xkcd

by **gmalivuk** » Sat Oct 13, 2007 7:31 am UTC

Edit: Probably the single most useful link so far provided in this thread is the IPA character picker.

It is often useful when talking about language to be able to clearly describe the pronunciation of words in a way everyone can compare. Trying to spell it out with English's limited number of letters doesn't work very well, since your own dialect will influence how you read those. Therefore, linguists use the International Phonetic Alphabet.

Now, obviously most of us are not linguists, and so the IPA can seem a bit intimidating, especially at first. But there are a few sites I think many people would find make dealing with it much easier.

First and foremost, the University of Victoria has a convenient IPA page where you can hear the pronunciation of each phoneme.
Secondly, there is the Wikipedia IPA entry, which you can use to read about it (obviously) as well as to copy and paste any characters your keyboard doesn't have.
If you're just talking about English, no need to try to find the right sound out of all the IPA characters. The IPA chart for English even goes so far as to arrange its vowel list by major English dialect. Also, the format makes copying characters with the mouse a bit easier than the general IPA page, especially the vowels.

Now I know that not everyone likes the existence of the IPA as much as I do, so know that I am certainly not requesting that it be made some kind of rule for this forum that people must use it. It's just that, when the conversation is about pronunciation, people can make their points far more clearly if they're using a standard representation of different sounds. Especially when it's an alphabet that was specifically designed to be able to represent every meaningful difference in pronunciation.

by **Laogeodritt** » Sat Oct 13, 2007 3:46 pm UTC

This is an IPA "keyboard" I found a while ago. It's extremely useful to type in IPA—still a bit tedious, but faster than copying and pasting from the Wikipedia page. (Evidently, it requires Javascript to be enabled.)

http://linguiste.org/phonetics/ipa/chart/keyboard/

I've been slowly learning it for two years or so, now. I'm far from mastering it though.

by **Dingbats** » Sat Oct 13, 2007 5:43 pm UTC

Since IPA can be hard to type and display for a lot of people, I would recommend X-SAMPA, an ASCII adaptation of IPA.

by **Soap** » Sun Oct 14, 2007 12:51 am UTC

This is a chart with both IPA and SAMPA marked together, which I've found to be easier to read than the chart on Wikipedia since it's organized by place and type of sound rather than being in alphabetical order.

http://www.3centsoap.com/sampa.gif

I don't know who the original author of this chart is, it's been being passed around the Internet for a long time now. In fact, it's technically out of date, since the IPA was revised in 2003 and this chart still shows the 1993 version ... but the changes from the 1993 IPA to the 2003 one are extremely minimal and the SAMPA symbols are bound to the IPA symbols rather than to sounds, so they would change in exactly the same ways. And so the chart is still basically correct.

by **fnordulicious** » Mon Oct 15, 2007 12:03 am UTC

Please don’t use SAMPA or X-SAMPA. In these days of Unicode they’re really irrelevant, and even harder to read than IPA. That’s my professional opinion as a linguist who’s working on orthographic issues right now.

For Windows users, here’s an IPA character map that allows for easy input using an IPA chart–like interface: IPA Character Map.

An equivalent for Mac OS X is an input palette which works similarly to the Character Palette or Japanese Kana Palette among others: IPA Palette.

And for a web-based solution, here’s a site full of Unicode character pickers that runs in a browser: Unicode Character Pickers. The IPA-specific one is the IPA Character Picker.

If you’d rather use keyboard input, then you can try a keyboard layout for your favorite OS. Two keyboard layouts for Windows are Richard Collins’s IPA Keyboard Layout for Windows and the collection of layouts from SIL International (née Summer Institute of Linguistics). If you want to fiddle with a layout or create your own, use the Microsoft Keyboard Layout Creator.

For Mac OS X, SIL also offers a good keyboard layout but if you have specific characters you use frequently I recommend editing it with Ukelele (sic), which is SIL’s Mac OS X keyboard layout editing program. There is another IPA keyboard layout I’ve used, but I don’t like it much and won’t recommend it here.

Lots of other information and links on keyboard layouts is available from SIL’s Input Resources page, a service of the Non-Roman Script Initiative.

