡ʃ t͡ʃ' ʃ ʒ ʃ' ɲ/ occasionally are written as <ti thi si zi shi ni> word finally, in contexts where there's an alternation due to a historic change of Ci--> C[PAL].
The syllable structure is (C)(C)V(C)(C).
Stress is always on the first syllables, and unstressed syllables induce some serious variation:
-The ejective consonants weaken to simply pharyngealized ones /ɓ t̪' t͡ʃ' k' s̪’ ɬ' ʃ'/-->[bˤ t̪ˤ t͡ʃˤ kˤ sˤ lˤ ʃˤ]
-ʀ becomes a fricative: /ʀ/-->[ʁ]
-The velar fricatives become epiglottal: /x ɣ/-->[ʜ ʢ]
-The short vowels /i e a u/ become the lax [ɪ ə æ ɔ]
With all these changes, especially the consonants, it's certainly possible to analyze it the other way around.
There are a few other allophonic changes:
-When one of [q ʀ ʁ ʜ ʢ] is the coda consonant closest to the vowel, values change regardless of stress or length: /i e a o/--> [ɪ æ ɒ ɔ]
-Stops are aspirated word initially.
-/ɲj/ is realized as [j̃] between vowels.
/p t̪ t͡ʃ k q ʔ ɓ t̪' t͡ʃ' k' f s̪ z̪ s̪’ ɬ ɬ' ʃ ʒ ʃ' x ɣ ʀ h m n̪ ɲ l j/<p t c k q ' b th ch kh f s z sh hl lh x j xh gh g r h m n ñ l y>
It's worth noting that the palatals /t͡ʃ t͡ʃ' ʃ ʒ ʃ' ɲ/ occasionally are written as <ti thi si zi shi ni> word finally, in contexts where there's an alternation due to a historic change of Ci--> C[PAL].
The syllable structure is (C)(C)V(C)(C).
Stress is always on the first syllables, and unstressed syllables induce some serious variation:
-The ejective consonants weaken to simply pharyngealized ones /ɓ t̪' t͡ʃ' k' s̪’ ɬ' ʃ'/-->[bˤ t̪ˤ t͡ʃˤ kˤ sˤ lˤ ʃˤ]
-ʀ becomes a fricative: /ʀ/-->[ʁ]
-The velar fricatives become epiglottal: /x ɣ/-->[ʜ ʢ]
-The short vowels /i e a u/ become the lax [ɪ ə æ ɔ]
With all these changes, especially the consonants, it's certainly possible to analyze it the other way around.
There are a few other allophonic changes:
-When one of [q ʀ ʁ ʜ ʢ] is the coda consonant closest to the vowel, values change regardless of stress or length: /i e a o/--> [ɪ æ ɒ ɔ]
-Stops are aspirated word initially.
-/ɲj/ is realized as [j̃] between vowels.
Re: Random phonology/phonemic inventory thread
A Proto-Lang:
/p pˤ pʼ b bˤ t tˤ tʼ d dˤ t͡ʃ t͡ʃˤ t͡ʃʼ d͡ʒ d͡ʒˤ k kʼ g q ɢ ʔ/
/m mˤ n nˤ ɲ ɲˤ/
/θ θˤ ð ðˤ s sˤ z zˤ ʃ ʃˤ ʒ ʒˤ x ɣ X ʁ ħ ʕ h ɦ/
/l/
/r/
/j w/
/ɨ ə ä/
/p pˤ pʼ b bˤ t tˤ tʼ d dˤ t͡ʃ t͡ʃˤ t͡ʃʼ d͡ʒ d͡ʒˤ k kʼ g q ɢ ʔ/
/m mˤ n nˤ ɲ ɲˤ/
/θ θˤ ð ðˤ s sˤ z zˤ ʃ ʃˤ ʒ ʒˤ x ɣ X ʁ ħ ʕ h ɦ/
/l/
/r/
/j w/
/ɨ ə ä/
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien
-JRR Tolkien
-
- sinic
- Posts: 401
- Joined: 21 Jul 2012 08:01
- Location: Buffalo, NY
Re: Random phonology/phonemic inventory thread
Concept for a SE Asian or eastern Himalayan-influenced phonology:
Vowels:
/i ɨ u/ <ii iu u>
/ɪ/ <i>
/ɜ ə ɚ/ <ee e er>
/æ ɑ/ <ae a>
/iə̯ uə̯/ <ie ue>
/eɪ̯ oɪ̯/ <eei oi>
/æɨ̯ ɑɪ̯/ <aeiu ai>
/ĩ ə̃ æ̃ ɑ̃/ <in en aen an>
/æ̃ɨ̯̃ ɑ̃ɪ̯̃/ <aeiun ain>
Every vowel without a glottal coda bears a contrastive tone:
high: á
low: a
rising: â
Consonants:
/m n ŋ/ <m n ng>
/p t t͜s k ʔ/ <p t ts k t> (glottal stop only appears in coda)
/ɓ ɗ/ <b d>
/v ʒ ɣ/ <v zh gh>
/ɸ θ ʃ x h/ <f th sh kh h>
/ɬ l r/ <lh l r>
/j w/ <y>
Possible initial clusters:
/ml pl ɓl/ <ml pl bl>
/θm θn θŋ θp θt θk θv θr/ <thm thn thng thp tht thk thv thr>
/xr xv/ <khr khv>
/θml θpl/ <thml thpl>
/rp rt rk rv/ <rp rt rk rv>
The final glottal stop may not appear after nasal or rhotic vowels.
