Although spectrograms contain much more detail, I'm curious how far phonologists and phoneticists can go with IPA alone. Can someone provide some examples of narrow, detail-rich IPA transcription for various words in English, Spanish, German, Icelandic, etc.? I'd also be interested in seeing how IPA can capture existing phonetic features such as allophones, etc.

Alternatively, what detail can be denoted in a spectrogram but not IPA?

I've found the following examples of what I'm looking for on Wikipedia:

  • pets [pʰɛʔts]
  • pretzel [ˈpʰɹ̥ʷɛʔt.sɫ̩]
  • 2
    "Most insane" seems to be too opinion-based. May 14, 2014 at 22:20
  • 1
    Yeah. There's always more phonetic detail that can be annotated; but if it's detail you're after, use a recorder and a spectrograph. IPA is for specifying particular features when they're necessary or relevant; the rest of the time loose phonemic transcription is all one needs.
    – jlawler
    May 14, 2014 at 22:50
  • @jlawler actually, it's just for see how far phonetics and phonology can go with IPA :) I would like to see an example with most phonetic (phonological) features generally accepted by linguists for English (for example). And i understand that IPA it's not replacement of spectrograph. It's just for theoretical interests. May 14, 2014 at 23:07
  • 1
    Well, then, one example I used to use when teaching phonetics was the American McDonald's server dialog. The servers at McDonald's have phrases they hafta say and the customers hafta know their lines, too. So you get to the head of the line and the server makes eye contact with you and says ['ᶄʰɛ̧ᵚᶈɨ], you give your order, and the server replies [zɛ̧ʸ'θɛ̧ᵚs].
    – jlawler
    May 15, 2014 at 0:07
  • @jlawler Nice example. Can you please rewrite this comment as answer? That a could accept it. May 15, 2014 at 4:56

1 Answer 1


There are various aspects of disordered speech that cannot be expressed using IPA 2005. This is why, e.g. the ExtIPA chart was constructed. This covers things like:

  1. interdental/bidental articulation;
  2. whistled articulation;
  3. sliding articulation.

IPA Extensions also covers thing like:

  1. ʩ - velopharyngeal fricative (snoring);
  2. ʬ - bilabial percussive (lip smacking);
  3. ʭ - bidental percussive (teeth gnashing).

Another thing is lip rounding. The IPA only supports rounded/unrounded, labiablised and more/less rounding. It does not describe things like compressed vs protruded rounding -- this is implied by the position of the vowels.

Also, with a spectogram you can identify the F1 and F2 frequencies of the different vowels in different speakers and get an accurate representation of where they are in relation to each other. IPA is ambiguous depending on the transcriber (e.g. 〈a〉 can be either front 〈a̠〉 or central 〈ä〉).

  • Rather, i mean how far ipa can go for non-disordered speech. Actually, how far we can go for details from ipa of dictionary to ipa, let's say of researcher using only conventional phonetic features. But thank you for this answer on extension of IPA, i didn't no. Generally, i would like to see real examples of such a transcription like in comment of jlawler for different languages... May 18, 2014 at 21:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.