Help talk:IPA/Azerbaijani

Discussions before the split can be found at Help talk:IPA/Oghuz languages/Archive 1.

Contents may be incomplete or incorrect

According to the wiktionary page for almaq, its IPA code is /ɑɫˈmɑχ/, which contains two consonants that are missing from this page: /ɫ/ and /χ/. At least for the latter, this seems to affect many verbs, so I'm surprised not to find it here.

Furthermore, here it says g is /ɟ/ while q is /g/, yet the page for getmək contradicts this, as g is /g/ (getmək: [ɡetˈmæk], [ɡetˈmæj]).

Personally, I don't know which of these is correct, but I believe I've seen /χ/ used in various sources, and g as /g/ also seems to be correct (at least sometimes). Maybe someone knowledgeable could help out. --Bfx0 (talk) 15:13, 4 August 2020 (UTC)

The transcription used here is consistent with that used at Azerbaijani language#Phonology. I don't see explicit citations there for those representations, but it may be a matter of convention rather than phonetic accuracy. We'd have to see what sources say. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 04:42, 6 August 2020 (UTC)

The page (Azerbaijani language) mentions the following:

In colloquial speech, /x/ (but not intramorpheme [x] transformed from /g/) is usually pronounced as [χ]

However, it contradicts the claim that almaq is pronounced /ɑɫˈmɑχ/, as that would fall under a /x/ transformed from /g/. Also, it is seemingly unsourced. Samiollah1357 (talk) 04:33, 9 November 2023 (UTC)

Unless I'm terribly mistaken (I'm neither a linguist nor Azerbaijani) I've also noticed that the example for /ɾ/ seems incorrect? I don't think "atom" matches the sound there! When compared to the IPA on Karabakh. 62.6.35.4 (talk) 13:56, 25 September 2024 (UTC)

the /ɪ/ sound in South Azerbaijani

@Aeusoes1: /ɪ/ is a vowel used as an allophone of both /i/ and /ɯ/ when they come at the end of a word or when they become the last vowel of a word regardless of being the actual last letter. Examples to this can be seen in 'Mahnı' pronounced /mɑhnɯ/ in the Republic of Azerbaijan but pronounced /mɑhnɪ/ in Iran with a sound that when paid perfect attention is between an /i/ and an /e/ which is exactly what /ɪ/ is. This is especially common in the Tabriz accent. Zxnith84 (talk) 15:52, 28 August 2020 (UTC)

Do you have a source? — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 22:59, 28 August 2020 (UTC)

This is only seen in colloquial speech and as South Azerbaijani is not standardized and there is no official written version, there is no written record of this Zxnith84 (talk) 08:26, 30 August 2020 (UTC)

Luckily, that's not quite how linguistic research works and, if this really does exist in the speech of South Azerbaijani speakers, then there may indeed be linguistic sources that document it. Until such sources are found, we can't really incorporate that feature in our transcriptions or include that claim in this IPA guide or elsewhere in Wikipedia articles. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 14:27, 30 August 2020 (UTC)

[1] gives [ɘ] as the typical South Azerbaijani realization of /ɯ/ (they don't mention /i/ being pronounced that way, they give [i] for it), thus [mɑhnɘ]. This is backer and lower than [ɪ] (but still similar, compare Polish and Czech ⟨y⟩) and is not limited to the word-final position. Sol505000 (talk) 12:03, 16 March 2023 (UTC)

Baku

The pronunciation of Bakı (Baku) in File:Az-baku.ogg is clearly not [bɑˈcɯ] but [bɑˈtʃɯ] 'Baçı'. If it's non-standard, it needs to be replaced or renamed (and presumably also removed from the article Baku). If it's not, the guide needs to be updated. Azerbaijani language#Phonology provides no information on this phenomenon. My suspicion is that those speakers who have [ts dz] for standard [tʃ dʒ] have [tʃ dʒ] for standard [c ɟ] (compare Old English [itʃ], with a postalveolar, with Old/Modern Dutch [ɪk], with a dorsal (phonetically pre-velar, but that's allophonic in Dutch), both meaning 'I').

Alternatively, [c ɟ] simply merge with [tʃ dʒ] for some speakers. I don't know. Either way, this needs to be cleared up. Sol505000 (talk) 11:56, 16 March 2023 (UTC)

Per [2], [bɑˈtʃɯ] could be a Tabriz pronunciation. Sol505000 (talk) 19:20, 20 March 2023 (UTC)