All processing runs in your browser — no files or inputs are uploaded to a server.
Two romanization systems for Korean coexist today. Revised Romanization (RR, 국립국어원) was adopted by the South Korean government in 2000 and replaced an earlier official system. McCune–Reischauer (MR), developed by two American academics in 1937, was the de facto standard for most of the 20th century and is still used by libraries, North Korea (with modifications), and most pre-2000 academic literature.
The two diverge in three visible ways. RR uses ASCII only; MR uses the diacritics ŏ and ŭ. RR uses plain letters for aspirated consonants (k, t, p, ch); MR marks them with an apostrophe (k', t', p', ch'). RR varies the spelling of a single Hangul letter by position (ㄱ is "g" syllable-initially, "k" syllable-finally); MR keeps a consistent letter (k) but the value depends on context too. The tool above lets you flip between them.
The 19 Hangul consonants and their mappings in both systems. The "initial" column applies when the consonant starts a syllable; the "final" column when it ends one. "—" indicates the position does not occur (the tense consonants ㄲ, ㄸ, ㅃ, ㅉ cannot stand finally on their own; only ㄲ does in some loanwords).
| Hangul | RR (initial) | RR (final) | MR (initial) | MR (final) | Example |
|---|---|---|---|---|---|
| ㄱ | g | k | k | k | 가구gagu / kagu |
| ㄲ | kk | k | kk | k | 꿈kkum / kkum |
| ㄴ | n | n | n | n | 나na / na |
| ㄷ | d | t | t | t | 다리dari / tari |
| ㄸ | tt | — | tt | — | 딸ttal / ttal |
| ㄹ | r | l | r | l | 라면ramyeon / ramyŏn |
| ㅁ | m | m | m | m | 말mal / mal |
| ㅂ | b | p | p | p | 바다bada / pada |
| ㅃ | pp | — | pp | — | 빵ppang / ppang |
| ㅅ | s | t | s | t | 사람saram / saram |
| ㅆ | ss | t | ss | t | 쌀ssal / ssal |
| ㅇ | — | ng | — | ng | 아이ai / ai |
| ㅈ | j | t | ch | t | 자전거jajeongeo / chajŏn'gŏ |
| ㅉ | jj | — | tch | — | 짜다jjada / tchada |
| ㅊ | ch | t | ch' | t | 차cha / ch'a |
| ㅋ | k | k | k' | k | 코ko / k'o |
| ㅌ | t | t | t' | t | 타다tada / t'ada |
| ㅍ | p | p | p' | p | 파도pado / p'ado |
| ㅎ | h | h | h | h | 하늘haneul / hanŭl |
The 21 Hangul vowels and diphthongs. Most map identically across both systems; the visible differences live on ㅓ, ㅕ, ㅝ, ㅡ, and ㅢ — RR spells them out in ASCII (eo, yeo, wo, eu, ui), MR uses the breve (ŏ, yŏ, wŏ, ŭ, ŭi).
| Hangul | RR | MR |
|---|---|---|
| ㅏ | a | a |
| ㅐ | ae | ae |
| ㅑ | ya | ya |
| ㅒ | yae | yae |
| ㅓ | eo | ŏ |
| ㅔ | e | e |
| ㅕ | yeo | yŏ |
| ㅖ | ye | ye |
| ㅗ | o | o |
| ㅘ | wa | wa |
| ㅙ | wae | wae |
| ㅚ | oe | oe |
| ㅛ | yo | yo |
| ㅜ | u | u |
| ㅝ | wo | wŏ |
| ㅞ | we | we |
| ㅟ | wi | wi |
| ㅠ | yu | yu |
| ㅡ | eu | ŭ |
| ㅢ | ui | ŭi |
| ㅣ | i | i |
The romanization of a Korean word does not come from a letter-by-letter substitution. The spelling represents the underlying form; pronunciation applies rules on top. Both RR and MR romanize the *pronunciation*, not the spelling, which is why 한국 (han-guk in transliteration) becomes Hanguk in RR and Han'guk in MR — and 신라 (sil-la in pronunciation) becomes Silla, not Sinra.
Four assimilation patterns cover most of the surprises. Liquid assimilation: ㄴ + ㄹ or ㄹ + ㄴ both become ll (신라 → Silla). Nasal assimilation: stops before nasals become nasals (국민 → gungmin, not gukmin). Tensification: certain consonant pairs tense the second (학교 → hakgyo in MR, hakgyo in RR — the second consonant is realized as ㄲ even though spelled ㄱ). Aspiration merging: ㅎ next to a plain stop merges into the aspirated form (좋다 → jota in RR, chot'a in MR). The tool above applies the common ones; the corner cases (place name conventions, hyphenation, given names) are codified in the official rules.