I apologize for not having a link to an X11 input method for IPA. I don’t use X11 for any sort of linguistic purposes so I’ve never needed to look for one.

(BTW, I am in no way affiliated with SIL International or Wycliffe International, other than being a happy user of their linguistic research and computer products. Heck, I’m not even a Christian. No ranting about SIL is invited here, please take it to a different forum.)

by **Laogeodritt** » Mon Oct 15, 2007 12:18 am UTC

Whoa.

Thanks for the keyboard layouts; I've been looking for some.

EDIT: I seriously need to get out of the habit of comma splicing.

by **Alpha Omicron** » Wed Oct 24, 2007 9:05 pm UTC

IPA charts with sound, done in flash:

http://www.paulmeier.com/ipa/charts.html

by **Ari** » Thu Oct 25, 2007 3:39 pm UTC

Echoing Fnord, SAMPA is really only to save time, or for those of us who like to have arguements over pronounciation on our cellphones. (don't ask) Real nerds use the IPA.

by **Memo** » Thu Oct 25, 2007 3:44 pm UTC

I love IPA, it makes learning non-phonetic foreign languages so much easier than trying to guess the pronunciation.

by **Ari** » Thu Oct 25, 2007 3:49 pm UTC

Memo wrote:I love IPA, it makes learning non-phonetic foreign languages so much easier than trying to guess the pronunciation.

It also makes it possible to discuss the highly amusing variation in vowel sounds among english accents without resorting to such faux pas as "fush and chups"

by **Pesto** » Thu Nov 01, 2007 8:17 pm UTC

Here's a link to some IPA fonts, free for download.

http://www.phon.ucl.ac.uk/home/wells/fonts.htm

by **stevekl** » Fri Mar 28, 2008 2:02 am UTC

I have a phonetic vowel chart, including a 'schwa' tattooed on my back. It's 15 years old and two-color; I should really get it touched up or redone.

by **gmalivuk** » Fri Mar 28, 2008 4:58 pm UTC

stevekl wrote:I have a phonetic vowel chart, including a 'schwa' tattooed on my back.

Why on Earth would anyone want a (presumably English) vowel chart that *didn't* include that most common of English vowel sounds?

by **stevekl** » Fri Mar 28, 2008 5:43 pm UTC

gmalivuk wrote:Why on Earth would anyone want a (presumably English) vowel chart that *didn't* include that most common of English vowel sounds?

It's not a tattoo of English vowel sounds. What's considered a vowel in English is actually rendered as a diphthong in the IPA anyway : the "a" in "paste" is is rendered like [eI] (where the I looks like a small cap letter I). It's just [a] [e] [i] [o] [u] and schwa placed on a quadrilateral representing the oral cavity, but when I have it redone, it'll be the full enchilada: http://tinyurl.com/3ar6sl

by **eierkopf** » Fri Apr 18, 2008 4:30 pm UTC

I just wish I had found this forum post when I had *started* graduate school. How impressed I am. I had to memorize the IPA chart for a test - and then promptly shoved it aside to make room for tree drawing and dangling participles. None the less, am pleased to know where to go for reference. Yay!

by **vorpal** » Tue Apr 29, 2008 9:38 am UTC

gmalivuk wrote:First and foremost, the University of Victoria has a convenient IPA page where you can hear the pronunciation of each phoneme

You mean phone.

by **BeetlesBane** » Mon May 19, 2008 3:15 am UTC

stevekl wrote:I have a phonetic vowel chart, including a 'schwa' tattooed on my back. It's 15 years old and two-color; I should really get it touched up or redone.

However do you manage to use it?

by **Laogeodritt** » Mon May 19, 2008 3:37 am UTC

Maybe it's not for his benefit but someone else's? 8D;

by **TheGrammarBolshevik** » Mon Jun 30, 2008 3:34 pm UTC

ipa.typeit.org is useful (and easy to remember) for times when you're unable to install another keyboard. It only has English sounds, but it does the job in a lot of cases. There are also separate pages on that site for writing (non-IPA) special characters from other languages.

