/pʰ tʰ~(ʔ) ʧʰ kʰ~x/
/b d ʤ g/
/m n/
/f v s z (θ ð) s z ʃ ʒ h/
/ɾ l w j/
I rarely have /ʔ/ for /t/, but it does creep in sometimes.
/x/ is an intervocalic and word-final allophone of /kʰ/.
/ð/ is nearly always /d/ for me in casual speech. /θ/ can tend towards dental /t/.
/ɾ/ may sometimes be /ɹ/ but I tend to have /ɾ/.
[ŋ] only occurs in /ŋg/.
/l/ tends to be less velarised than in some varieties.
Long vowels:
/i: ʉː/
/e:~ɪ: ɔ:/
/a:~ɑ:/
I tend towards /ɪ:/.
Short vowels:
/ɪ ʊ/
/ɛ ɑ/
/a ǝ/
/a:i a:u ɛi ɔu ɔi/
I wonder if anyone can guess my accent based on that.
Davush wrote:I wonder if anyone can guess my accent based on that.
Is that Liverpool?
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
h/m n ŋ/
/p pː t tː k kː/
/p͡f t͡s t͡ʃ/
/f fː s sː x h xː~hː/
/ʋ l ɾ j/
Gemination may not be phonemic in all consonants in all environments. Initial k: is realized as aspirated and contrasts with marginal [k:] (found mostly in loans from romance words starting with a tenuis plosive, or as reduction of /tk/). /x:/ is often realized as [h:], especially intervocally.
/i y u/
/ɪ ʏ ʊ/
/e ø o/
/ə/
/ɛ œ ɔ/
/a/
/ɪɐ̯ ʏɐ̯ ʊɐ̯/
/(eɪ̯) ɛɪ̯ ɔɪ̯/
/aɪ̯ aʊ̯/
Length may be contrastive in a number of vowels, but the large inventory makes true minimal pairs statistically unlikely. I believe vowel length can be determined by the environment but haven't been able to prove it. In final unstressed position, only two vowels appear: [ɪ ɐ], the latter being a common allophone of the schwa. [eɪ] is marginal, occurring mostly in loanwords (baby) and as a possible allophone of /egɪ/.
At kveldi skal dag lęyfa,
Konu es bręnnd es,
Mæki es ręyndr es,
Męy es gefin es,
Ís es yfir kømr,
Ǫl es drukkit es.
Unlike my peers, my voiceless phonemes are nearly unaspirated, which I attribute to growing up around Indian parents.
This is strictly inventory: allophones are not included if they're represented by another phoneme (ie. no /t~ʃ/, just /t ʃ/
/p t~ʔ~ɾ kʰ b d~ɾ g~ɣ/
/f v θ~ð (ð) h~χ/
/l~ɫ~ɰ ɹ~ɻ~ɾ j~j̥ w(~ʍ)*/
Stressed
/æ ɛ ɪ i a(~ɒ)* ʉ ɯ̽ eɪ̯(~eː)* ʌʊ̯ oɪ̯ aɪ̯ æʊ̯/
Pre-R
/ɛ ɪ i ə ɝ ɑ ɒ ɔ o ʉ e~eɪ̯(~æ)* eɪ̯ə oɪ̯ oɪ̯ə aɪ̯ə æʊ̯ə/
Pre-L (Vocalization)
/æɰ ɛɰ ɪɰ iɰ eɰ ʉʊ̯ɰ əɰ ɑɰ(~aɰ)* ɔː oː ɯ̽ː ʉː aɪ̯ɯ oɪ̯ɯ/
Pre-L (No vocalization)
/æ ɛ e ɪ i ʉ ə o aɪ̯ ɯ oɪ̯/
Unstressed
/ə ɨ/
/p b t d t͡ʃ d͡ʒ k g/
/f v θ ð s z ʃ ʒ h/
/m n ŋ/
/w j ɹʷ l/
Voiceless stops and fricatives are aspirated in the onset except after /s/ and in the onset of an unstressed syllable other than the first one. Aspirated stops tend to be somewhat affricated.
/t/ and /d/ both become [ɾ] intervocallically when not in the onset of a stressed syllable or the first syllable of a word, and /d/ can also become [ɾ] word-finally.
/t/ becomes [ʔ] in the coda.
/h/ is sometimes deleted word-initially in casual speech.
/θ/ and /ð/ are interdental, and tend to be realized as [t̪͆] and [d̪͆], respectively, utterance-initially and after nasals.