Vowels:
/i ɨ u/ <ii iu u>
/ɪ/ <i>
/ɜ ə ɚ/ <ee e er>
/æ ɑ/ <ae a>
/iə̯ uə̯/ <ie ue>
/eɪ̯ oɪ̯/ <eei oi>
/æɨ̯ ɑɪ̯/ <aeiu ai>
/ĩ ə̃ æ̃ ɑ̃/ <in en aen an>
/æ̃ɨ̯̃ ɑ̃ɪ̯̃/ <aeiun ain>
Every vowel without a glottal coda bears a contrastive tone:
high: á
low: a
rising: â
Consonants:
/m n ŋ/ <m n ng>
/p t t͜s k ʔ/ <p t ts k t> (glottal stop only appears in coda)
/ɓ ɗ/ <b d>
/v ʒ ɣ/ <v zh gh>
/ɸ θ ʃ x h/ <f th sh kh h>
/ɬ l r/ <lh l r>
/j w/ <y>
Possible initial clusters:
/ml pl ɓl/ <ml pl bl>
/θm θn θŋ θp θt θk θv θr/ <thm thn thng thp tht thk thv thr>
/xr xv/ <khr khv>
/θml θpl/ <thml thpl>
/rp rt rk rv/ <rp rt rk rv>
The final glottal stop may not appear after nasal or rhotic vowels.
- Frislander
- mayan
- Posts: 2088
- Joined: 14 May 2016 18:47
- Location: The North
Re: Random phonology/phonemic inventory thread
/t̪~θ k~x ʔ/ <t k ʻ>
/s ɬ/ <s l>
/n̪ ɾ j/ <n r y>
/i iː o oː/ <i ii o oo>
/a aː aɪ̯ aːɪ̯/ <a aa ai aai>
Additionally each vowel may take a mid, high or low tone, with the latter two being indicated with acute and grave accents respectively. With the long vowels the accent is written on both vowels, except with the diphthongs where the tone is not written on the <i> (it is still written on the <a>(s) however).
Syllable structure is CV(C), where any consonant may appear in either onset or coda, bar /j/, which may not appear in the coda.
The oral stops are usually realised as fricatives, except when following another fricative (this includes the fricative allophones of the stops in coda position, such that, e.g. /t̪t̪/ is realised as [θt̪]). Additionally, all voiceless oral consonants are realised as voiced when they occur intervocalically, so e.g. a word like /t̪àɬkóːt̪i/ is realised as [θàɬkóːði].
/s ɬ/ <s l>
/n̪ ɾ j/ <n r y>
/i iː o oː/ <i ii o oo>
/a aː aɪ̯ aːɪ̯/ <a aa ai aai>
Additionally each vowel may take a mid, high or low tone, with the latter two being indicated with acute and grave accents respectively. With the long vowels the accent is written on both vowels, except with the diphthongs where the tone is not written on the <i> (it is still written on the <a>(s) however).
Syllable structure is CV(C), where any consonant may appear in either onset or coda, bar /j/, which may not appear in the coda.
The oral stops are usually realised as fricatives, except when following another fricative (this includes the fricative allophones of the stops in coda position, such that, e.g. /t̪t̪/ is realised as [θt̪]). Additionally, all voiceless oral consonants are realised as voiced when they occur intervocalically, so e.g. a word like /t̪àɬkóːt̪i/ is realised as [θàɬkóːði].