Talk:IPA vowel chart with audio

This article is entitled "IPA vowel chart with audio", and the chart in the article is labelled "IPA vowel chart". The lead paragraph says "Below is a chart depicting the vowels of the IPA". The article needs to make it clear that the chart is NOT the vowel chart of the International Phonetic Association. The authentic IPA vowel chart may be examined at [2], and it can be seen that the Wikipedia chart has had half a dozen vowels added, for reasons that are not explained.

Some of the recorded vowel qualities bear little or no resemblance to the vowel qualities familiar to most phoneticians trained within the IPA tradition. RoachPeter (talk) 19:28, 25 June 2014 (UTC)

Yes, a more complete perfect title would be like "Vowel chart based on the IPA vowel chart with additions and with audio". Or: "Vowel chart" (without any IPA claim).

Actually, your IPA link says it is from 2005. (2005!). Nine years without change - and this is internet. I propose to leave IPA authority behind. They have made not a single step development after 2005 (check the wiki consonants tables or this). Both Unicode and the missionaries have evolved the phonetic alphabet greatly in between.

You second remark, about vowel sound quality: this is wikipedia. You and everyone ares invited to propose or upload improvements. -DePiep (talk) 21:05, 25 June 2014 (UTC)

A secondary reply. I just saw your background in phonetics, so I think a clarification might help (after maybe sounding cynical) ;-). I am not familiar with phonetics at all, and so not with IPA. It is, three years ago [3] I started building the IPA charts into tables for HTML and wiki. Afterwards I made those "with audio" in scheme. They are still alive, so I think I did something good. (Had many talks with User:Kwami).

My point is that the IPA standard is not clear, and not up to date. Grey areas all over. Do I really have to buy a book to learn the standard? A 2005 book? No thanks. And if IPA does not respond to SIL or Unicode (that's real life), the IPA is in the waiting room for me. Meanwhile, we at wikipedia provide more accurate, complete and correct diagrams. With audio. -DePiep (talk) 22:58, 25 June 2014 (UTC)

DePiep, I don’t want to disagree with you that the International Phonetic Association and its Alphabet are far from perfect, and that some updating is needed. I also agree that it would be useful to have some or all of the material of the IPA Handbook available on-line rather than in a printed book. My main point is that it would remove possible confusion if the article pointed out where your chart is different from the IPA’s – this clarification is done effectively on the WP’s article on the IPA, using grey print for innovations, and an explanatory note. Apart from that, the audio-active chart that you have produced is a very useful tool, and my criticisms of that are ones of detail rather than of principle.

On some wider points, and in defence of the IPA, I should say that I was involved in the organization of the 1989 Kiel conference of the IPA when I was Secretary of the IPA - see for example Ladefoged, P. and Roach, P.J. `The future of the I.P.A.', Journal of the International Phonetic Association, vol.16, pp. 5-14. (1987) - and you can probably imagine how difficult it is to assemble a world-wide group of academics, put them in one place and get them to agree on a complicated set of proposals. When you have got something agreed, it’s natural to want to preserve that consensus, even if you are unhappy (as I am) with some of the decisions made. Symbols can be added to or removed from the IPA Chart by the Association’s Council (and ultimately by the whole membership) if a member can establish, usually by means of publishing a paper in the Journal of the IPA, the need for such a change. I don’t know if any such changes are in the pipeline, as I am no longer active in IPA affairs.

Finally, I do realize that if I comment adversely on the quality of some of the recorded sounds, it is a fair response for you to say that I can always offer to do better myself. I am thinking about this (after all, I did teach these things for over 35 years). As someone pointed out in this talk page, it’s best if all the vowels are recorded by the same person; ideally they should all be on the same pitch and with the same length and loudness. Even better would be two sets of recordings by different voices, if possible one female and one male. There are already sets of expert recordings of these vowels in existence, but as I am coming to understand WP rules about submitting audio files I can see that probably none of these would be acceptable (or available) for a WP article. I could record the vowels of the official IPA chart, but I would struggle to do the extra ones that have been added. RoachPeter (talk) 15:28, 26 June 2014 (UTC)

Thanks for these thoughtful replies. I myself am have an engineering background, so when a table gets diffusive, that makes me, uh, restless (as you could read). Then you having to gather & make conclude academics, in Kiel -- wow, respect ;-).