Use Revised Romanization for anything official in South Korea published after 2000: road signs, government documents, the official spellings of place names (Busan, Daegu, Incheon), South Korean passports issued through the standard process, and the Korean Wikipedia's convention. SEO and social-media transliterations of new Korean place names and brand names follow RR by default.
Use McCune–Reischauer for older literature (Pusan, Taegu, Inch'ŏn appear in books from before 2000), library catalogs (LC, OCLC still index under MR), most academic romanization in Korean studies outside the most recent decade, and North Korean place names where international convention still leans MR. Personal names are their own category — South Koreans choose their own romanization for passports, and family names rarely follow the rules exactly (이 is "Lee" for most, "Yi" in MR; 김 is almost always "Kim" even though strict RR would write "Gim").
Paste Korean text, pick a system (Revised Romanization — the South Korean government standard since 2000 — or McCune–Reischauer — the academic and pre-2000 standard, still used by libraries and North Korea), and the romanized output appears alongside. Consonant assimilation (the way `한국` becomes `Hanguk` rather than `Hangug` because the final `ㄱ` softens before the next syllable) is applied for the common cases each system specifies.
Reach for this when transliterating names for English correspondence, generating an ASCII version of a Korean article title, or filling out a passport-style romanization. Note that Korean *names* often follow personal preferences that diverge from official rules — `이` is officially `I` but most people romanize it as `Lee`, `김` is officially `Gim` but almost universally `Kim`. The tool gives the standard transliteration; for names, ask the person whose name it is.
system: Revised Romanization input: 서울 부산 인천 대구 광주 대전 울산 제주
Seoul Busan Incheon Daegu Gwangju Daejeon Ulsan Jeju
These are the official spellings on Korean road signs, passport names, and the .kr country domain. `부산` switched from `Pusan` (McCune–Reischauer) to `Busan` (Revised) in 2000, which is why older guidebooks still spell it the old way. The change was mostly cosmetic — same syllable, different convention for the initial `ㅂ` consonant.
system: McCune–Reischauer input: 부산 광주 평양
Pusan Kwangju P'yŏngyang
McCune–Reischauer (1939) uses apostrophes for aspirated consonants (`p'` for `ㅍ` vs `p` for `ㅂ`) and breves on vowels (`ŏ` for `ㅓ`). Academic Korean studies and older library catalogs still use it — Library of Congress kept it as the standard transliteration until 2008. North Korea continues to use McCune–Reischauer as its official system, which is why `평양` is `P'yŏngyang` in international press rather than `Pyeongyang`.
system: Revised Romanization input: 국립국어원 (the National Institute of Korean Language) literal letters: g+u+k + l+i+p + g+u+k + eo + w+o+n
Gungnipgugeowon (not "Guklipgugeowon")
When `ㄱ` (final consonant of one syllable) meets `ㄹ` (initial of the next), both shift: the `ㄱ` becomes `ng` (`ㅇ` sound) and the `ㄹ` becomes `n`. So `국립` is `Gungnip`, not `Guklip`. This is one of about a dozen assimilation rules Korean speakers apply unconsciously; the romanizer encodes the common ones. For pure letter-by-letter transliteration without phonetic assimilation, look for a "transliteration" mode or use a McCune–Reischauer strict variant.
Korean personal names follow individual preference, not the official Revised Romanization rules. The 2000 standard would render `김` as `Gim`, `이` as `I`, `박` as `Bak` — but only a small fraction of Koreans use these forms. `Kim` (∼95% of `김` bearers), `Lee` / `Yi` / `Rhee` for `이`, `Park` for `박` are entrenched through generations of passports and English correspondence. The Korean passport office accepts any spelling the applicant requests for the first time, then locks it in for life. The romanizer applies the standard rule; for legal name spelling, use whatever the passport says.
They are both correct in different contexts. **Revised Romanization** (2000) is what South Korea uses for road signs, government documents, the `.kr` domain, and most modern English-language Korean content. **McCune–Reischauer** (1939) is what academic Korean studies, libraries, and pre-2000 publications use; it represents Korean phonology more precisely (apostrophes for aspirated consonants, breves for distinct vowel qualities) at the cost of needing special characters. North Korea uses McCune–Reischauer as official, which is why international press writes `P'yŏngyang` rather than `Pyeongyang`. For new web content targeting English readers, use Revised; for academic / library work, use McCune–Reischauer.
