How to use the furigana converter
- Paste or type Japanese into the box. Line breaks are kept, so you can paste a whole article, song lyrics or a page of a manga script.
- Read it in the Furigana tab. Small hiragana appear above each kanji word. The okurigana (the kana endings, like the べる in 食べる) are left alone, so you only see readings where you need them.
- Fix or hide readings. Tap any word to type the correct reading, or hide the furigana for kanji you already know. Your change applies to every place that word appears.
- Copy the format you need: pure hiragana, katakana, romaji, bracket notation for documents, or HTML
<ruby>tags for a website. You can also print a clean reading sheet.
What is furigana?
Furigana (振り仮名) are small kana written next to or above kanji to show how they are read. Japanese uses three scripts together: kanji carry most of the meaning, while hiragana and katakana are phonetic. Since a single kanji often has several readings, and the right one depends on the word it is in, even native speakers rely on furigana for rare words, names and children’s books.
That is also why a converter that looks up kanji one at a time gets things wrong. 生 is read せい in 学生 (gakusei, student), なま in 生ビール (nama bīru, draft beer) and う in 生まれる (umareru, to be born). This tool splits your text into words with a Japanese morphological analyser, then takes the reading for each whole word in context.
Hiragana, katakana or romaji: which output should you use?
| Output | Looks like | Best for |
|---|---|---|
| Furigana | 漢字を読む | Learning to recognise kanji while still being able to read the sentence. |
| Hiragana | かんじをよむ | You know kana but not the kanji yet, or you want text for a young reader. |
| Katakana | カンジヲヨム | Pronunciation notes, and checking how a text sounds without kanji. |
| Romaji | kanji o yomu | Complete beginners, singing along to lyrics, or showing pronunciation to someone who can’t read kana. |
| Brackets | 漢字（かんじ）を読（よ）む | Word, Google Docs, Anki cards and anywhere ruby text isn’t supported. |
How the romaji is written
The default is modified Hepburn, the system used by most textbooks and by the Japanese government for passports and signs:
- Particles are written as they sound: は → wa, へ → e, を → o.
- Long vowels get a macron: 東京 → Tōkyō, コーヒー → kōhī. Pick “Typed kana” for toukyou or “Plain” for Tokyo.
- Small っ doubles the next consonant (切符 → kippu, 抹茶 → matcha), and ん before a vowel takes an apostrophe (禁煙 → kin’en), so it isn’t mistaken for に or ね.
- Words are separated by spaces, while verb endings stay attached to their verb (ikimashō, matte imasu).
When the reading is wrong (and how to fix it)
Automatic readings are a very good first draft, but they are not perfect. These are the usual problems:
- Names. 中田 could be Nakata or Nakada, and many given names have several possible readings. Only the person knows. Always check names.
- Words with two valid readings. 今日 is usually きょう (today) but can be こんにち (nowadays). 日本 can be にほん or にっぽん. We default to the more common reading.
- Numbers and counters. Readings like 一本 (いっぽん), 三本 (さんぼん) and 六本 (ろっぽん) change with the number. Digits are left as digits.
- Very old or poetic text. Classical Japanese often uses readings that modern dictionaries don’t list.
Tap the word, type the right reading in hiragana, katakana or romaji, and press Save. Every output updates, romaji included.
Tips for learners
- Read with furigana, then hide it. Read a paragraph with full furigana, hide the readings of words you think you know, and read it again.
- Build flashcards fast. The Brackets output pastes straight into Anki or a spreadsheet.
- Print it. “Print reading sheet” leaves out everything except your text with furigana, in a large font for writing notes.
Frequently asked questions
Is this furigana converter free?
Yes. There is no sign-up and no length limit, and nothing you paste is uploaded. The conversion runs on your own device after a one-time dictionary download.
How accurate are the readings?
For everyday text, such as news, textbooks, menus and signs, most words come out right, because the tool reads whole words in context rather than looking up one kanji at a time. It gets names, place names, rare words and some number + counter combinations wrong more often. Kanji with more than one valid reading, like 今日 (きょう or こんにち), can also come out wrong. Tap any word to correct it, and your correction is applied everywhere that word appears.
Why is は written “wa” and を written “o” in the romaji?
When は, へ and を are used as particles they are pronounced wa, e and o, and Hepburn romaji spells them the way they sound. The converter checks the grammar of each word, so the particle は becomes “wa” while the は inside a word like はし (hashi) stays “ha”.
What is the difference between the romaji styles?
Hepburn with macrons (Tōkyō) is the standard in textbooks and dictionaries. “Typed kana” (toukyou) spells long vowels the way you type them on a Japanese keyboard, which is handy for IME practice and for searching. Plain (Tokyo) drops long-vowel marks completely, the way you see on signs and in English-language media.
Can I hide furigana for kanji I already know?
Yes. Tap a word and choose “Hide reading”. The word stays in kanji in the furigana view, in the bracket and HTML output, and in the hiragana output, so you get a graded-reader style text that only helps where you need it.
Can I put the furigana on my website or in Word?
Copy the HTML tab to get standard <ruby> tags, which every modern browser displays. For Word, Google Docs or plain-text notes, use the Brackets tab, which writes the reading in brackets after the kanji: 漢字（かんじ）.
Does it work offline or on my phone?
It works on any modern phone or computer browser. After the first visit the dictionary is cached, so later visits load almost instantly and conversion keeps working even on a weak connection.
Is my text private?
Yes. Your text is analysed by JavaScript inside your browser and is never sent to a server. It is saved only in your browser’s local storage so it is still there when you come back.