The 107 Consonants and 31 Vowels: How the International Phonetic Alphabet Maps All Human Speech Sounds

When a primary school teacher points to the alphabet on the classroom wall, the implicit claim is that those 26 letters are the foundational building blocks of human speech. But the anatomical reality of language contradicts that premise entirely. The English language alone contains roughly 44 distinct sounds, forcing its 26 Latin letters into chaotic, overlapping combinations. Because English has a deep orthography, a single letter like "c" can be pronounced as an "s" or a "k," and the sound /m/ can be spelled "m," "mb," or "mn." To actually map human speech, linguists abandon standard spelling entirely. They recognize that an alphabet tied to historical etymology cannot accurately record the physical reality of how humans talk.[3]

Instead, researchers and educators rely on the International Phonetic Alphabet (IPA). First published in 1888 by a group of French and British language teachers led by Paul Passy, the IPA was designed to do what no national orthography could: provide a single, unambiguous symbol for every distinct sound produced by the human vocal tract. "The International Phonetic Alphabet solves the problem of limited sound representation by making use of non-Latin characters," notes Voices.com, allowing it to offer "distinct representations for sounds from any language or dialect on earth." By decoupling sound from spelling, the IPA created a universal standard for transcription.[2][3]

Today, the full IPA chart contains exactly 107 consonants and 31 vowels, alongside 31 diacritical marks that modify them. It is a comprehensive map of human speech, stripping away the historical baggage of spelling to focus purely on the physical mechanics of how a sound is made. There are no letters that change their pronunciation based on context, and no silent characters. Every symbol corresponds to one specific acoustic event. This one-to-one mapping allows a linguist to look at a transcription of a language they do not speak and pronounce it with near-perfect accuracy.[1][3]

The mechanism behind the IPA is strictly anatomical. The chart does not organize sounds by their history or their alphabet, but by where and how they are physically produced in the mouth and throat. For consonants, the chart operates on a rigid grid of "place" and "manner" of articulation. The columns represent the place—where the airflow is obstructed by the tongue, teeth, or lips. This ranges from the front of the mouth (bilabial sounds made with both lips) to the teeth (dental), all the way back to the glottis deep in the throat.[1][4]

The rows represent the manner—how the airflow is manipulated once it is obstructed. A "plosive" involves a complete block and sudden release of air, like the /p/ in "prawn," while a "fricative" squeezes air through a narrow channel, like the /f/ in "fish." Within each cell on this grid, sounds are further divided by voicing: whether the vocal cords vibrate during the sound, as in a voiced /z/, or remain completely still, as in a voiceless /s/. This matrix accounts for every consonant humans naturally produce.[1][4]

Vowels, meanwhile, are mapped on a separate trapezoidal chart that mimics the open space inside the human mouth. Because vowels do not involve obstructing the airflow the way consonants do, they cannot be categorized by strict friction or closure. Instead, the vertical axis of the vowel trapezoid tracks how high or low the tongue sits in the mouth, while the horizontal axis tracks how far forward or back the tongue is positioned. For example, the /i/ sound in "fleece" requires the tongue to be pushed high and to the front, while the /ɑ/ sound in "father" drops the tongue low and to the back. A third dimension for vowels is roundedness—whether the lips are pursed or relaxed. In the IPA vowel chart, unrounded vowels sit on the left of a given position, while rounded vowels sit on the right.[1]

This anatomical precision allows the IPA to capture sounds that simply do not exist in European languages. It includes specific symbols for the click consonants found in some African languages, the ejectives of the Caucasus, and the tonal variations essential to Mandarin. When the standard 107 consonants and 31 vowels are insufficient, linguists deploy the 31 diacritical marks to indicate subtle shifts, such as nasalization, aspiration, or a sound being held for twice as long. There is even an extended set of IPA symbols designed to represent atypical speech qualities, such as lisping or sounds made with a cleft palate, ensuring that speech-language pathologists have the tools they need for clinical diagnosis.[1][3][4]

Despite its global ambition, the IPA retains a heavy bias toward the Latin and Greek scripts. Its 19th-century creators explicitly aimed to use as few non-Latin letters as possible to maintain international harmony, modifying existing letters or borrowing from Greek when necessary. For instance, the symbol for the glottal stop, /ʔ/, originally took the form of a question mark with the dot removed. Other symbols are simply inverted or rotated Latin letters, such as the schwa /ə/ or the turned 'a' /ɐ/. This reliance on familiar letterforms makes the system easier for Western scholars to adopt, but it also reflects the European origins of the International Phonetic Association, which was initially founded to help French and British students learn each other's languages.[2][3][4]

The system also distinguishes between "broad" and "narrow" transcription, a crucial feature for researchers. Broad transcription, enclosed in forward slashes like /kæt/, captures the general phonemes of a word without worrying about the speaker's specific accent. Narrow transcription, enclosed in square brackets like [kʰæt], uses diacritics to record the exact, physical pronunciation of a specific speaker. This narrow approach captures regional accents, subtle variations in breath, and the precise moment a consonant is aspirated, allowing linguists to document how a language actually sounds in the wild rather than how it exists in a dictionary. As Voices.com points out, "a native Maine fisherman likely will pronounce that word quite differently from a native Kansan," and narrow IPA transcription is the only way to accurately record that divergence on paper.[2][3]

While it never became the universal everyday writing system its founders envisioned, the IPA remains the absolute bedrock of modern linguistics. From speech-language pathologists diagnosing articulation issues to actors mastering foreign accents and software engineers training voice recognition models, the 107 consonants and 31 vowels of the IPA ensure that a sound made in one corner of the world can be accurately reproduced anywhere else. It stands as a testament to the complexity of the human vocal tract, proving that the sounds we make are far too intricate to be contained by just 26 letters. By mapping the physical reality of speech rather than the arbitrary rules of spelling, the International Phonetic Alphabet provides the only truly universal code for human language.[2][3]