The "How do You Pronounce X" Thread

[Zaarin]

dénouement - [ˈdɛɪnʉmə̃nʔ] (spelling pronunciation, I'm not familiar with this word).

It's a literary term that refers to the "moment of revelation" at the climax or conclusion of a tragedy, which brings about the catharsis for the viewers and the catastrophe for the characters.

[finlay]

Three
Arthritis
Asthma
They
Brother

I pronounce them as closely as I can to RP (sorry America, I don't like [ɹ̠ʷ]): [ˈθɹ̠ʷiː] [ˈɑːθɹ̠ʷäɪ̯tɪs] [ˈæsθmə] [ðe̞ɪ̯] [bɹ̠ʷɒðə], but I still (it's 3 years since I discovered IPA etc.) can't distinguish 'three'-'free' and 'breathe'-'breve' and I want to ask you - how do you hear the difference between these two places of articulation.

if i may, the th in asthma is silent (in AmE it's usually /æzmə/ and in BrE it's usually /æsmə/), and brother should be /ʌ/, not /ɒ/ (a STRUT vowel, not LOT). as for three / free, learning a distinction that's not in your native language is always difficult and the key is practice. or you could just learn to speak a lower register of london english where they're not distinguished.

[jmcd]

AAVE can pronounce <th> as [f] (or [d] for the voiced version) as well.

[Travis B.]

I normally pronounce initial /ð/ as [t] except after a vowel (where then it is [ð]) or a nasal (where then it is [n̪~n]) except in higher-register speech.

(My daughter has the same pronunciation it seems, and she seems to have lost all the kinds of pronunciations younger kids have by this point.)

[ˈd̪ʲɛ.gɔ kɾuˑl̪]

Some Polish stuff (I'll give both (my) Polish and English orthographies and pronunciations):
bigos - [ˈbʲiˑ.ɢʌ̙s̪] [ˈbʲiːˑ.gɔs] /ˈbiː.gɒs/
kiełbasa/kielbasa - [kʲɛ̝wˈbäˑ.sä] [kʰʲjuˑˈbɑ̟ː.sɐ] or [kʰʲjɛ̈lʶ'bɑ̟ː.sɐ] /kjuːˈbɑː.sə/ or /kjɛlˈbɑː.sə/
pączki/paczki - [ˈpɔ̃n̠t̠͡s̠.kʲi] [ˈpʰɔ̃n̠ʲt͡ʃ.kʲi] or [ˈpʰat͡ʃ.kʲi] /ˈpɒnt͡ʃ.ki/ or /ˈpæt͡ʃ.ki/
pierogi - [pʲɛˈrʌ̙ˑ.gʲi] [pʰʲjəˈɹ̠ʷɔgʲ.i] /pjəˈrɒg.i/
rędzina/rendzina [rɛ̝̃ɲ̟ˈd͡ʑĩˑ.n̪ä] [ɹ̠ʷɛ̃n̠ʲˈd͡ʒĩːˑ.nɐ] /rɛnˈd͡ʒiː.nə/
ogonek - [ʌ̙ˈɢʌ̙̃ˑ.n̪ɛq] [ɔʊ̯ˈgɔn.ək] /oʊˈgɒn.ək/
złoty/zloty - [ˈz̪ʷwɔˑ.t̪ɪ̽] [ˈzʷwɔt.i] or [ˈzlɔt.i] /ˈzwɒt.i/ or /ˈzlɒt.i/
Also tell me if something like [ɛʊ̯] would be accepted for "kiełbasa".

[Sumelic]

Also tell me if something like [ɛʊ̯] would be accepted for "kiełbasa".

I think the only native English speakers who would use [ɛʊ̯] in "kielbasa" in English are people with generalized l-vocalization in syllable codas, so it would still be /ɛl/ phonemically. But to some extent, non-native speakers may be expected to use pronunciations closer to their native language for words from their native language.

[Jonlang]

The dictionary on my Mac* gives the pronunciation of hyperbole as /hʌɪˈpəːbəli/ which, to me, just sounds wrong. I've never heard it pronounced like that - definitely not with two schwas. The first schwa there, in everywhere I can recall hearing it, is pronounced like the er in British English her, but I'm not sure which IPA vowel that would be and I only have minute or two to write this post.

* Which is apparently the Oxford Dictionary of English.

[KathTheDragon]

It's ɜ: (but phonologically, it does pattern as a long counterpart of ə)

[Sumelic]

The dictionary on my Mac* gives the pronunciation of hyperbole as /hʌɪˈpəːbəli/ which, to me, just sounds wrong. I've never heard it pronounced like that - definitely not with two schwas. The first schwa there, in everywhere I can recall hearing it, is pronounced like the er in British English her, but I'm not sure which IPA vowel that would be and I only have minute or two to write this post.

* Which is apparently the Oxford Dictionary of English.