Your primal point being that the title is incorrect, I do admit right now. As I recall from my earlier talks with user:kwami, we need a graphic notation (like a grey tone) to denote "not official IPA". On the other hand, that "not offical IPA, but very useful" is a status phonetic alphabet needs (is why I mentioned Unicode and SIL). I can reply more in substance later on. -DePiep (talk) 19:58, 26 June 2014 (UTC)

I had the title move, yesterday, reverted. The new name did not reflect this discussion, and did not take titling guidelines in account. Phonetically savvy, here it was said that "IPA" might not be correct in the title (because there are more non-IPA sounds in there). However, I do not think that is major enough to be mentioned in the title. A footnote could do that. Then, the bad title said "examples". First, if I understand IPA well, there can only be one "example" for each sound. So that word is superfluous (so, bad in itself). Second, a WP:TITLE does not have to describe the topic. That is what the article itself is for, especially the lede. It just needs to say what it is. And it does, now. I disagree with this topic's sectiontitle, saying it is "misleading" (imprecise? incomplete? not-common?). I agree that a better title might be possible, within the title guidelines. -DePiep (talk) 08:05, 6 February 2015 (UTC)

I recommend calling the template

{{Extended IPA vowel chart with audio}}

My reasoning: This is not an ecyclopedia article, whose name is a heading under which the information is listed for consultation by a reader. It is a technical tool, whose name is a specification of what is wanted for use by a writer. I would expect, as a writer, that if I specified "IP vowel chart with audio" I would get something different from what this template provides. The content provided to writers by this template is related to encyclopedia article Extensions to the International Phonetic Alphabet. --Pi zero (talk) 12:27, 6 February 2015 (UTC)

"Extended" already has an association with IPA, so better not use it to say "outside of IPA'. I'll have to re-read this whole thread carefully, Roach makes good points. Improving the article text can be done always. Maybe simply merge into the vowel article? -DePiep (talk) 16:11, 6 February 2015 (UTC)

Thank you all for creating this page! It helps me as an English teacher and will help many English students. DBlomgren (talk) 04:22, 17 January 2017 (UTC)

Maybe I have misunderstand something, but Ä here doesn't sound anything like Ä in Finnish language (I'm native Finnish speaker). Is it supposed? 91.158.165.178 17:51, 5 February 2015 (UTC) — Preceding unsigned comment added by 91.158.165.178 (talk)

ä (lowercase) is not tied to any language. It may also not sound like the German "ä" in "während" either. It is a symbol for a well-defined vocal sound, as is every symbol int the IPA alphabet. It is always the same sound. Someone who can read & speak IPA symbols, can pronounce that word correctly -- without knowing its meaning of course (every opera singer can sing perfectly Czech this way!).

And from the opposite side: the Finnish "ä" has a IPA symbol too. Help:IPA for Estonian and Finnish suggests it is IPA symbol æ as in pöytä. Check the audio. More Finnish examples [4]. See IPA 'translations' in Arvo Pärt, Pesäpallo, Matti Ijäs. Hyva matka. -DePiep (talk) 19:58, 5 February 2015 (UTC)

Got it! Thanks to both of you! 91.158.165.178 (talk) 20:56, 5 February 2015 (UTC)

You lucky Finnish people. One letter = one sound! Compare that to English (they really need IPA. Their "a" in cat, back, bag). And also, the stressed syllables are visible in a word: every first, third, fifth, seventh, ninth, eleventh, thirteenth, fifteenth, seventeenth, nineteenth, twenty-first syllable. Lucky people. -DePiep (talk) 21:06, 6 February 2015 (UTC)

Many of these samples are recorded with an extreme amount of creaky voice, which I find makes them very difficult to distinguish. Can we change such recordings to more neutral ones? timothymh (talk) 07:31, 22 November 2016 (UTC)