Edit: Fixing linkfail.

by **Kosana** » Fri Sep 26, 2008 7:46 pm UTC

Actually Vorpal, I think he does mean Phoneme.

Long live the IPA!
It makes me happy inside. And diphthongs really are strange things, aren't they?

by **gmalivuk** » Wed Oct 01, 2008 3:29 pm UTC

Ari wrote:Echoing Fnord, SAMPA is really only to save time, or for those of us who like to have arguements over pronounciation on our cellphones. (don't ask) Real nerds use the IPA.

Seconding this, as sort of a linguistics forum-wide notion.

If you use SAMPA, I will think you are as inept at making things that look good as I will in the math forum if you try to type out everything on one line instead of using superscripts and subscripts and, even better, the nifty TeX interpreter that's right on the forum.

It might be easier to type, but it's a hell of a lot harder to read, and more people read each post than type it, so using IPA really is worth your time if you expect people to read your posts.

by **goofy** » Wed Oct 01, 2008 6:22 pm UTC

Kosana wrote:Actually Vorpal, I think he does mean Phoneme.

I think it should be phone. A phone is a speech sound. A phoneme is a contrastive sound in a language - the smallest unit that distinguishes meaning. If you say a sound is a phoneme, you're making a claim about a language's phonological system. So the IPA chart is a chart of phones, because it is not language-specific.

by **Qoppa** » Sun Oct 05, 2008 11:40 pm UTC

gmalivuk wrote:It might be easier to type, but it's a hell of a lot harder to read, and more people read each post than type it, so using IPA really is worth your time if you expect people to read your posts.

Lies. I post a lot on the Zompist bboards where X-SAMPA is the norm. I can read X-SAMPA just as easily, if not easier than IPA. It just depends on what you're used to. Nevertheless, I'll use IPA on this board, but it's really not any easier to read, especially when lots of diacritics are involved.

by **Dingbats** » Mon Oct 06, 2008 8:45 pm UTC

Qoppa wrote:
gmalivuk wrote:It might be easier to type, but it's a hell of a lot harder to read, and more people read each post than type it, so using IPA really is worth your time if you expect people to read your posts.
Lies. I post a lot on the Zompist bboards where X-SAMPA is the norm. I can read X-SAMPA just as easily, if not easier than IPA. It just depends on what you're used to. Nevertheless, I'll use IPA on this board, but it's really not any easier to read, especially when lots of diacritics are involved.

I agree. It's just a matter of getting used to either system. X-SAMPA is more longwinded, but then you don't have to squint to see the diacritics.

by **BrainMagMo** » Thu Oct 16, 2008 7:01 am UTC

I googled quite a bit but I can't get what I'm looking for.
What I want is a certain font face, so that if I type something in (X)SAMPA, it gives an IPA-looking character. ¿Does this exist? if so, a link would be appreciated.

by **ZLVT** » Thu Oct 16, 2008 9:55 am UTC

Yuo mean somethign to convert your X-Sampa to IPA? Well, there is a chart in the links above giving the conversions and a thingy which will type IPa for you but I don't know if they have a converter.

by **Alpha Omicron** » Thu Oct 16, 2008 12:41 pm UTC

BrainMagMo wrote:I googled quite a bit but I can't get what I'm looking for.
What I want is a certain font face, so that if I type something in (X)SAMPA, it gives an IPA-looking character. ¿Does this exist? if so, a link would be appreciated.

You could make your own, with a good Unicode keyboard layout creator.

by **Supaiku** » Thu Apr 16, 2009 3:28 pm UTC

So I was working on transcribing a sizeable list of works and I've run into some frustration over how to transcribe some of the vowels.
One is with words like
"lay", "bayed"
I have a Cambrdige published Handbook of the IPA and when references "American English" (i realize this is still argueable with reguard to dialects) it says you should use /e , eː , eː , eʲ/depending on the precision you demand.
However, I learned in univeristy /eɪ/ for this. Dictionary.com's IPA for"bayed" also has /eɪ/
I feel like I even saw /ɜɪ/ somewhere once.

they also give "bode" as an example and recommend /o, oː, oː, oʷ/.
This should be the same vowel sound as "go" which I learned as /goʊ/ː Dictionary.com has /oʊ/ for both "go" and "bode."