That's Clive Upton's scheme for transcribing British English. The length mark after the first schwa shows that this is supposed to be the NURSE vowel, not the lettER vowel. Another oddity of this scheme is the use of /ʌɪ/ for PRICE.

[Travis B.]

I personally do not like transcription systems that treat SSBE in a vacuum, like Clive's - like the whole nonsense of writing /ɛ/ as "/e/" despite that this transcription does not translate well to NAE...

[Sumelic]

I personally do not like transcription systems that treat SSBE in a vacuum, like Clive's - like the whole nonsense of writing /ɛ/ as "/e/" despite that this transcription does not translate well to NAE...

People write /e/ for [ɛ] in transcriptions for many languages... for example, Polish. It's just an example of the principle of using normal letters in preference to IPA-specific letters, like transcribing the rhotic phoneme as /r/ rather than something like /ɹ/.

[Travis B.]

I personally do not like transcription systems that treat SSBE in a vacuum, like Clive's - like the whole nonsense of writing /ɛ/ as "/e/" despite that this transcription does not translate well to NAE...

People write /e/ for [ɛ] in transcriptions for many languages... for example, Polish. It's just an example of the principle of using normal letters in preference to IPA-specific letters, like transcribing the rhotic phoneme as /r/ rather than something like /ɹ/.

The problem is when your language has dialects that contrast [ɛ] and [e] without a length contrast...

[Pole, the]

People write /e/ for [ɛ] in transcriptions for many languages... for example, Polish.

There's nothing more hilarious than native speakers of English complaining about Polish vowels being too similar to one another…

Also, typing /e/ for /ɛ/ in Polish in not a common convention; it is, however, common for its allophone appearing in proximity of palatal consonants (e.g. „dziesięć” is more [dʑeɕeɲtɕ] than anything else).

The problem is when your language has dialects that contrast [ɛ] and [e] without a length contrast...

I thought your language has dialects that constrast anything with anything…

[Jonlang]

Ah this explains a few things. I don't think I'm a fan of the Clive Upton system, I think I prefer literal transcriptions.

[Travis B.]

The problem is when your language has dialects that contrast [ɛ] and [e] without a length contrast...

I thought your language has dialects that constrast anything with anything…

I meant that there are varieties of English that contrast [ɛ] and [e], namely Scottish English varieties and some varieties of North American English (i.e. varieties which lack a length contrast and which have a monopthongal realization of /eɪ/).

[Sumelic]

Ah this explains a few things. I don't think I'm a fan of the Clive Upton system, I think I prefer literal transcriptions.

Well, any phonemic transcription will be somewhat removed from the literal phonetic realization of a word. My TRAP vowel varies greatly depending on if the following consonant is nasal or non-nasal, and my GOAT and FACE vowels have fairly distinct allophones before tautosyllabic /l/. Upton's transcriptions were actually meant to be more phonetically accurate than previous systems, though, not less.

The problem is when your language has dialects that contrast [ɛ] and [e] without a length contrast...

I thought your language has dialects that constrast anything with anything…

I meant that there are varieties of English that contrast [ɛ] and [e], namely Scottish English varieties and some varieties of North American English (i.e. varieties which lack a length contrast and which have a monopthongal realization of /eɪ/).

We get similar or worse issues with low and low-back vowels: depending on the dialect, /a/ /ɒ/ could represent TRAP LOT or LOT THOUGHT (and then there's /ɑ/ /æ/ /ɔ/). As long as you know the dialect and what system is being used, it's not a problem. And if you don't know the dialect, you'll run into other problems anyway, like unexpected non-rhoticity or intrusive "r".

[Travis B.]

The thing is I would prefer phonemic transcriptions that cover as wide of a range of dialects as possible. E.g. one phonemic transcription system to cover all rhotic North American English varieties. I could make a phonemic transcription system just for my own dialect, but it would be essentially useless to people who speak other dialects, giving them no indication of correspondences between other dialects and my own dialect.

[Sumelic]

The thing is I would prefer phonemic transcriptions that cover as wide of a range of dialects as possible. E.g. one phonemic transcription system to cover all rhotic North American English varieties. I could make a phonemic transcription system just for my own dialect, but it would be essentially useless to people who speak other dialects, giving them no indication of correspondences between other dialects and my own dialect.

I understand that preference, but if we're going with actually phonemic rather than diaphonemic transcriptions, I don't think it's really possible to have a united transcription of rhotic and non-rhotic accents. There are too many phonemic differences between them. And as far as I know, there is no non-rhotic accent where it would make sense to use /e/ for FACE, so there is no possible confusion with using /e/ for DRESS in transcription of non-rhotic accents.

I mean, I also prefer /ɛ/ for DRESS, I just don't think /e/ is that bad a choice.

[Travis B.]

The thing is I would prefer phonemic transcriptions that cover as wide of a range of dialects as possible. E.g. one phonemic transcription system to cover all rhotic North American English varieties. I could make a phonemic transcription system just for my own dialect, but it would be essentially useless to people who speak other dialects, giving them no indication of correspondences between other dialects and my own dialect.