There is currently no license information for the audio files. Would these be therefore covered under the default Creative Commons Attribution-ShareAlike License?kdelwat (talk) 06:35, 16 December 2016 (UTC)

Guys, I'm sorry, I don't want to be negative, but... How were these sound recordings made? Most of them sound wrong. I'm dead serious when I say that I could make a better recording of all of these vowels. At least then the pitch, volume and length would be consistent, and there wouldn't be creaky voice. — Preceding unsigned comment added by 84.2.59.196 (talk) 05:07, 9 June 2017 (UTC)

Same guy. The recordings made by Daniel Jones are available online. It would be a very good idea to start from these, as they're definitely acoustically equidistant as far as the primary cardinals are concerned and the secondary cardinals have the same tongue-position as the primary counterparts. In the current recording for [y], the speaker articulates the sound of French 'lune' (moon). That vowel is way more open and back than the cardinal i/y. If you take away the lip-rounding from it, it becomes the Engish half-open [ɪ]. — Preceding unsigned comment added by 62.165.216.137 (talk) 07:47, 10 June 2017 (UTC)

Your example is very wrong. The [y] sound record is definitely not the french "u" (I'm native standard French speaker and used to phonetics). The french sound is much more closed and front than this sound, thus much closer to the cardinal [y]. On the other hand, I agree that the recorded sound is obviously very far from the cardinal [y] (far too open and back). denis 'spir' (talk) 15:59, 16 September 2019 (UTC)

There are definitely a few bad ones. For instance what it claims is [ʌ] = [ə], and [ɐ] = [ɑ]. 68.228.246.127 (talk) 15:28, 27 August 2018 (UTC)

Using the formants F1, f2 and f3, It would be clearer for users to add a 3d interactive plot object of the IPA vowels. I have read R software could make it, but I am afraid I have never used it. --Backinstadiums (talk) 18:49, 13 December 2017 (UTC)

Sorry if this is not the place to say so, but I just want to congratulate and to thank whoever took the trouble to do this. It's a great job, and it's extremely useful. Thank you very much! --Eduarodi (talk) 18:07, 8 May 2018 (UTC)

First, thank you very much for this page, a extremely useful idea indeed !

But: Where do these sounds come from? Like others (see comments above) I find that many of them obviously are very far from correct. How were they produced? Why is their source not even mentionned?! denis 'spir' (talk) 15:59, 16 September 2019 (UTC)

The sounds for the vowel and pulmonic consonant pages come from Peter Isotalo, as mentioned here : http://www.ipachart.com. These files can be found at several places on the internet. Peter Isotalo is an amateur linguist and a regular contributor to Wikipedia. He presents himself here : User:Peter_Isotalo

He did a good work (thank you !), although I must say that his voice quality is a bit too guttural. Another amateur, Vincent Ramos, also did good recordings (merci également à lui) : http://sivanataraja.free.fr/api/. The best recordings would be those by Peter Lagefold, because he is an authority in the world of phonetics. I am not sure that his sounds could be used here because of copyrights. http://www.phonetics.ucla.edu/course/chapter1/chapter1.html. UBC Video Speech also made videos of API Sounds : https://enunciate.arts.ubc.ca/linguistics/world-sounds.

I do not know additional references with audio (do you ?), but those already mentioned could be added to a link section. Also, Peter Isotalo should probably be acknowledged on this page.

Pinging : User:Peter_Isotalo User:Denispir User:DePiep User:RoachPeter User:Vincent_Ramos https://www.search.com.vn/wiki/fr/Utilisateur:Vincent_Ramos

LinguisticStudent (talk) 10:15, 1 October 2019 (UTC)

Not all of them are Peter Isotalo's. Click the blue 'i' icon to see the attribution of each file.

