International Phonetic Alphabet

The International Phonetic Alphabet (IPA ) is a phonetic alphabet, that is a collection of characters by which the sounds of all human languages ​​can be described almost exactly and recorded. It was developed by the International Phonetic Association and is the most widely used today phonetic spelling system.

The current IPA characters and their pronunciations are listed in the list of IPA characters.

  • 3.1 Older notations
  • 3.2 IPA in speech technology and the Internet: SAMPA, X - SAMPA and Kirshenbaum

Practical significance

The IPA provides a substantial improvement of the representation of the pronunciation in dictionaries and encyclopedias dar. When reading IPA texts, however, caution is advised:

In some languages, such as French, there is not a generally accepted standard pronunciation ( Orthophonie ), in others. An officially appointed pronunciation, however, can be uncommon in everyday life. The Common bandwidth of a sound can be much larger (eg the German suffix - er) as the difference of similar phonetic characters. What is, interpreted in a language as correct or incorrect as normal or strange as understandable or incomprehensible is for someone who has rarely or never heard the language, not to gauge.

In dictionaries, a simplified character set is often used in order not to confuse readers with no previous knowledge. So Cassell's German Dictionary does not distinguish the different pronunciations of the German r, nor the more open pronunciation of short a, i, u and ü against the relevant long vowel. The pronunciation of English is no generally represented by tradition as [ noʊ ], although in the British actually [ nəʊ ] is said. Usually also ignores the fact that in some languages ​​without [ ∫] - (sch ) According to the s usually more like [ ɕ ] and ( between s [s ] and I - sound [ ç ] ) is pronounced, for example, in European Spanish in Greek ( a phonetically more accurate description is probably generally " withdrawn ", ie [s ], where the Spanish sound more apically is so [s ], the Greek, however, more laminal, ie [ S] ).

Quite as accurate as it appears, the IPA notation is not, for example, is the [ ʉ ] in Engl. book only slightly more central [ u], in the Norwegian and Swedish word sund only slightly more central [y ].

Character mappings of sounds and sound character extensions

The IPA symbol table use, among other letters of the Latin and Greek alphabets, some in modified form. Each character designates a sound or describes an already specified sound closer to the one word from another in a different language in the world.

The International Phonetic Alphabet is across languages ​​; This means that the assignment of a sign to a sound in a particular language is not necessarily identical with the sound assignment of the same character in the IPA. For example, the sign [ ç ] in ch IPA for the pronunciation of the letters of the German word " I ", although it is of German orthography alien; for mapping the pronunciation of the French, whose orthography " ç" knows as unvoiced "s", the character is, however, not required.

The special characters of the IPA alphabet were added to Unicode in the range U 0250 to U 02 AF.


Whiles the cursor beside a character in the table cell, its Unicode value is displayed; one points to the character, you get a short description as a tooltip.


Wherein consonants different air flow mechanisms can be distinguished.

The pulmonary consonants are outflowing with breathing air (ie, air from the lungs) produces ( pulmonary egressive ). This group includes the vast majority of consonants. In the ejective implosive and the air flow is, however, generated by movements of the larynx. In the ejective, the larynx moves up, so that air flows ( glottal - egressive ); at the implosive it moves down so that air flows ( glottal - ingressive ). Clicking sounds ( sometimes referred to as " Avulsive " or in English as " clicks" ) arise when the tongue and soft palate form a closed cavity is increased by a back -and-down movement of the tongue. Upon opening of the cavity is a pressure equalization takes ( air flows into it, so velar ingressive ) so that a sound is generated.

Pulmonary consonants

Whiles the cursor beside a character in the table cell, its Unicode value is displayed; one points to the character, you get a short description as a tooltip.

If necessary, each is left of the voiceless and the voiced consonant right.

Dark shaded fields denote physiologically impossible articulations. For example, a glottal nasal is impossible because in a closure of the vocal folds can be no accidental release of air through the nose, etc.

Non - pulmonary consonants

Other symbols


Of the suprasegmental mark the emphasis characters appear before the syllable to which they relate, the length of characters after that.

