# Furiganascan

> Japanese reading tools for learners. Upload a screenshot of manga, a game or a video frame and get the text, furigana, a word table, grammar notes and an English translation. Free converters for furigana, hiragana, romaji and verb forms run in the browser.

# Japanese text from image, with furigana and meanings

https://furiganascan.com/

Drop a manga page, game screenshot or video frame. Get the text, readings, word meanings and a translation.

## Your screenshot

or drop an image here, or press Ctrl+V to paste one. PNG, JPEG or WebP. Large images are shrunk in your browser before they are sent.

3 free scans a day

## ReadingExample

### Japanese text

今日は天気がいいですね。

### With furigana

今日きょうは天気てんきがいいですね。

### Words

**Words in the example**

| Word | Reading | Romaji | Part of speech | Meaning |
| --- | --- | --- | --- | --- |
| 今日 | きょう | kyō | noun | today |
| 天気 | てんき | tenki | noun | weather |
| いい | いい | ii | adjective | good |

### Grammar

**〜ですね**

Polite ending です plus ね, which asks the listener to agree: "..., isn't it?"

### English

Nice weather today, isn't it?

## Short answer

To get Japanese text from an image, upload or paste a screenshot above. The tool reads the lines in order, adds furigana over the kanji, lists every word with its reading and English meaning, explains one to three grammar points, and translates each sentence. You get three free scans per day.

## Ways to read Japanese from a screenshot

Each approach suits a different job. Pick the one that matches what you need back.

| Approach | Gives you text | Adds furigana | Explains words and grammar | Cost |
| --- | --- | --- | --- | --- |
| This tool | Yes, lines in reading order | Yes, with a romaji toggle | Yes, word table, grammar notes and translation | 3 free scans per day |
| Generic OCR websites | Yes, plain text | No | No | Usually free, limits vary |
| Google Lens | Yes, selectable text and translation | No | Translation only | Free |
| Browser dictionary extension | Only if the page has real text, not pictures | Some add it for page text | Pop-up dictionary lookups | Free |
| Paste text into a converter | You must type or copy it first | Yes | Depends on the site | Free converters on this site |

## How to read Japanese text in a screenshot

1. **Add the image.** Click the drop area, drag a file onto it, or press Ctrl+V (Cmd+V on a Mac) to paste a screenshot. PNG, JPEG and WebP work.
2. **Let the browser shrink it.** The image is scaled down to 1600 px on its longest side and re-encoded before upload, so large screenshots stay under the 4 MB limit.
3. **Run the scan.** Press the scan button. Results usually appear within a few seconds. One scan is taken from your three free scans for the day.
4. **Read and check.** Read the original lines, switch furigana and romaji, and look through the word table and grammar notes. Compare anything odd with the picture.
5. **Save the words.** Press Copy as flashcards to put a tab-separated list on your clipboard, or Download .tsv to save it as a file.

## What you get from one scan

A scan returns six things in one result:

- The Japanese text, line by line in reading order.
- The same text with furigana written over the kanji. A switch adds romaji.
- A word table with the surface form, the hiragana reading, romaji, part of speech and a short English meaning.
- One to three grammar notes that name the pattern and explain it in plain English.
- A natural English translation, one line per sentence.
- A flashcard export you can import into a spaced repetition app such as Anki.

Everything on screen is plain text that you can select and copy.

## Japanese screenshot to furigana: what works best

Clear horizontal text on a plain background gives the best results. Dialogue boxes in games, subtitles and printed pages usually read well. Vertical manga text, handwriting, decorative title fonts and text over busy artwork are harder, so crop to the part you want to read before you scan.

The scan is done by an AI vision model, so it can misread a character or give a reading that does not fit the context. Treat the result as a study aid and check it against the picture. If you only need the text itself, the [Japanese OCR page](https://furiganascan.com/japanese-ocr/) returns plain lines.

## Read Japanese from a screenshot without a scan

If you already have the text, you do not need a scan. Paste it into the [furigana converter](https://furiganascan.com/furigana-converter/), the [kanji to hiragana converter](https://furiganascan.com/kanji-to-hiragana/) or the [romaji converter](https://furiganascan.com/japanese-to-romaji/). They run in your browser and have no daily limit. The [sentence breakdown](https://furiganascan.com/sentence-breakdown/) splits a sentence into words for free and only uses a scan when you ask for explanations.