As for general references with audio, I recommend the four listed at Help:IPA#External links. They include audio recorded by renowned phoneticians who have served as IPA President (Ladefoged, Wells, Esling, Keating) and work in modern browsers. Nardog (talk) 11:20, 1 October 2019 (UTC)

Thanks a lot Nardog for these references. This is exactly what I was looking for, very useful. LinguisticStudent (talk) 18:53, 1 October 2019 (UTC)

It would certainly in the best interest of this article to redo the sounds, many of which seem inaccurate, and all of which are painful to listen to. Hyxl4161 (talk) 22:05, 19 March 2020 (UTC)

Some people say "[ɒ] is never found in actual languages", but why it's exist. Who can explain this? Juidzi (talk) 02:04, 26 June 2020 (UTC)

It is in many languages. Even in English. --YodaMaster445 (talk) 09:43, 27 December 2020 (UTC)

@Juidzi: Lindsey does not say this. He says that ⟨ɒ⟩ is not a useful symbol and the rounded vowel in the open back area is adequately covered by the symbols ⟨ɑ⟩ and ⟨ɔ⟩. Personally, I agree with him. Sol505000 (talk) 12:51, 10 January 2021 (UTC)

Yes! but more details are clarify to me. Juidzi (talk) 11:35, 2 January 2021 (UTC)

A click takes you to Open back rounded vowel. Love —LiliCharlie (talk) 12:20, 2 January 2021 (UTC)

Just to make things quite clear, the opinion expressed by Geoff Lindsey refers to the symbols used for British English. That does not mean that the IPA does not need or want an open back rounded vowel [ɒ] for general use. The vowel chart would be incomplete without it. RoachPeter (talk) 10:00, 11 January 2021 (UTC)

@RoachPeter Thanks! for some clarification. Juidzi (talk) 10:49, 11 January 2021 (UTC)

Spanish E (e̞).--Manfariel (talk) 16:12, 23 July 2020 (UTC)

No, that one is there. Mgasparin (talk) 07:18, 13 November 2020 (UTC)

It wasn't when the above comment was made. Nardog (talk) 14:40, 18 February 2021 (UTC)

There are two vowels with missing sounds: ø̞ and ɤ̞. Perhaps someone involved in this WikiProject could attend to this. Mgasparin (talk) 07:21, 13 November 2020 (UTC)

I added ø̞. I could also add ɤ̞. --YodaMaster445 (talk) 20:40, 24 December 2020 (UTC)

The person in the audio pronounces [ɔ] and [ɒ] the same. I'm convinced it's actually the same voice clip. Could someone please fix this? — Preceding unsigned comment added by Omoutuazn (talk • contribs) 15:23, 3 January 2021 (UTC)

The sounds are almost the same auditorily and the distinction is meaningless in most cases. Vowels that are transcribed with ⟨ɒ⟩ in world's languages can be (and indeed, often are) transcribed with ⟨ɔ⟩ instead (in British English, the latter vowel is usually closer to [oː], so that cot is [kɔt] and caught [koːt]). Geoff Lindsey argues that ⟨ɒ⟩ is a superfluous symbol and I completely agree with him (though I prefer to replace it with ⟨ɔ⟩ rather than ⟨ɑ⟩, so that Swedish jag is written /jɔːɡ/), though rounding is sometimes so weak that ⟨ɑ⟩ may be more appropriate. The same goes for e.g. ⟨ɶ⟩, which is adequately covered by ⟨œ⟩ and even ⟨ɐ⟩. Sol505000 (talk) 05:57, 5 January 2021 (UTC)

As I've just noted in the discussion of [ɒ] above, it's important at all times to distinguish between the choice of a set of IPA vowel symbols for a particular language and the symbols used by the IPA's Cardinal Vowel system which provides a framework for classifying any vowel of any language in the world. You can easily do without [ɒ] for British English if you want, but it would not make sense to remove it from the Cardinal Vowel system. RoachPeter (talk) 10:05, 11 January 2021 (UTC)

@RoachPeter: Lindsey does seem to argue that ⟨ɒ⟩ is superfluous in any language, and that he's not aware of any language in which genuinely back [ɑ, ɒ, ɔ] (even better: [ɑ, ɒ, ɔ, o], or even better [a, ɑ, ɒ, ɔ, o]) would contrast phonemically based on height and roundedness alone. Do you know a language like that? I'm being genuinely curious. Sol505000 (talk) 10:59, 11 January 2021 (UTC)

