Get your
Korean name.

It picks from a table — it does not transliterate

There are three ways a tool like this could turn your name into a Korean one: write the sound of it in Hangul, find characters that match its meaning, or choose from a list prepared in advance. This page does the third, with the second laid on top when the name you type is one it already knows. Nothing here sounds a name out — the "Write your own name in Hangul" link under the result goes to a different page, and that one does.

The lists are short and fixed: twenty surnames, and twenty-four syllables that given names are built from. A given name is always two of those twenty-four, and never the same one twice. So what comes back is one of the combinations we prepared beforehand, not a character invented on the spot.

The surname is drawn off the table mechanically and has nothing to do with the meaning of what you typed. If Kim or Lee or Park comes up, there is nothing in that to read.

What happens when your name is in the dictionary

There are two tables of what names are usually taken to mean: thirty-nine Japanese given names and sixty-three from other languages. Two of those sixty-three repeat a Japanese entry and are never reached, because the Japanese table is read first — so a hundred distinct names can actually be looked up. The Japanese entries can be reached in kana as well as in romaji, with thirty-nine kana spellings listed as keys and katakana folded to hiragana before the lookup, so サクラ and さくら land on the same entry.

Only the first word you type is looked up. The input is split on spaces and the first piece alone becomes the key, so a name typed surname-first is searched on the surname, is not found, and drops through to the route for names that are not in the dictionary — and typing the two run together makes no difference, because that is one word too. Type the given name on its own.

Kanji spellings are not looked up at all. The keys are thirty-nine romaji and thirty-nine kana, and there is no kanji key anywhere in the tables. A reading does not fix one spelling, and guessing wrong would be worse than not guessing, so kanji input matches nothing and is handled like a name that is not in the dictionary.

Each entry holds a one-line gloss of what the name means and three Korean syllables that suit it — three exactly, in every entry. The Japanese entries give the spelling the name is usually written with, the rest give an etymology. Yui's three are Yu, Eun and A. The two syllables of the name are taken from that shortlist, which is what makes the result come from the meaning rather than the sound.

Choosing a feminine or masculine style drops any of the three marked the other way. The mood does not narrow the three at all on this route; it only feeds the number. Only if the style empties the shortlist does the tool go back to the twenty-four — and even there it narrows by gender first and mood second, using the mood recorded against the name rather than the one you picked on screen, because the name is the more specific of the two.

Which parts are our choice

Each of the twenty-four syllables carries a meaning, but that meaning is our own one-line summary of how the hanja are commonly used. A row in that table holds five things — romaji, hangul, a gender mark, moods and that meaning — and there is no field for a character. What you see under a syllable is wording, not a character assigned to you.

The name glosses are written the same way. Twenty of the thirty-nine Japanese entries decline to fix a spelling — usually 結衣, usually 陽菜 — and the other nineteen name a character outright, 桜 or 蓮, which still does not make it the character in your name. Either way the spelling never enters the calculation: what decides the result is the three Korean syllables recorded beside it.

The twenty-four are each marked feminine, masculine or either. Feminine leaves eighteen candidates and masculine seventeen, while the neutral style does no narrowing at all and leaves all twenty-four standing. Neutral does not mean "only the neutral syllables". That behaviour is a decision of ours, and it is written here rather than left to be inferred.

Why the same input always gives the same name

Your name, the style and the mood are joined into a single string, that string is turned into a number by a fixed procedure, and the number decides where the tables are read. Nothing is rolled fresh each time, so the same three inputs return the same name for as long as the tables stand. Change only the mood and the number changes with it — though a different number can still land on the same syllable, and on the dictionary route, where the shortlist is only three long, the same name coming back is common rather than rare.

For a name that is not in the dictionary, the spelling itself is the material the number is made from. It is not being sounded out into Hangul, so changing one letter changes the number completely and a near spelling does not give a near name. It does not guarantee a different name either: with twenty surnames and twenty-four syllables, two spellings can land on the same one by coincidence. Capitals make no difference. A space before the name is dropped, but a space after it is not — the name, style and mood are joined first and trimmed second, so a trailing space ends up inside the string and changes the number.

Making the name finishes inside this page. There is no request to a server anywhere in the code that makes it, and no model is called; loading the page itself is another matter. Which said: this is a name for fun. It is not a name for documents, and since no birth date is taken it has nothing to do with the chart calculations elsewhere on this site.