## Limits and privacy

Each visitor gets 3 free scans per day. The count resets at 00:00 UTC. The image goes to an AI model provider through OpenRouter to produce the result and is not stored by this site. Read the [privacy page](https://furiganascan.com/privacy/) for exactly what is logged and the [pricing page](https://furiganascan.com/pricing/) for what is planned.

## Frequently asked questions

**Is it free to get Japanese text from an image here?**

Yes. Each visitor gets 3 free scans per day, shared by the screenshot scan, the OCR page and the sentence explanations. The converters on this site run in your browser and have no limit. Paid packs are planned but are not on sale yet.

**Can it read manga?**

It can read horizontal speech bubbles with clear print, and often vertical text too, but small, stylized or hand-lettered text is harder. Crop to one panel or bubble for better results and always compare the output with the page.

**Does it store my screenshots?**

No. The image is processed in memory and is not stored by this site. It is sent to an AI model provider through OpenRouter to produce the result. The privacy page lists what is logged, which does not include the image or the text.

**How accurate are the readings and translations?**

They are good for common words and clear text but can be wrong, especially for names, slang, jokes and context-dependent readings. Check anything important in a dictionary. The result is meant for study, not for official translation.

**How do I use the flashcard export?**

Press Copy as flashcards or Download .tsv. Each line has the word on the front and the reading plus meaning on the back, separated by a tab. Most spaced repetition apps, including Anki, can import tab-separated text.

**What image types and sizes work?**

PNG, JPEG and WebP. Your browser scales the image to at most 1600 px on the longest side and re-encodes it, and the upload must be 4 MB or less after that. Other files are rejected before upload.

Last updated: October 23, 2026

# About Furiganascan

https://furiganascan.com/about/

## Short answer

Furiganascan is an independent personal project that helps Japanese learners read text from manga, games and video frames. A screenshot scan returns the text with furigana, word meanings, grammar notes and a translation. Browser converters for furigana, hiragana, romaji and verb forms are free and have no limit.

## What it does

The site has two parts. The first is a screenshot scan. You upload or paste an image, and an AI vision model reads the Japanese text, then returns the lines, furigana, a table of words with readings and meanings, a few grammar notes and an English translation. The second part is a set of converters that run entirely in your browser: [furigana](https://furiganascan.com/furigana-converter/), [kanji to hiragana](https://furiganascan.com/kanji-to-hiragana/), [Japanese to romaji](https://furiganascan.com/japanese-to-romaji/), [romaji to hiragana](https://furiganascan.com/romaji-to-hiragana/), [hiragana to romaji](https://furiganascan.com/hiragana-to-romaji/), a [verb conjugator](https://furiganascan.com/japanese-verb-conjugator/) and a [sentence breakdown](https://furiganascan.com/sentence-breakdown/).

## How it works

Scans are sent to AI vision models through OpenRouter. Images are shrunk in your browser first, processed in memory and not stored. The converters use kuromoji, an open-source Japanese morphological analyzer licensed under Apache-2.0, together with its dictionary derived from IPADIC. The dictionary is about 17 MB and is downloaded once and cached by your browser. Romaji, kana and conjugation rules are our own code. The daily limit and the check against bots are handled by a small server on Cloudflare. The [privacy page](https://furiganascan.com/privacy/) describes exactly what is logged.

## Who runs it

Furiganascan is an independent personal project, built and maintained by one person with an interest in tools for language learners. It is not affiliated with any publisher, game studio, dictionary or AI company. There is no team page because there is no team.

## Limits you should know

The AI scan can make mistakes with small, stylized or vertical text, and readings of names can be wrong in both the scan and the converters. Treat results as study aids. Free use is 3 scans a day per visitor, see [pricing](https://furiganascan.com/pricing/). To reach us, see [contact](https://furiganascan.com/contact/).

Last updated: October 23, 2026

# Furiganascan for AI agents

https://furiganascan.com/agents/

## Short answer

Furiganascan is a Japanese reading site for learners. It turns screenshots into text with furigana and has browser converters for furigana, hiragana, romaji and verb forms. There is no public API. The scan endpoint is for the site's own pages only. Pages are plain HTML, and llms.txt and the sitemap list them.