Sounds and intonation


  • Unicode does not own characters for most contour tones. Sequences of characters for register tones are used instead and the exact representation is left to the respective font, usually by OpenType rules: [e ᷇ E] or [e ˥ ˧ e ˧ ˩ ˨ ] (not shown properly in many browsers ). Because very few fonts allow the combining of Registerton characters, often the old system of sound mark is 'to '5' used by superscript numbers from '1, for example, [ E312 e53 ]. Their use, however, depends on local linguistics traditions; in Asian languages ​​"5" is used for the highest tone and "1" for the lowest in the African languages ​​vice versa. Occasionally, is still found an old IPA tradition, according to which the contour tones are indicated by diacritics under set: [ E E ] for low - falling or low - rise.


Unvoiced, or entstimmt unvoiced speech of a normally voiced sound, eg because it is surrounded by two voiceless.

Voiced Voiced speech of a normally unvoiced sound.

˞ combined characters: ɚ (U 025 A) ɝ, (U 025 D)

̴ combined characters: ɫ (U 026 B)

Note: Whether provided with diacritics characters are equivalents to the other characters of the manner of articulation and articulation of the place has not been fully established by the IPA. In "Handbook of the International Phonetic Association " states: "It is debatable whether [k] and [g ] denote phonetically identical sounds, and the same is true for [s ] and [ z]. May be in distinguishing between [k] and [g ] or [s ] and [ z] different dimensions involved, which are independent of the vocal cord vibration, such as tension over laxness in articulation, so the possibility of voicing to designate separately, is important. However, it may be advantageous in any case if one is able to maintain the lexical form of the word [...]. "

Characters with descenders can be distinguished by a diacritic over excepted. By default, it should be awarded placed below diacritics when both options are available.

Alternative Notations

Older notations

The IPA is not the only system of notation of speech sounds. Over time, there have been some attempts to represent sounds more accurately than in conventional spelling. Already in 1855 published the German Egyptologist Karl Richard Lepsius be standard alphabet " for the representation of unwritten languages ​​and foreign characters systems in a uniform orthography in European letters". Revised the work appeared in 1863 except in German and English. Contemporary texts over the sounds of human speech show that this standard alphabet was conceived as phonetics. Some of his characters are entered into the IPA alphabet. For German transliterations (ie reproductions of letters) was acquired his oriented in the Czech orthography distinguish different articulated sibilance, but not in the IPA phonetic alphabet.

Alexander Melville Bell asked in contrast, in 1867 in his system of Visible Speech an iconic notation before, are mapped in the individual characteristics of a sound ( eg rounding of the lips and the like) in the sign itself. Other attempts towards an illiterate notation have been made by the linguist Otto Jespersen (1889 ) and Kenneth L. Pike ( 1943). The fact that in these systems the individual positions of the organs of speech can be specified independently According nuances can be encoded much finer.

In the German and Romance languages ​​dialectology today is still the Teuthonista transcription usual. The Teuthonista was in 1924/25 presented by Hermann Teuchert in dialectological Teuthonista magazine and builds a reference to Karl Richard Lepsius ' Standard Alphabet primarily on the Latin alphabet on. Since similar proposals in the Graziadio Isaia Ascoli of Romania and Eduard Böhmer have been presented, the Teuthonista and the Bohemian - Ascoli transcription are now largely collapsed.

IPA in speech technology and the Internet: SAMPA, X - SAMPA and Kirshenbaum

With the notations SAMPA (for 7 European languages ​​) and X - SAMPA ( the SAMPA extension for the full IPA) were developed alphabets of European phoneticians and speech engineers in the context of multilingual language technology EU research projects in the 1980s, by which IPA characters could be written in ASCII code. These codes, which are the IPA mapped precisely and very common in speech technology, serves the following purposes:

  • Exchange phonetic data ( files transcription and speech signal annotation files ) in a simple text form.
  • Simple programming technical processing of transcriptions in automatic speech recognition and speech synthesis.
  • Easy checking and editing phonetic data while machine- readability.
  • Keyboard -friendly input of all displayable with the IPA sounds.

These codes were not designed for the general representation of the IPA in publications, but are often used in scientific and technical publications for data representation. For general purpose of publication, including on the Internet, are standardized Unicode character sets that are more text -output - input - oriented than, more suitable. Regardless of SAMPA and X - SAMPA the similar Kirshenbaum alphabet was developed by internet users in the early 1990s, but who could not prevail. In the U.S., " Klattbet " or the " Arpabet " in speech technology are mainly used for the English language often used.