And then there's "boughed" transcribed with a /aʊ,aʊ,aʊ,aʷ/.

So, are these indeed diphthongs and why doesn't this book recognize them as such? (Or is it and I just don't know that those cute little quasi consovoles are the trick) I would expect this book to be accurate as it was published by the International Phonetic Association in 1999 (http://www.langsci.ucl.ac.uk/ipa/handbook.html)

There is one sentence in the book that saysː
'All of these (and several other) styles of transcription are properly reguarded as IPA transciptions of California english, provided they are accompanied by suitable conventions." (page 42)
Whatever suitable conventions means, I thought this was a standard... I can deal with differing levels of precision but several other styles? Does this mean that both the /oʷ/ and the /oʊ/ are valid transcriptions of a diphthong?

If you think I might find a better answer in another place please send me a link, this seems like a neat place, so hopefully someone can give me some insightː)
Thanks
CJ

by **gmalivuk** » Thu Apr 16, 2009 3:56 pm UTC

Personally, I treat long 'a', long 'o', and long 'i' as diphthongs (/eɪ/, /oʊ/, and /aɪ/) when transcribing words for my ESL class, and I think most dictionaries do as well. The alternatives with little superscript characters and such seem more for saying which exact phones a person uses, rather than phonemes. (Bough would for me be /baʊ/.) Long 'e' and 'u' aren't so much, and would typically be /i:/ and /u:/ in words like meet and moot, and /i/ and /u/ in words like happy and Malibu.

by **hocl** » Thu Apr 16, 2009 6:16 pm UTC

Be careful about the wikipedia pronunciations, though. They are quite different from the ones given on the IPA website. Plus the audio files seem to have a sort of vibrato to them.

According to wikipedia, for the second vowel sound in "English" for Americans, Canadians, Scots, and Welsh, the IPA symbol is ɨ and the pronunciation is http://upload.wikimedia.org/wikipedia/c ... _vowel.ogg

WTF?

by **Supaiku** » Fri Apr 17, 2009 1:15 am UTC

gmalivuk wrote:Personally, I treat long 'a', long 'o', and long 'i' as diphthongs (/eɪ/, /oʊ/, and /aɪ/) when transcribing words for my ESL class, and I think most dictionaries do as well. The alternatives with little superscript characters and such seem more for saying which exact phones a person uses, rather than phonemes. (Bough would for me be /baʊ/.) Long 'e' and 'u' aren't so much, and would typically be /i:/ and /u:/ in words like meet and moot, and /i/ and /u/ in words like happy and Malibu.

What I don't get is how /eɪ/ even begins to qualify as a single long vowel... how is not not clearly a diphthong?
Indeed, this is for my ESL class SO I must use a the more common double vowel interpertation or else the kindergardener's parent's brains might malfunction, but why isn't that a standard? What is a long vowel?

I just noticed: "/e/ and /o/ are usually slightly diphthongized"...
Gah. What does slightly even mean? Ahhh! :shock:

I guess including preciese numbers would only exacerbate the problem (of IPA being inaccessable to most), but like.. don't these sound like pretty def inite diphthongs to you?

Oh well. :roll:

by **gmalivuk** » Fri Apr 17, 2009 5:17 am UTC

Of course it's phonologically a diphthong or glide, but the issue with phonemic transcription is that the mental phoneme might not necessarily be a diphthong. (It is, I think, with "long I", or /aɪ/, but not necessarily with "long A", or /eɪ/, because the difference between /baɪt/ and /bat/ makes them different words, while the difference between /beɪt/ and /bet/ is more of an accent difference, I think.)