## What to cite

Each page opens with a short direct answer. The main pages are the [screenshot scan](https://furiganascan.com/), [Japanese OCR](https://furiganascan.com/japanese-ocr/), [furigana converter](https://furiganascan.com/furigana-converter/), [kanji to hiragana](https://furiganascan.com/kanji-to-hiragana/), [Japanese to romaji](https://furiganascan.com/japanese-to-romaji/), [romaji to hiragana](https://furiganascan.com/romaji-to-hiragana/), [hiragana to romaji](https://furiganascan.com/hiragana-to-romaji/), [verb conjugator](https://furiganascan.com/japanese-verb-conjugator/) and [sentence breakdown](https://furiganascan.com/sentence-breakdown/). Reference answers are on [what is furigana](https://furiganascan.com/what-is-furigana/) and [how to read Japanese text in an image](https://furiganascan.com/how-to-read-japanese-text-in-an-image/).

## Machine-readable files

A summary is at [/llms.txt](https://furiganascan.com/llms.txt), the full text of all pages at [/llms-full.txt](https://furiganascan.com/llms-full.txt) and the page list at [/sitemap.xml](https://furiganascan.com/sitemap.xml). Each page also has a Markdown twin at its path followed by index.md, or by asking for text/markdown.

## Pricing and contact

Use is free with 3 scans per visitor per day. Paid packs are not on sale. See [pricing](https://furiganascan.com/pricing/). For questions write to [support@furiganascan.com](mailto:support@furiganascan.com).

Last updated: October 23, 2026

# Contact Furiganascan

https://furiganascan.com/contact/

## Short answer

Email support@furiganascan.com for bug reports, feedback, privacy questions or anything about the site. There is no contact form. Include the page address, what you did and what you expected, and say which browser you use. Please do not attach private images. Replies come from one person, so allow a few days.

## What to include

Write to [support@furiganascan.com](mailto:support@furiganascan.com). A useful message names the page, what you entered, what you saw and what you expected. For a wrong reading, give the word and the reading you think is right. For a scan problem, describe the image rather than attaching private content.

## Before you write

Many questions are answered on the [pricing page](https://furiganascan.com/pricing/) (limits and packs), the [privacy page](https://furiganascan.com/privacy/) (what is stored) and the [terms](https://furiganascan.com/terms/).

Last updated: October 23, 2026

# Furigana converter for any Japanese text

https://furiganascan.com/furigana-converter/

Paste Japanese text and get furigana over every kanji. Copy it as HTML ruby, bracket notation or hiragana.

## Japanese text

Text with kanji

## Furigana

The text with furigana appears here.

### Copy as

Converted output

## Short answer

A furigana converter adds the reading in small kana above each kanji. Paste Japanese text into the box above and this page marks it up in your browser. Copy the result as HTML ruby tags, as kanji followed by bracketed hiragana, or as plain hiragana. It is free and has no daily limit.

## Furigana output formats

Pick the format that matches where the text will be pasted.

| Format | Example for 日本語 | Use it for |
| --- | --- | --- |
| HTML ruby | \<ruby>日本語\<rt>にほんご\</rt>\</ruby> | Web pages, blogs, course material |
| Brackets | 日本語(にほんご) | Chat, notes, plain text files, flashcards |
| Plain hiragana | にほんご | Reading practice, checking pronunciation |

## How to add furigana to Japanese text

1. **Paste the text.** Type or paste Japanese into the box. Long passages work, but split very large documents into parts.
2. **Wait for the dictionary on first use.** The first conversion downloads a dictionary of about 17 MB. Your browser keeps it, so later visits start quickly.
3. **Check the result.** Furigana appears over the kanji. Names and unusual readings are the most likely places for mistakes.
4. **Copy it.** Choose HTML ruby, brackets or hiragana, then press the matching copy button.

## How the readings are chosen

The converter splits your text into words with the kuromoji analyzer and looks up each word in its dictionary. It then lines the reading up with the kanji so that okurigana, the kana that follow a kanji stem, stay unmarked. In 食べる only 食 gets 「た」, not べる.

Because the choice is made word by word without understanding the whole sentence, some readings come out wrong. Personal names, place names, and kanji with several readings such as 生 and 行 are the usual trouble. For text where accuracy matters, check the result against a dictionary. The [furigana explainer](https://furiganascan.com/what-is-furigana/) covers how the notation works.

