The NATO phonetic alphabet contains exactly 26 standardized code words, one for each letter A through Z, drawn from the international ICAO radiotelephony spelling alphabet and used by aviation, maritime, military, and emergency services to spell names and identifiers clearly.
Learning the list is much easier when every word sits on one page, when you can double-check the spelling — because Alfa is spelled with an f, Juliett carries two t's, and the letter X is written as X-ray — and when you can then test yourself by turning ordinary text into code words and turning those code words back into letters again. The NATO Phonetic Alphabet Translator does exactly that. It encodes any ASCII text against the verified 26-word table and decodes those code words back to plain letters, all in your browser with nothing uploaded to a server. That makes it a practical memorization aid for students, radio hobbyists, support agents, dispatchers, and anyone preparing for a classroom drill, a callsign exercise, or a real call where spelling has to be unambiguous. The tool is a spelling utility only. It will not pronounce the words for you, will not turn digits into aviation phrases such as Niner or Fife, will not teach radio procedure, and is not a substitute for qualified instruction in safety-critical environments. Treat it as a memorization surface: read the table first, encode a few words, then decode them to verify what you have learned.
The 26 Verified Code Words, in Order
The complete NATO phonetic alphabet — technically the ICAO radiotelephony spelling alphabet, even though "NATO phonetic alphabet" is the everyday name — is a single fixed table of 26 letter-to-word pairs. The translator you will use to encode and decode is built directly on top of this table, with each row checked against the FAA's operational ICAO phonetics table and ICAO Annex 10 Volume II. Memorization begins here. Read the table straight through once, then again in reverse, then cover the right-hand column and recite each code word from the letter alone. Once that cold-recall pass is comfortable, move on to encoding real words.
The full ICAO spelling table
| Letter | Code word | Letter | Code word |
|---|---|---|---|
| A | Alfa | N | November |
| B | Bravo | O | Oscar |
| C | Charlie | P | Papa |
| D | Delta | Q | Quebec |
| E | Echo | R | Romeo |
| F | Foxtrot | S | Sierra |
| G | Golf | T | Tango |
| H | Hotel | U | Uniform |
| I | India | V | Victor |
| J | Juliett | W | Whiskey |
| K | Kilo | X | X-ray |
| L | Lima | Y | Yankee |
| M | Mike | Z | Zulu |
Two rows above include spellings that learners tend to get wrong on first contact. Alfa ends in -a, not -pha. Juliett ends in -tt, with two final t characters, not the single t of the English given name. Everything else in the table is a normal English word whose spelling is the spelling you would expect. Treat the table above as the authority for every later exercise in this article and as the ground truth for every answer the translator will produce.
Where the Easy Misspellings Hide
Three entries in the table trip up the majority of new learners, and a fourth catches people who confuse this alphabet with the International Phonetic Alphabet (IPA) used by linguists to transcribe speech sounds. Naming them in advance stops the wrong shape from getting drilled in.
- Alfa. The official row uses an f, not ph. The spelling is from the ICAO radiotelephony alphabet and is independently confirmed in the FAA's published ICAO phonetics table at JO 7110.10EE.
- Juliett. The official row ends in two t's. A single-t Juliet looks more natural to many English speakers and is the form you will see in older or informal listings online. Trust the table in this article, not those.
- X-ray. The official row keeps the hyphen. The translator renders it as X-ray in the output. Treating it as two tokens or as the bare word "Xray" produces a decode error.
- The spelling alphabet, not the IPA. This is a code-word spelling alphabet, not a pronunciation key. The code words are designed to be unambiguous when spoken over a noisy radio channel. They are not transcriptions of how letters sound, and they overlap with — but are not the same as — the IPA symbols used in dictionary entries.
Because the translator's decoder deliberately rejects Alpha and treats other misspellings as unknown tokens rather than guessing, you can audit any training material you are given by running its example sentences through decode mode. The strictness is a feature: a plausible misspelling should never silently pass as an official match, and if your training sheet lists Juliet with one t, that tells you the sheet needs updating, not the alphabet.
Encoding Your First Word With the Translator
- Open the NATO Phonetic Alphabet Translator in your browser.
- Leave the direction set to "letters to code words" (encode).
- Type a short word or phrase into the input box. Start with a familiar example such as SOS or your own name. Letter case does not matter — the spelling alphabet identifies letters, not typography.
- Click Translate.
- Read each visible token in the result. For SOS, expect Sierra Oscar Sierra. Output tokens are joined with single spaces. Any run of spaces, tabs, or line breaks in your source is collapsed to a single visible slash, and a literal source slash is escaped as [/] so it cannot be confused with a word boundary.
- Click Copy result only after the status line confirms the browser accepted the clipboard write. If a permission error appears, select the visible result manually. The copied string matches the displayed output exactly, including slashes, spaces, digits, and punctuation.