Also, there's the fact that certain sounds can be accurately represented in multiple ways using IPA characters. I think it's somewhat of a stylistic determination based in part on the phonemes being worked with.

by **Apeiron** » Tue Jun 30, 2009 4:45 pm UTC

Dingbats wrote:Since IPA can be hard to type and display for a lot of people, I would recommend X-SAMPA, an ASCII adaptation of IPA.

i've made my own system i call 'I Can't Believe It's Not SAMPA'. IPA is untypeable, SAMPA uses multiple characters for single sounds, so i use this creation of mine. It's usefulness is limited to American English, but i think it's vastly more intuitive to American English speakers. This is what i'd use for a phonetic system to teach kids.

by **Bobber** » Thu Jul 02, 2009 3:41 am UTC

Apeiron wrote:SAMPA uses multiple characters for single sounds

Huh? Are you sure that you are not misinterpreting some of the characters of mishearing some of the sounds? I am almost 100% certain that all characters used in SAMPA represent distinct sounds.

by **gmalivuk** » Thu Jul 02, 2009 4:16 am UTC

Bobber wrote:I am almost 100% certain that all characters used in SAMPA represent distinct sounds.

No, not all of them. There are several sounds represented by pairs of characters, though I don't think any of them are used in English. So if you are just focusing on English, I'm not sure why you'd prefer something else over SAMPA or IPA, which at least have the benefit that other linguistically inclined people can read them.

by **Laogeodritt** » Thu Jul 02, 2009 6:16 am UTC

gmalivuk wrote:There are several sounds represented by pairs of characters, though I don't think any of them are used in English.

Well, there's the English R, which is an alveolar approximant [ɹ], [r\] in X-SAMPA... granted, I believe some English IPA transcriptions use the symbol [r] for it anyway, though.

by **Apeiron** » Thu Jul 02, 2009 3:21 pm UTC

gmalivuk wrote:
Bobber wrote:I am almost 100% certain that all characters used in SAMPA represent distinct sounds.

No, not all of them. There are several sounds represented by pairs of characters, though I don't think any of them are used in English. So if you are just focusing on English, I'm not sure why you'd prefer something else over SAMPA or IPA, which at least have the benefit that other linguistically inclined people can read them.

i'm focused on my own ConLang, but i'm limiting it's sounds to those of English. SAMPA, as i said, sometimes uses multiple characters for a sound (even sounds within English {OI, 3: ju:), which does not meet my 1:1 ratio requirement. IPA is, as i said, untypeable. i have a set of symbols i can write and type easily (the ASCII set) (w/o doing any escape sequences, w/o using a virtual keyboard, w/o needing to load additional character sets, w/o using more than one symbol per sound[for the sounds i'm using]). i came up with this so i could have phonetic spelling of the words in my ConLang with minimum effort.

kate = kAt
cat = kat

It jives with the long and short vowels most American kids learned in school, which of course linguistically inclined people don't use. For them, vowels are rounded, open and what not. Not being a linguist, and being lazy, ICBINS works perfectly for me.

If i were to publish any part of my ConLang, it would have it's own symbols with IPA in the chart for all the language nerds.

With a lil more work it could be something English dictionaries could use for pronunciation guides.

by **Bobber** » Thu Jul 02, 2009 11:27 pm UTC

gmalivuk wrote:
Bobber wrote:I am almost 100% certain that all characters used in SAMPA represent distinct sounds.

No, not all of them.

The chart's pretty big, and I can't find any sound on it represented by two or more of the blue characters. Help me out

EDIT: I found two: the labiodental approximant can be P or v\, and the nasalized diacritic can be either ~ or _~

by **gmalivuk** » Thu Jul 02, 2009 11:43 pm UTC

Rhoticized schwa, which is pretty common in English, is @`, and there's the already mentioned true English r, which is r\ (since SAMPA r is technically the trill and not the approximant.)
Several of the sounds farther back in the mouth include ` or \
All of the non-pulmonic consonants and suprasegmentals and tone indicators involve multiple characters.

You didn't actually look very hard, it seems...

by **Laogeodritt** » Fri Jul 03, 2009 12:09 am UTC

I think we're interpreting Bobber's message differently...

gmalivuk and I understand "any single character in SAMPA represents a distinct sound", while Bobber seems to have meant "any phone is represented by a single SAMPA symbol (character or combination)" (although I would have understood this interpretation the opposite way).

Who is online