No — the tool romanizes the Hangul phonetic spelling, regardless of whether the word came from native Korean, Sino-Korean (Chinese-origin), or English loanwords. So `학교` (school, Sino-Korean) becomes `hakgyo` from its Hangul pronunciation, not from the Chinese reading `xuéxiào`. This is by design: modern Korean is written and read in Hangul; the underlying etymology only matters for specialist work. If you need the Hanja (Chinese-character) form for a name, look it up in a name dictionary separately.
Three reasons. **Different standards** — `Naver` dictionary often shows Revised; `LingoDeer` and academic resources may default to McCune–Reischauer; older `Wiktionary` entries can be either depending on contributor. **Romanization vs transliteration** — some tools apply phonetic assimilation rules (Revised does), others give letter-by-letter (`Yale Romanization` does this strictly). **Personal-name override** — bilingual sites often use the name's self-chosen spelling instead of the standard. Always check which system a dictionary or article is using before comparing.
No, and that is intentional. Both Revised and McCune–Reischauer are phonetic romanizations — they encode how the word *sounds*, not which exact Hangul letters spell it. Multiple Hangul spellings can produce the same romanization (e.g., `남녀` and `남여` both become `namnyeo`), so the reverse mapping is ambiguous. For strict letter-by-letter conversion that round-trips losslessly, use **Yale Romanization** (1942, academic linguistics standard) — it represents each jamo distinctly. Yale is unsuited for casual reading but ideal for linguistic data interchange.
URL Slug runs a romanizer as one step, then applies additional URL-safety rules — strips punctuation, replaces spaces with hyphens, lowercases everything, optionally truncates to a max length. Korean Romanizer is the underlying transliteration step alone: it preserves the original spaces, punctuation, and capitalization, and emits multi-word output. If you want a clean URL slug, use the URL Slug tool; if you want the readable romanization as text for a heading, citation, or correspondence, use this one.
Korean romanization is the problem of representing Hangul, an alphabetic script designed in 1443 specifically for Korean, in the Latin alphabet used by English and most international communication. The two major standards are **Revised Romanization** (개정 로마자 표기법, 2000) issued by South Korea's National Institute of Korean Language as the official replacement for the government's 1959 system, and **McCune–Reischauer** (1939) created by two American academics and adopted by libraries, North Korea, and the Library of Congress (until 2008). A third, **Yale Romanization** (1942), is used in academic linguistics for its strict 1-to-1 jamo mapping.
The core challenge is **phonetic assimilation**: Hangul is written by stacking jamo into syllabic blocks, but spoken Korean modifies sounds at syllable boundaries. `한국` is written `ㅎㅏㄴ ㄱㅜㄱ` (`Han + Guk`) but pronounced (and romanized) as `Hanguk`, with the final `ㄱ` of `한` softening into the initial `ㄱ` of `국`. There are about a dozen common assimilation rules — final-consonant softening, ㄹ becomes ㄴ after certain consonants, palatalization before ㅣ — and a romanizer either applies them (phonetic, like Revised) or ignores them (transliteration, like Yale). The choice fundamentally changes the output.
Three adjacent ideas matter. **IDN / Punycode for `.kr` URLs** (`한국.kr` → `xn--3e0b707e.kr`) handles non-ASCII in domain names without going through romanization — different problem, different solution. **Yale Romanization** is the linguistic-grade alternative when round-tripping matters; libraries used to combine McCune–Reischauer for cataloging with Yale for textual analysis. **Names follow their own rules**: Korean immigrants in the early 20th century established `Kim` (instead of `Gim`), `Lee` (instead of `I`), `Park` (instead of `Bak`) through passports and ship manifests, and personal preference now overrides the standard. The romanizer gives you the rule-based answer; the human in the room gives you the right one for their name.
Convert Korean addresses to English form for international shipping, visas, and forms. Built-in city/province table plus auto-romanization of road and district names.
Count characters, words, lines, and bytes — including a UTF-8 byte length for non-ASCII.
Convert text between camelCase, PascalCase, snake_case, kebab-case, CONSTANT_CASE, and Title Case.
Generate placeholder text by words, sentences, or paragraphs. Choose to start with the classic "Lorem ipsum…".
Common assimilations (liaison, nasalization, lateralization) are applied. Specialist work — proper nouns, contested boundaries — should be cross-checked against the official guidelines.
