Japanese Romaji Converter

Convert hiragana/katakana to Hepburn, Nihon-shiki, or Kunrei-shiki romaji

Convert Japanese kana to Latin letters in the three standard romanization schemes — Hepburn, Nihon-shiki, and Kunrei-shiki — with correct yōon, sokuon doubling, and long-vowel handling. Compare schemes side by side, all in your browser. It runs free in your browser on Gera Tools, with nothing uploaded.

Last updated Source: Gera Tools

What is the difference between the three schemes?

Hepburn writes sounds as English speakers hear them (し = shi, ち = chi, つ = tsu). Nihon-shiki is the most systematic, mapping each kana by its grid position (し = si, ぢ = di). Kunrei-shiki is the Japanese government and ISO 3602 standard, a near-merge of Nihon-shiki that simplifies a few pairs (し = si, ぢ = zi).

Convert kana to romaji in three schemes

Romaji is not one system. This converter transliterates hiragana and katakana into the three standard schemes — Hepburn, Nihon-shiki, and Kunrei-shiki — with correct handling of yōon, sokuon, and long vowels, so you can pick the right one for documents, search, or linguistics.

How it works

Each kana (longest match first, so yōon like きゃ are caught) is looked up in the table for the chosen scheme, then two rules adjust the output:

っ + か  →  kka   (sokuon doubles the next consonant)
ー        →  repeats the previous vowel (コーヒー → koohii)
  • Multi-character kana are matched before single kana so combined syllables romanize as one unit.
  • The sokuon っ doubles the following consonant; before Hepburn ch it yields tch.
  • The long-vowel mark ー repeats the preceding vowel; unknown characters pass through unchanged.

Where the three schemes differ

The three romanization systems agree on most sounds but diverge on the irregular and voiced rows. Here are the key differences:

KanaHepburnNihon-shikiKunrei-shiki
shisisi
chititi
tsututu
fuhuhu
jizizi
jididi
zududu
しゃshasyasya
ちゃchatyatya
つぁtsatuatwa

Hepburn was designed by an American missionary in the 19th century and writes sounds the way an English speaker would naturally pronounce them. It is the most readable for Western audiences and is the basis for most printed phrasebooks, place-name signs in Japan, and passports.

Nihon-shiki is the most phonologically systematic — it maps the kana grid directly without irregularities. Each consonant column is consistent: sa si su se so, ta ti tu te to. Useful for linguistics but less intuitive for English readers.

Kunrei-shiki is standardised by the Japanese government (JIS X 4012) and is close to Nihon-shiki, collapsing a few distinctions (ぢ and じ both become zi). It is used in some official and educational contexts.

Practical guidance

  • Use Hepburn for place names, personal names in English documents, tourism copy, or any audience unfamiliar with Japanese.
  • Use Kunrei-shiki when submitting to a Japanese government body or following ISO 3602.
  • Use Nihon-shiki for academic linguistics work or when you want a perfectly systematic scheme.

きょうとし becomes kyoutoshi in Hepburn and kyoutosi in Kunrei-shiki (し = si). ふじさん is fujisan in Hepburn but huzisan in Nihon-shiki and Kunrei-shiki. All conversion runs in your browser with nothing uploaded.