The Amstetten dialect of Bavarian as described in Ladefoged & Maddieson (1996:290) comes close. Note that the vowel transcribed with the symbol ⟨a⟩ is much backer than /æ/ and /ɶ/ (which BTW contrast with /ɛ/, /e/, /œ/ and /ø/). It would be nice to get a recording, but it is quite usual for Bavarian dialects to realise /a/ as a vowel that is much closer to cardinal [ɑ] than to [a]. Love —LiliCharlie (talk) 12:28, 11 January 2021 (UTC)

We need to get a few basic principles straight. The article's title refers to the IPA vowel chart, so it cannot be a presentation of a radically different view of vowel classification put forward (very cogently) by Geoff Lindsey. As long as there is an official IPA vowel chart and a set of official auditory vowel qualities corresponding to the points on it, that's what you have to work with IF your article is called "IPA vowel chart with audio". I have no objection to someone creating a different article called "Lindsey vowel chart with audio". Secondly, the question of whether or not a phonemic contrast exists somewhere in the world between two particular vowels has nothing to do with whether a particular vowel should appear on the IPA vowel chart. The chart (like the full IPA chart) is supposed to represent the phonetic capabilities of human speakers, a set of templates that students of phonetics can learn and use as a standard for comparison in working with different languages and dialects. In this way, if some field worker stumbles on a previously unnoticed vowel quality in a hitherto unrecorded language or dialect, there is a symbol there for it in the phonetician's toolkit, and a set of known vowel qualities to use as a comparative standard. The argument about whether [ɒ], [ɑ] and [ɔ] are truly distinct can get a bit like medieval theologians arguing about how many angels can dance on a pin-head. But the two qualities recorded for the vowels [ɔ] and [ɒ] for this chart (by me) are different from each other auditorily, articulatorily and acoustically. The candidate usually trotted out to challenge the IPA's Cardinal Vowel symbol set is the front open rounded [ɶ], on the grounds that no known language possesses a phoneme for which this would be the phonetically appropriate symbol. However, any phonetician (including me) who has had a thorough training in traditional phonetics can produce this vowel and recognize it. It may well be time to consider changing the system, but as far as I know Wikipedia is not the place to promote changes to established academic frameworks. Some WP editors have already made widespread revisions to the official (full) IPA Chart, work which (impressive though it is) I still consider to be a clear case of OR. RoachPeter (talk) 15:47, 11 January 2021 (UTC)

@RoachPeter: Who said anything about changing this article? I certainly didn't. On what basis could it be done anyway? Lindsey's WP:FRINGE viewpoint expressed in a blogpost? My only question was: are you aware of a language that features a contrast between [ɑ], [ɒ] and [ɔ]. If you're not, that's perfectly fine. But the question was (and still is) relevant to the issue at hand, per WP:NOTAFORUM. @LiliCharlie: thanks for the example. Sol505000 (talk) 17:57, 11 January 2021 (UTC)

@Sol505000:I'm sorry for causing confusion with my over-interpretation of the discussion in this section. I read (above) the statement "Geoff Lindsey argues that ⟨ɒ⟩ is a superfluous symbol and I completely agree with him (though I prefer to replace it with ⟨ɔ⟩ rather than ⟨ɑ⟩" and I interpreted that to mean that you felt the IPA Chart ought to be changed to do without the [ɒ] vowel. That would, of course, entail a change to this article. On other points, I feel that there are many interesting things that could be said about alternative vowel classification schemes, and the WP article Vowel diagram is an excellent basis. It used to be claimed (as I am sure you know) that English Received Pronunciation (early to mid 20th century) had three contrasting vowel phonemes in the back open area that could be symbolized /ɑ/, /ɒ/ and /ɔ/ (though it was never claimed that these vowels were identical to Cardinal values). Ida Ward, for example, cited 'calm', 'not', 'bought'.^[1] Most modern writers reject such an analysis if it leaves out the length factor, and the phonetic quality of these RP phonemes has shifted significantly since Ward's time. Finally, I am afraid I don't understand what you mean by WP:NOTAFORUM, though I did look it up. RoachPeter (talk) 11:03, 12 January 2021 (UTC)