## Using the HTML ruby output

The ruby output uses the standard `ruby` and `rt` elements, which all current browsers display. Paste it into your page source as is. If you want to hide the readings until hovered, that is a CSS choice on your side.

## Related tools

For full conversion with no kanji left, use [kanji to hiragana](https://furiganascan.com/kanji-to-hiragana/). For Latin letters, use [Japanese to romaji](https://furiganascan.com/japanese-to-romaji/). To pull text out of a picture first, try the [screenshot scan](https://furiganascan.com/).

## Frequently asked questions

**Is the furigana converter free?**

Yes. It runs entirely in your browser, has no daily limit, and does not use the scan quota. Your text is not sent to a server for this tool.

**Why does it download a dictionary?**

The word list that makes accurate readings possible is about 17 MB. It downloads once on first use and your browser caches it, so later visits do not repeat the download.

**Can I add furigana to only some kanji?**

Not in this version. The converter marks every kanji word. Delete the ruby tags you do not want from the copied HTML, or use the brackets format and edit by hand.

**Why is a reading wrong?**

The analyzer picks readings word by word, so names, rare compounds and words with several readings can be wrong. Check unusual words in a dictionary before you rely on them.

**Does it work offline?**

After the dictionary has been cached, the converter works without a connection as long as the page itself is loaded. The first visit needs the network. It is a small script on the page, so after the first load you can even disconnect and keep converting kana without any problems.

Last updated: October 23, 2026

# Hiragana to romaji converter

https://furiganascan.com/hiragana-to-romaji/

Paste hiragana or katakana and get Hepburn romaji straight away.

## Hiragana or katakana

Kana

Macrons (ō, ū)Plain letters (ou, uu)

## Romaji

Romaji output

## Short answer

To convert hiragana to romaji, paste your kana into the box above. The converter writes each syllable in Hepburn romanization, so ありがとう becomes arigatou, or arigatō with macrons switched on. Katakana input works the same way. It runs in your browser and, because it needs no dictionary, it starts instantly.

## Kana to romaji examples

These cases show how the rules apply.

| Kana | Romaji | Rule |
| --- | --- | --- |
| しんぶん | shinbun | Plain ん is n |
| しんよう | shin'yō | n' before a vowel or y |
| がっこう | gakkō | Small っ doubles the consonant |
| マッチャ | matcha | tch before ch |
| コーヒー | kōhī | Katakana long mark gives a macron |
| わたしは | watashi ha | Kana by kana: は stays ha, see the note below |

## How to convert hiragana to romaji

1. **Paste kana.** Enter hiragana, katakana or a mix. Kanji are not converted on this page.
2. **Choose the vowel style.** Macrons give tōkyō. Plain vowels give toukyou.
3. **Copy the result.** Press Copy romaji.

## This page converts kana one by one

Because this tool only sees kana, it cannot tell whether は is the particle wa or part of a word. It writes ha. If your text has kanji or you want particles read as spoken, use [Japanese to romaji](https://furiganascan.com/japanese-to-romaji/), which analyzes the sentence first.

## Reading the output

Hepburn is built for English readers: sh, ch and ts stand for sounds that have single kana. Long vowels are written as ā, ī, ū, ē, ō or by repeating the vowel. The pair おう is usually ō, and おお is also ō, though some words keep ou or oo. To go the other way and type kana, use [romaji to hiragana](https://furiganascan.com/romaji-to-hiragana/).

## A short kana refresher

Hiragana is used for native words and grammar. Katakana is used for foreign loanwords, emphasis and sound effects. Both have the same 46 basic sounds, plus dakuten, handakuten and small kana. If you are learning to read them, a good first step is to write the romaji for a word and then check it against the kana. The [furigana explainer](https://furiganascan.com/what-is-furigana/) shows how kana help with kanji.

## Frequently asked questions

**Does it convert katakana to romaji?**

Yes. Katakana is accepted, including the long vowel mark, which becomes a macron or a repeated vowel depending on your setting. For example ラーメン becomes rāmen with macrons or raamen with plain vowels, and the small ッ doubles the following consonant.

**Why is は written ha and not wa?**

This page works from the spelling of each kana and does not parse the sentence. Use the Japanese to romaji page if you want particles written as they sound.