Light /l/ is interdental. In the coda, /l/ is realized as [ɤ̯ˤ] after unrounded vowels and [o̯ˤ] after rounded vowels (e.g. <still> [stɪɤ̯ˤ], <pool> [pʰuo̯ˤ]), with the same amount of rounding as the preceding vowel. /oʊ̯l/ is realized as [oˤ] (e.g. <old> [oˤd]). When a word-final or morpheme-final /l/ is followed by a vowel-initial word, it becomes [lˤ] (e.g. <taller> [ˈtʰɒlˤɚ]).
I'm still not entirely sure whether I have [oʊ̯] or [əʊ̯], or whether I have [ɐ] or [ʌ].
Apparently it's [ɝ] in stressed syllables and [ɚ] in unstressed syllables, although I personally can't hear the difference.
/aɪ̯/ becomes [ɐɪ̯] before unvoiced consonants and sometimes before flapped /d/ [ɾ], as well as in a few unusual cases before other voiced consonants: <onion> [ˈɐɪ̯njən], <cyborg> [ˈsɐɪ̯bɔɚ̯g].
/æ/ becomes [eə̯] before /n/ and /m/ and [æɪ̯~ɛɪ̯] before /ŋ/.
/ɝ/ is bunched, just like /ɹ/ above (but not rounded).
High German (albeit influenced by certain Swabian and Saxonian dialects)
/b̥ pʰ d̥ tʰ g̥ kʰ/
/͡pf t͡s t͡ʃ d͡ʒ/
/m n ŋ/
/f v s z ʃʷ ʒʷ x h/
/l j ʁ/
Like in most areas of Southern Germany, the difference between voiced and unvoiced stops is rather a lenis-fortis distinction in my idiolect. Unlike the speakers around me, I still aspirate my fortis consonants.
Unlike most speakers around me, I retain the voiced phonemes /z/, /ʒʷ/ and /d͡ʒ/ (the latter two only occurring in loanwords). On the other hand, I lack [ɱ] as an allophone of /m/ in words like fünf.
Born and raised by Saxonians, I had a difficult time with /x/'s allophone [ç], constantly pronouncing it [ʃ] instead. I hypercorrected that a little, resulting in forms like Wäsche: [vɛçə] instead of [vɛʃə]
Vowels:
/a aː/
/eː ɛ øː œ oː ɔ ə/
/iː ɪ yː ʏ uː ʊ/
The pronunciation of my vowels almost exactly follows what has been "taught" for Standard German. As many speakers of standard German, I lack /ɛː/ (I pronounce it as /eː/ instead, making Bären and Beeren homophonous). However, I almost never pronounce the schwa, with the effect that I'm sometimes hard to understand (although not as hard as dialect speakers). I don't feature Swabian nasal vowels, and whenever I was trying, I was constantly told that the sound "weird".
Wipe the glass. This is the usual way to start, even in the days, day and night, only a happy one.
Consonants:
/m n ŋ/
/pʰ~p~p̚ b~b̚ tʰ~t~ɾ~t̚ d~ɾ~d̚ kʰ~k~k̚~kʼ g~g̚/
/t͡ʃʰ~t͡ʃ d͡ʒ/
/f v θ ð s z ʃ ʒ h/
/w ɹ j/
/ɫ/
- among plosives, aspirated forms are used initially and, apart from [tʰ], medially unless preceded by /s/ or /ʃ/
- unreleased forms used syllable finally unless emphasised in which case they may be aspirated, or in the case of /k/ ejected.
- [ɾ] appears between vowels after stress or even prestress when followed by a morpheme boundary
- /dj tj sj zj/ of other dialects merge with /t͡ʃʰ~t͡ʃ d͡ʒ ʃ ʒ/ and /stj/ may further become [ʃt͡ʃ].
- no ɱ, /mf/ becomes [mpf] !?
- Superscript characters indicate the the epenthetic /ɹ/ or /w/ that arises when followed by a vowel to prevent hiatus. Vowels that lack this are never followed by another vowel.
- /ɪː/ can become [ɪɐ̯] word finally.
- /ə/ becomes [ɐ] word finally.
- in unstressed syllables, /ə/ and /ɪ/ are in almost complementary distribution with /ɪ/ appearing before postalveolar and velar consonants and /ə/ elsewhere although there are a few words with /ə/ before velars.
- the CURE vowel is not distinct, either merging with /oːʳ/ as in "poor", "sure" or else bisyllabic /ʉːʷəʳ / as in "cure", "tour",or simply /ʉːʷ/ as in "Europe", "fluoro"
- Superscript characters show epenthetic consonants that appear before vowels to prevent hiatus.
