I have been introduced to the IPA diagram (the triangle/trapezoid) for vowel sounds and I find it interesting as a model of pronunciation as it represents the mouth position so it can be very informative if one wants to understand how to pronounce one vowel.

I was wondering whether there was any more advanced model/theory that would be a little more general, allowing one to understand how the pronunciation ability of more and more complex sounds, starting from simple vowels, diphthongs, and consonant sounds (even though I'm mostly interested in vowels only). Such a model would for example explain in more detail the tongue placement, tension etc...

What I would be interested in is to map any accoustic recording to a mouth position.

  • 2
    I would really encourage you to watch this video by Dr. Geoff Lindsey: youtu.be/FdldD0-kEcc . It will greatly improve your understanding of the vowel space, as it did mine.
    – Graham H.
    Jul 9, 2023 at 13:33
  • @GrahamH. Thanks for your suggestion, I just watched it and found it very interesting and relevant!
    – MiKiDe
    Jul 16, 2023 at 3:39

1 Answer 1


It is unclear what you what a more general model of (it's not clear to me whether you understand what the IPA is a model of, and what it would mean to make it more general). My best bet is that you are interested in articulatory synthesis, where you specify the state of the vocal tract, and get some resulting sound. This will give you some directions to look. However, the reverse process of estimating what vocal tract configuration might have created a particular acoustic output is a much more complicated problem, since waveform-to-vocal tract mapping is many-to-many.

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.