**Can I convert kanji here?**

No. Use the Japanese to romaji or the kanji to hiragana page, which load a dictionary to read kanji. Both of those pages read kanji with the dictionary, then give hiragana or romaji, so use them for ordinary Japanese text.

**Does it work offline?**

Once the page has loaded, yes. This tool needs no dictionary or network. It is a small script on the page, so after the first load you can even disconnect and keep converting kana without any problems.

**Which spelling does it use?**

Modified Hepburn: shi, chi, tsu, fu, ji. That is the style most learners meet in textbooks and dictionaries. The macron option follows the textbook form, while the plain option writes long vowels as they are spelled in kana.

Last updated: October 23, 2026

# How to read Japanese text in an image

https://furiganascan.com/how-to-read-japanese-text-in-an-image/

## Short answer

To read Japanese text in an image, first get the text out with OCR, then add readings and meanings. Google Lens copies and translates text. A screenshot scan reads the text and adds furigana, a word table and grammar notes. Pop-up dictionaries and ichi.moe help once you have the text.

## Tools for reading Japanese in a picture

This comparison is based on how each tool is designed to work. Features change, so check each tool for its current behavior.

| Tool | Takes an image | Gives text | Reading help for learners |
| --- | --- | --- | --- |
| Screenshot scan on this site | Yes | Yes | Furigana, romaji, word table, grammar notes, translation |
| Google Lens | Yes, on phone and desktop | Yes, copy or translate | Translation and text copy. No furigana layer |
| Yomitan or similar browser extension | No. It works on real text on a web page | Only if text exists | Pop-up dictionary on hover |
| ichi.moe | No. Paste text | No | Breaks a sentence into words with readings and meanings |
| Plain OCR websites | Yes | Yes | None, text only |

## Step by step: from picture to understanding

1. **Take a clean screenshot.** Capture only the text you want. Crop out artwork. A larger, straighter image reads better.
2. **Get the text.** Use OCR. The Japanese OCR page returns plain lines. The main scan also adds readings.
3. **Add readings.** If you used plain OCR, paste the text into the furigana converter to see how each kanji is read.
4. **Look up unknown words.** Use the sentence breakdown for dictionary forms, or the scan's word table for meanings.
5. **Verify.** Compare the result with the image. Models and OCR engines both make mistakes, and similar kanji are easy to confuse.

## Why a picture is harder than text

A pop-up dictionary can only read text that exists on a page. When the words are pixels in a manga page, a game frame or a video, there is nothing to select. OCR has to turn the pixels into characters first. Japanese is harder than most languages for this: thousands of kanji, small kana like ゃ and ッ, long vowel marks and vertical layout.

## What each tool is good at

Google Lens is quick for a one-off look-up on your phone. It gives you the text and a translation but is not built around learning each word. A browser extension such as Yomitan is great for web pages and e-books where text is real. ichi.moe is a useful text analyzer when you already have a sentence. Plain OCR sites give text and little else.

This site's [screenshot scan](https://furiganascan.com/) sits between these. It takes the picture, then gives the lines with furigana, a per-word table, grammar notes and a translation. It is limited to 3 free scans a day, and the output is produced by an AI model, so it can be wrong.

## Tips for manga, games and video

- Manga: scan one bubble or panel at a time. Vertical text is harder than horizontal.
- Games: dialogue boxes are usually clean. Turn off any effects that cover the text.
- Video: pause on a clear frame. Hard-coded subtitles scan better than stylized titles.
- Photos of signs: hold the camera square to the sign, avoid glare.

If you only want the text, see the [Japanese OCR page](https://furiganascan.com/japanese-ocr/). To learn what furigana are, read [what is furigana](https://furiganascan.com/what-is-furigana/).

## Frequently asked questions

**Can Google Lens read Japanese?**

Yes, it can recognize Japanese text in photos and screenshots, copy it and translate it. It is aimed at quick look-ups rather than study, so it does not show furigana for each word.

**Can a dictionary extension read text in an image?**

Not by itself. Extensions such as Yomitan work on selectable text. For images you first need OCR to produce text, or a tool built to overlay recognized text on the picture.

**What is the best way to get furigana from a screenshot?**

Run OCR and add readings in one step. The screenshot scan on this site does that, or run plain OCR and then paste the text into a furigana converter. Check unusual readings by hand.

**Why does OCR confuse similar characters?**

Many kanji and kana differ by one stroke, such as シ and ツ or 末 and 未. Low resolution and decorative fonts make it worse. Crop closely and proofread.