The exercise above gives you an immediate visible reward for the first letters you have memorized and exposes any gap on the spot. If you typed SOS and got Sierra Oscar Sierra, your S and O recall is good. Any deviation tells you exactly which row in the table to revisit. Once you have a small vocabulary under your belt, push yourself: type your own full name, a street, a model number, or a callsign. Real content produces real learning, because the letters you actually need to spell are the ones most worth memorizing. Keep the input well below the 100,000 Unicode code points the translator accepts — far above anything you will plausibly need for spelling practice, but useful to know if you ever paste a large document into encode mode by mistake.
Decoding Code Words Back to Letters
Decoding is the second half of learning. Each code word you can produce from memory, you should also be able to recognize when you read or hear it. The decode mode of the translator lets you self-test that recognition without needing anyone to read code words at you.
- Switch the direction in the NATO Phonetic Alphabet Translator to "code words to letters" (decode).
- Paste a string of code words separated by single slashes. Slashes mark word boundaries: Sierra Oscar Sierra / 2 ! becomes SOS 2!. A run of spaces, tabs, or line breaks between words in your source counts as a boundary too.
- Click Translate and read the plain-text result. Compare it with what you expected, letter by letter.
- If a token is rejected with an error, read the token name you typed and check it against the reference table above. The decoder is case-insensitive and requires the exact spelling — Alpha is refused even though most other entries would normally decode, because the embedded row is Alfa. That refusal is intentional and protects you from training on a misspelling.
Use decode mode to verify any outside material you are studying. A training handout that mixes Juliet and Juliett, or that spells an entry as Xray without the hyphen, will trip the decoder's strict checks rather than silently passing. That makes the tool a small auditor as well as a learner — you can paste a row from a printed sheet and immediately see whether the sheet matches the ICAO / FAA standard.
What the Translator Will Not Do
The 26-letter spelling table is the entire factual scope of the tool, and the table below spells out how each character type is handled. Several things you might expect from a "phonetic alphabet" product are deliberately outside the scope, and recognizing them prevents misuse.
| Character type in the input | Encode behavior | Decode behavior |
|---|---|---|
| A–Z ASCII letter | Replaced with the official code word | Recognized case-insensitively, with the exact spelling required |
| Lowercase a–z | Same as uppercase; the spelling alphabet ignores case | Accepted as the letter |
| Digit (0–9) | Kept as a literal token | Kept as a literal token |
| Space, tab, line break | Each run collapsed to one slash word boundary | A slash between code words becomes one ordinary space |
| Literal slash | Escaped as [/] to mark that it is not a boundary | [/] returns the literal slash |
| Punctuation | Kept as a literal token | Kept as a literal token |
| Non-ASCII letter, accent, CJK, emoji | Kept as a literal token, not transliterated | Passed through unchanged on the encode-then-decode round trip |
| Unknown alphabetic token | N/A | Rejected with an explicit error (e.g., "Alpha", "Juliet") |
Stated plainly, the tool does not substitute aviation number pronunciations, does not produce spoken phraseology or pronunciation coaching, does not teach radio procedure or call-sign rules, and does not silently fix misspellings on decode. It also does not transmit, store, or upload your text: all tokenization, validation, mapping, and clipboard writing happen locally in your browser. You can study sensitive names, identifiers, or test material through the translator without those strings leaving your device. That confidence is also its limit: the translator knows exactly 26 rows and treats that knowledge literally.
Pairing the Tool With a Real Learning Routine
Learning a 26-row table is a memorization problem with a known shape, and the two directions of the translator cover both the encoding and the recognition half of the task. A routine that takes a week or two and sticks usually looks like this.
- Day 1–2, cold read. Read the table in order out loud twice, then in reverse. Cover the code-word column and recite from the letter alone. Repeat the same pass before bed.
- Day 3–5, encode practice. Pick five real items — your name, a street, a model number, a callsign, a memorable word — and encode each one in the translator. Read the result aloud once. The translator has done the lookup; your job is to internalize the mapping.
- Day 6–7, decode practice. Switch to decode mode and prepare code-word lists for yourself by hand. Mix in tricky rows like Alfa, Juliett, and X-ray. Run each list through the tool. Any decode error tells you which row to revisit.
- Week 2, audit mode. Find any third-party handout, cheat sheet, or website listing the alphabet, and decode a sample from it. Mismatches show up immediately, because the decoder rejects Alpha on principle and other misspellings as unknown tokens.
The pattern repeats itself in many language-learning tools: a reliable reference, a quick encoder to use what you know, and a strict decoder to verify what you have learned. The NATO Phonetic Alphabet Translator offers all three in one tab, with the additional benefit of a hard cap on input — 100,000 Unicode code points — that keeps tokenization, rendering, and clipboard payloads predictable. That bound is more than any human will paste into a spelling exercise, but it matters because predictable boundaries are exactly what a memorization tool needs.
If you study alongside someone else, take turns encoding and decoding for each other. The encoder reads the result aloud; the decoder writes the letters down, then runs the result through the tool to confirm. The strict decoder turns the exercise into a self-grading quiz rather than a discussion, which is the point. Once you can encode and decode A–Z under that loop without an error, you have learned the NATO phonetic alphabet the official way, and you can move on to whatever radio practice, call-sign drill, or classroom exercise your context actually calls for.
Related reading: Convert Numbers to Words Without Excel Code.