NATO phonetic alphabet

What is the NATO phonetic alphabet?

It is a set of 26 words, one per letter, chosen so that none can be mistaken for another over a bad radio link. Alfa, Bravo, Charlie, Delta and so on. The everyday name credits NATO, but the alphabet is not NATO's: it is defined in ICAO Annex 10, the international standard for aeronautical telecommunications, and reproduced in Appendix 14 of the ITU Radio Regulations, which covers maritime and general radio. It came into use on 1 March 1956, and the words have not changed since.

It is also not phonetic in the linguists' sense — it does not describe sounds, it substitutes whole words for letters. The specifications call it a spelling alphabet, which is exactly what it is.

The words on this page are transcribed from those two documents rather than from other websites, which matters more than it should, because the two spellings people most often get wrong are in the first half of the alphabet.

How to use the phonetic alphabet converter

1. Type or paste your text. Letters, digits and spaces all work, and the output updates as you type. Case is ignored, and a space between words becomes a slash so the grouping survives being read aloud.
2. Choose which numbers you want. Aviation gives the ICAO pronunciations — the digits said almost as English, respelled to survive a bad channel. Maritime gives the ITU figure code, which replaces them with entirely different words. The 26 letters are identical either way.
3. Read it off, or copy it. The reference tables below the output list every letter and figure with the pronunciation as printed in the specification. Anything the alphabet cannot express is left alone and counted, rather than quietly dropped.

One alphabet, two completely different sets of numbers

Diffing the ICAO table against the ITU table row by row, the letters are the same. Same 26 words, same pronunciations, with exactly one exception in the entire table: ICAO prints Hotel as HO TELL and the ITU prints HOH TELL. That is the whole disagreement between aviation and maritime on the letters.

The numbers are not the same at all. Aviation keeps the English digits and respells them to survive interference: three is TREE, four is FOW-er, five is FIFE, nine is NIN-er. The syllables in capitals are the stressed ones — the specification notes that ZE-RO takes equal emphasis on both syllables while FOW-er stresses the first.

The maritime figure code throws the English words away entirely. Zero is Nadazero, one is Unaone, two is Bissotwo, then Terrathree, Kartefour, Pantafive, Soxisix, Setteseven, Oktoeight and Novenine. Each is three syllables with the English digit buried at the end, so that even if the first part is lost the number survives. Of the eleven entries in each table, exactly one word appears in both: Decimal.

So the common claim that “there are several NATO alphabets” is wrong in an interesting way. There is one alphabet, and there are two unrelated ways of saying the digits. Which one you want depends on whether you are on an aircraft or a ship.

Alfa and Juliett are the spellings

Both documents print Alfa, not Alpha, and Juliett with two Ts, not Juliet. This is not a typo that got standardised — it is deliberate. “Ph” for an F sound is an English-language convention that speakers of most other languages do not share, and a final silent T is exactly the kind of letter that gets dropped. Spelling the words the way they must be said removes the ambiguity for a Spanish or Indonesian or Japanese speaker reading them cold.

The word Alpha does appear in ICAO Annex 10 — ten times — but never in the spelling alphabet. Every occurrence is “International Alphabet No. 2”, “No. 3” or “No. 5”, which are teletypewriter character encodings and have nothing to do with radio speech. It is an easy thing to search for and get wrong.

Two more that get corrupted in transit: it is Whiskey with an E, and X-ray with a hyphen.

Honest limitations

The stress marks for the letters are missing here, on purpose. Both specifications mark the emphasised syllable in the letter table by underlining it, and an underline does not survive when the published PDF is read as text. Rather than guess which syllable is underlined — and be wrong on a page whose whole point is fidelity to the source — the letters are given without stress. The number tables encode stress as capital letters instead, so those are reproduced exactly.

The alphabet covers A to Z and the ten digits and nothing else. There is no word for an accented letter, so this tool folds one to its base — café spells as Charlie Alfa Foxtrot Echo — which is a real loss of information and is stated rather than hidden. Punctuation and non-Latin scripts are left exactly as typed and counted separately, because a spelling alphabet that invents a word for ب would be worse than one that admits it cannot.

National variants are legitimate and are not shown here. Appendix 14 closes by permitting stations of the same country, communicating between themselves, to use any other table their administration recognises. That is the legal basis for the police alphabets — Adam, Boy, Charles — which people often mistake for this one. They are different systems with different purposes, not incorrect versions of this table.

The data is held to 44 assertions: that every code word begins with its own letter, that no folk spelling has crept in, that the ICAO and ITU tables differ on exactly one row and that row is Hotel, that the two figure sets share exactly one word, and that ordinary text reports nothing as unspellable.

Why is it free?

It is a lookup table and a loop, running in your browser as you type. Nothing you type is sent anywhere, there is no account, and there is no limit.

The reference tables are not held back either — the point of the page is that they are correct and sourced, which is worth nothing if you have to sign up to read them.