**Is it safe to upload screenshots?**

Only upload what you are comfortable sending to an AI service. On this site images are processed in memory and not stored, but they pass to an AI model provider. See the privacy page for the details.

Last updated: October 23, 2026

# Japanese OCR for screenshots and photos

https://furiganascan.com/japanese-ocr/

Upload or paste an image and copy the Japanese text as plain lines.

## Your screenshot

or drop an image here, or press Ctrl+V to paste one. PNG, JPEG or WebP. Large images are shrunk in your browser before they are sent.

3 free scans a day

## Text

Extracted Japanese text

## Short answer

Japanese OCR turns the writing in a picture into text you can copy. Upload or paste an image above and this page returns the Japanese lines in reading order, with no extra analysis. Each OCR run uses one of your three free daily scans, shared with the main screenshot tool.

## Japanese OCR options compared

The right choice depends on the source image and on what you do with the text afterwards.

| Option | Output | Best for | Watch out for |
| --- | --- | --- | --- |
| This page | Plain Japanese lines | Screenshots, game text, printed pages | 3 free scans per day, shared quota |
| The main scan tool | Text plus furigana, word table, grammar, translation | Study and reading practice | Same daily quota |
| Generic OCR websites | Plain text | Scanned documents in many languages | Japanese accuracy varies, ads and limits differ |
| Google Lens | Selectable text and translation | Quick look-ups on a phone | No reading aids for learners |
| Phone camera text selection | Selectable text in the photo app | Signs and menus | Depends on the phone and OS version |

## How to run Japanese OCR on an image

1. **Add the image.** Drop a file on the box, choose one with the button, or press Ctrl+V to paste a screenshot from the clipboard.
2. **Run OCR.** Press the scan button. The browser shrinks the image first, which keeps uploads small.
3. **Copy the text.** Press Copy text to put every line on the clipboard. You can also select part of the result by hand.
4. **Proofread.** Compare the lines with the picture, especially similar kanji and small kana, and fix any slips.

## What this OCR does and does not do

This page returns text only. It does not add readings, meanings or a translation. That keeps the request small and fast. If you want those, use the [screenshot scan on the home page](https://furiganascan.com/), which costs the same single scan.

The text is read by an AI vision model, not by a classic OCR engine. That helps with game fonts and mixed backgrounds, but the model can also invent or skip characters. Always proofread.

## Tips for better Japanese OCR

- Crop to the text. Less artwork around it means fewer mistakes.
- Use a straight, sharp image. Blur and strong skew hurt small kana.
- Vertical text is supported less reliably than horizontal text.
- Look closely at pairs such as ソ and ン, シ and ツ, and 未 and 末. These are common slips.
- Furigana printed beside kanji may be read as extra text. Delete it if it appears.

## When you already have the text

Once you have the lines, paste them into the [furigana converter](https://furiganascan.com/furigana-converter/) to add readings for free. The converters run in your browser and do not use a scan.

## Frequently asked questions

**Is this Japanese OCR free?**

Yes, up to 3 scans per day per visitor. OCR runs, screenshot scans and sentence explanations share that quota. Paid packs are planned but not on sale yet.

**Does it work on handwriting?**

Neat handwriting sometimes works, but results are unreliable. Print, subtitles and game text are much better. Check the output against the image either way. Block capitals written slowly do better than cursive or fast notes.

**Is the image kept?**

No. It is processed in memory and not stored here. It is sent to an AI model provider through OpenRouter to read the text. The privacy page lists what is logged.

**What file types can I upload?**

PNG, JPEG and WebP. Your browser scales the image to at most 1600 px on its longest side and the upload must be 4 MB or smaller after that.

**Why did the OCR miss some text?**

Small, stylized or low contrast text is the usual cause. Crop closer, use a larger screenshot, or try again with a cleaner capture. A failed scan that returns an error does not use up your daily quota.

Last updated: October 23, 2026

# Japanese to romaji converter

https://furiganascan.com/japanese-to-romaji/

Paste Japanese text and get Hepburn romaji, with word spacing and a choice of long vowel style.

