Universal Phonetic Alphabet
I always wondered, why is the IPA, whose name is International Phonetic Alphabet, based on the Latin alphabet. Not only that the Latin alphabet is far from being international; its shapes are not anyhow related to their sounds. Not to mention that by adding all those diacritical marks, hooks etc. it becomes quite ugly. The aim of this construction is to create a visually appealing phonetic alphabet which is more universal than the IPA but much easier to master, since it is based on a logical symbol construction process.
As you could have already noticed, consonants are defined by two “large” parameters and one “small” parameter – manner of articulation and place of articulation are the large parameters in the sense that there are many possible combinations of both, and being voiced/voiceless is the small parameter (since it takes only two possible values). For the indication of the value I use the following symbols:
![]()
The slash-like symbol is sort of central, the rounded symbols are somewhat opposite of the spiky ones. However, as you will see, they also maintain a linguistical logic. There will actually be more symbols, but those additional symbols are used for rarely used (and sort of complementary) concepts.
In the table of pulmonic consonants you can see probably all simple pulmonic consonants that a human can produce. The place of articulation axis is ordered by increasing depth in mouth/throat. The manner of articulation axis is sorted by increasing air flow through mouth.

Not all pulmonic consonants are of this kind. Some consonants are doubly articulated, t͡s, d͡z, t͡ʃ and d͡ʒ being the most common. These are marked in UPA by joining the bottoms of the two consonants, like (for the four mentioned):
![]()
From this notation you immediately see that d͡ʒ is only a voiced t͡ʃ etc.
Some consonants are only coarticulated (second articulation is done in the same manner). The second place of articulation is denoted by a modifier symbol attached to the consonant on the left. So, labialized (dʷ), labio-velarized (dʷ), velarized (dˠ), labio-palatalized(dɥ), palatalized (dʲ), and pharyngealized (dˁ) d would be:
![]()
Nasalization of a consonant is denoted like if it were “glottalization”:
![]()
Probably the most important concept for an English speaker is w and ʍ. Those are labialized velar approximants and would be written:

Another important concept is the release. Nasal release (dⁿ) is marked by attaching the nasal symbol to the right, voiceless aspiration (tʰ) is marked as “approximal release”, voiced (dʱ) then by adding a line to voiceless, lateral release (dˡ) by adding a lateral symbol and no audible release (d̚) by adding a dot:
![]()
Vowels form another important concept. Again, they are characterized by two large and one small parameters – closeness and frontness being the large, roundedness being the small.
In the table of vowels the frontness axis should be understood just as a parameter being relative to the position „as front as possible“ for the current closeness (it is not possible to pronounce fully open and front a).

Reverse symbols for vowels allow us not only to use the same components as for consonants, but also provide a good basis for kerning. Look at the following example that shows the word international (written as ɪntənæʃənəɫ):
![]()
It is esthetically very pleasant.
Diphthongs are in the IPA written using diacritics. English contains for example ɔɪ̯, aɪ̯, ʊə̯. In the UPA, diphthongs are formed by joining the two characters by an arc. The three mentioned would be:

Triphthongs are marked in the same manner, ɔɪ̯ə would be:

The length modifier is a small line on top of a vowel. In British English, “are” is pronounced [ɑː]. In UPA, this is written:
![]()
Nasalization is denoted in the similar way as the nasal release for consonants, ã would be:
![]()
To make our list of possible sounds complete, we need to consider non-pulmonic consonants:

For I am not really well acquainted with the concept of clicks, I won’t go to any further details here.
Stresses are indicated in the same manner as in the IPA, that is by a small preceding line (either top or bottom), like
![]()
in the word ɪntə’næʃənəɫ. Tones could be indicated by some diacritics above the letters, however I’ll let somebody else to decide the best way to do that. Punctuation in longer texts may be written in any way that suits one’s needs, however I recommend using a vertical line instead of dot and a „half vertical line“ instead of comma, like:
![]()
Here is a sample text in English (without stresses), written in the UPA, the IPA and the Latin alphabet:

ɔːl hjuːmən biː.ɪŋz ɑː bɔːn friː ænd iːkwəɫ ɪn dɪgnɪti ænd raɪts. ðeɪ ɑː ɪndaʊd wɪθ riːzn ænd kɒn ʃns ænd ʃʊd ækt təwɔːdz wʌn ənʌðə ɪn ə spɪrɪt ɒv brʌðəhʊd.
All human beings are born free and equal in dignity and rights. They are endowed with reason and conscience and should act towards one another in a spirit of brotherhood.
This is fascinating! Do You have a font file for use on a computer?
Yes and no. The font file I used to create the images in the article contains only parts of these characters (with zero width) and the spacing is done manually using spaces of different width (so it’s actually quite difficult to write the whole symbols). I hope that in the future I’ll make some font that will offer an easier way to write the basic IPA characters. Meanwhile, I could publish some specification of the font I use now, but unfortunately I am now quite busy
.