- All diphthongs display allophonic length changes, short before voiceless consonants, long before voiced. (This seems to be blocked by phonemic length distinctions in the monophthongs.)
- /ɪi̯/ shortens to [ɪ̆i̯] when unstressed.
- /æʊ̯/ unrounds to [æɯ̯̽] before most consonants and may not be distinguished from /æː/ before /ɫ/.
- /ɐʉ̯/ backs to [ɔʊ̯] or possibly [ʌʊ̯] before /ɫ/.
Last edited by Imralu on 21 Mar 2017 09:19, edited 3 times in total.
It really depends on how good I'm trying to speak so I have either a really German accent or quite a faked British accent. I ordered the vowels according to how they would be pronounced in GAE (monophthongs, diphthongs, r-coloured vowels, unstressed vowels).
Not trying at all:
/ɛ(ː)1 äː ɔ o̞(ː)ɐ̯ ɪ iː ɛ ä ʊ uː/ trap bath/palm/start lot/cloth thought/north/force kit fleece dress strut foot goose
/ɛɪ̯ äɪ̯ ɔɪ̯ o̞ʊ̯ äʊ̯/ face price choice goat mouth
/œ(ː)ɐ̯ i(ː)ɐ̯ e̞(ː)ɐ̯ ɔɐ̯~u(ː)ɐ̯2/ nurse near square cure
/ɐ3 i/ 1 The lengthening is only there to differentiate /æ/ [ɛː] and /ɛ/ [ɛ], I'd say 2 I somehow pronounce cure a [kʰjɔɐ̯~k͡çɔɐ̯] but sure as /ʃʷu(ː)ɐ̯/ instead. I don't know why to be honest. 3 This might be due to German's pronunciations of Komma and Bäcker for example.
/m n ŋ/
/pʰ~p b tʰ~t~d d t͡ʃ1 d͡ʒ1 kʰ~k g/
/f v θ2 ð2 s z ʃʷ ʒʷ h/
/l3 ɹ̠ʷ j w/
1 These may deaffricate depending on how German I speak and aren't labialised compared to the post-alvealor fricatives. 2 Again, depending on how German I speak the dental fricatives either stay dental or they merge with /s/ and /z/. 3 /l/ is always /l/ and never velarizes in the coda.
Allophones:
- The distribution of aspirated and unaspirated voiceless stops is just like in English. At the end of a word the stops aren't as aspirated as I normally aspirate them in German.
- Voiced consonants devoice at the end of words.
- /j/ after /t/ and /k/ often turns into [ç].
- /t/ is realised as [d] intervocally.
- /tɹ̠ʷ/ and /dɹ̠ʷ/ are realised as /t͡ʃʷ~t͡ʃw/ and /d͡ʒʷ~d͡ʒw/.
1 /ɒ/ is incredibly hard for me to pronounce and I noticed that I constrict something down my throat which is most likely the pharynx or the epiglottis.
/m n ŋ/
/pʰ~p b tʰ~t d t͡ʃ1 d͡ʒ1 kʰ~t g/
/f v θ2 ð2 s z ʃʷ ʒʷ h/
/l ɹ̠ʷ j w/
1 Compared to the simple post-alvealor fricates, the affricatives aren't labialised. 2 Depending on how non-standard I want to speak, I pronounce the dental fricatives as /f/ and /v/.
Allophones:
- The distribution of aspirated and unaspirated voiceless stops is just like in English. At the end of a word the stops aren't as aspirated as I normally aspirate them in German.
- /t/ is realised as a glottal intervocally and at the end of words, which is a bad attempt at trying to pronounce it as [ʔ͡t]
- /tɹ̠ʷ/ and /dɹ̠ʷ/ are realised as /t͡ʃʷ~t͡ʃw/ and /d͡ʒʷ~d͡ʒw/.
- Coda /l/ is vocalises into /o̯/ and merges with the vowel(s) in front of it. Diphthongs change their last part into /o̯/ as well, e.g. /ɪ̟i̯l/ is [ɪ̟o̯] and /e̞ɪ̯l/ is [e̞o̯]. At the end of words /əl/ is simply realised as /o/.
- /h/ and /k/ are pronounced as /ħ/ and /k͡ħ/ before /ɒˤ/, so cot is [k͡ħɒˤʔ] which is pretty fucked up in my opinion.