I understand that preference, but if we're going with actually phonemic rather than diaphonemic transcriptions, I don't think it's really possible to have a united transcription of rhotic and non-rhotic accents. There are too many phonemic differences between them. And as far as I know, there is no non-rhotic accent where it would make sense to use /e/ for FACE, so there is no possible confusion with using /e/ for DRESS in transcription of non-rhotic accents.

I mean, I also prefer /ɛ/ for DRESS, I just don't think /e/ is that bad a choice.

Even if it is not possible to unify certain dialect groups' transcription, I would still be for following consistent standards across transcription systems, e.g. always using /ɛ/ for DRESS no matter how it is realized.

[Vlürch]

hyperbole as /hʌɪˈpəːbəli/ which, to me, just sounds wrong.

My pronunciation of English tends to be pretty bad, so maybe this is as well, but I've always pronounced "hyperbole" something like [hɑi̯pʰəɹboʊ̯ɫ̪~hɑi̯pʰɛɾbɔu̯ɫ̪] at least in my head. I know the English /r/ is never supposed to be [ɾ], except maybe in some Scottish or Irish accents, but it feels like the most natural pronunciation for me; obviously that's because my first language is Finnish, where /r/ is always [r~ɾ], though, but well. The "bole" part being anything except [boʊ̯ɫ̪~bɔu̯ɫ̪] makes no sense to me and doesn't only sound wrong, but feels wrong.

[Zaarin]

hyperbole as /hʌɪˈpəːbəli/ which, to me, just sounds wrong.

My pronunciation of English tends to be pretty bad, so maybe this is as well, but I've always pronounced "hyperbole" something like [hɑi̯pʰəɹboʊ̯ɫ̪~hɑi̯pʰɛɾbɔu̯ɫ̪] at least in my head. I know the English /r/ is never supposed to be [ɾ], except maybe in some Scottish or Irish accents, but it feels like the most natural pronunciation for me; obviously that's because my first language is Finnish, where /r/ is always [r~ɾ], though, but well. The "bole" part being anything except [boʊ̯ɫ̪~bɔu̯ɫ̪] makes no sense to me and doesn't only sound wrong, but feels wrong.

Hyperbole is a Greek loanward, hence the finale /i/. Cf. names like Chloe /kloʊ.i/ or Zoe /zoʊ.i/.

[vokzhen]

hyperbole as /hʌɪˈpəːbəli/ which, to me, just sounds wrong.

The "bole" part being anything except [boʊ̯ɫ̪~bɔu̯ɫ̪] makes no sense to me and doesn't only sound wrong, but feels wrong.

I have bad news for you concerning epitome and syncope. Pronouncing epitome as /'EpIto:m/ instead of /@'pIt@mi/ is a well-known, widely-laughed-at spelling pronunciation. In my experience it lacks the vitriol associated with other "errors," you're likely to get corrected but less harshly or patronizingly than a lot of other "errors." I'm not sure I've ever heard a native speaker use it for hyperbole except ironically, though. I definitely thought that's how syncope was supposed to be pronounced when I got into linguistics, and just learned in the last few years it's one of the ones that has a final <-e> /i/.

[finlay]

hyperbole as /hʌɪˈpəːbəli/ which, to me, just sounds wrong.

My pronunciation of English tends to be pretty bad, so maybe this is as well, but I've always pronounced "hyperbole" something like [hɑi̯pʰəɹboʊ̯ɫ̪~hɑi̯pʰɛɾbɔu̯ɫ̪] at least in my head. I know the English /r/ is never supposed to be [ɾ], except maybe in some Scottish or Irish accents, but it feels like the most natural pronunciation for me; obviously that's because my first language is Finnish, where /r/ is always [r~ɾ], though, but well. The "bole" part being anything except [boʊ̯ɫ̪~bɔu̯ɫ̪] makes no sense to me and doesn't only sound wrong, but feels wrong.

The /i/ at the end isn't being contested here - people are nitpicking over whether it should be /əː/ or /ɜː/ in the middle, which I rather think misses the point of the IPA and of phonemic transcriptions. Basically as long as we accept that we're transcribing a non-rhotic dialect, either way this is a central vowel and I personally can't hear the difference between ə and ɜ.

But yeah sorry sweetheart, it's /bəli/.

[Pole, the]

Inspired by this comment: do you pronounce “Prussian” as [pɹʌsiən] or [pɹʌʃən]?

[Qxentio]

złoty/zloty - [ˈz̪ʷwɔˑ.t̪ɪ̽] [ˈzʷwɔt.i] or [ˈzlɔt.i] /ˈzwɒt.i/ or /ˈzlɒt.i/

I am currently experiencing a strong case of Mandela Effect. I remember learning the word złóty, pronounced zwutɨ or something along those lines, before travelling to Poland for the first time several years ago.
Please tell me that there is a form of the word or a dialectal variation (Małopolska?) with /u/ in the first syllable.