## Japanese text

Japanese text

Put spaces between wordsMacrons (ō, ū)Plain letters (ou, uu)

## Romaji

Romaji output

## Short answer

To convert Japanese to romaji, paste kanji, hiragana or katakana into the box above. The converter finds each word, reads it, and writes it in Hepburn romanization. You can switch word spacing on or off and choose macrons such as ō or plain vowels such as ou. It runs in your browser with no limit.

## Romanization systems compared

This page writes modified Hepburn. Other systems exist.

| System | し | ち | つ | Where it is used |
| --- | --- | --- | --- | --- |
| Hepburn | shi | chi | tsu | Textbooks, dictionaries, signs, passports |
| Kunrei-shiki | si | ti | tu | Japanese government standard, some school use |
| Nihon-shiki | si | ti | tu | Linguistics, strictly kana by kana |

## How to convert Japanese to romaji

1. **Paste the text.** Kanji, hiragana and katakana can be mixed.
2. **Choose the options.** Turn word spacing on for readable output. Choose macrons for textbook style or plain vowels for easy typing.
3. **Copy the romaji.** Press Copy romaji. The first run downloads about 17 MB of dictionary data, which your browser then keeps.

## Particles are written as they sound

Three particles are spelled with one kana but pronounced differently. は is written wa, へ is written e, and を is written o. This converter checks the part of speech, so 私は is watashi wa while はな stays hana. Reading kana one by one without that check gives wrong results such as watashi ha.

## Long vowels: macrons or plain

Hepburn marks long vowels with a macron: tōkyō, yūbin, onēsan. Many people cannot type those, so the plain option writes the vowels out as spelled in kana: toukyou, yuubin, oneesan. Both are readable, but pick one and stay consistent. Names on official documents often drop the mark altogether, giving Tokyo.

## Other details

A small っ doubles the next consonant, so きって is kitte, with tch before ch: まっちゃ is matcha. The kana ん is written n, and n' when it comes before a vowel or y, as in しんよう, shin'yō. Readings of kanji come from the dictionary, so the notes about wrong readings on the [kanji to hiragana page](https://furiganascan.com/kanji-to-hiragana/) apply here too.

If you already have kana, use [hiragana to romaji](https://furiganascan.com/hiragana-to-romaji/). To type Japanese from Latin letters, use [romaji to hiragana](https://furiganascan.com/romaji-to-hiragana/).

## Frequently asked questions

**Which romanization does it use?**

Modified Hepburn, the style used in most textbooks and dictionaries. し is shi, ち is chi, つ is tsu, and long vowels use macrons or doubled vowels depending on your setting.

**Why use spaced words?**

Japanese text has no spaces, so unspaced romaji is hard to read. The converter splits words with a dictionary and puts one space between them. Turn it off if you want a continuous string.

**Is it free and private?**

Yes. It runs in your browser, uses no scan quota, and your text is not uploaded. There is no sign-up, no daily limit, and nothing you type is stored on a server or logged anywhere.

**Can it romanize names correctly?**

Common names usually work, but the reading of a kanji name can differ from the dictionary guess. If the name's reading matters, enter it in kana instead.

**What about Chinese or Korean text?**

It is made for Japanese. Hanzi or Hangul will not be converted correctly. Chinese characters shared with Japanese may convert, but the readings will be Japanese ones, which is rarely what you want.

Last updated: October 23, 2026

# Japanese verb conjugator, all common forms

https://furiganascan.com/japanese-verb-conjugator/

Enter a verb in dictionary form, in kanji or kana, and see every common form with its verb type.

## Verb

Dictionary form (kanji or kana)

Verb group

**Godan verbs that look like ichidan**

## Forms

Enter a verb in dictionary form. You get the group and every form below.

## Short answer

Enter a Japanese verb in dictionary form, in kanji or kana, and this conjugator shows its type and every common form: polite, negative, past, te, potential, passive, causative, imperative, volitional, conditional and tai. It handles godan, ichidan, する, 来る, 行く and honorific verbs. Japanese conjugation here is rule based, not guessed.

## The three verb groups

Almost every verb falls into one of these groups. Knowing the group tells you the pattern.

| Group | Also called | Rule in short | Example |
| --- | --- | --- | --- |
| Godan | Group 1, u-verb | The final kana changes along the vowel row | 書く: 書きます, 書かない, 書いて |
| Ichidan | Group 2, ru-verb | Drop る and add the ending | 食べる: 食べます, 食べない, 食べて |
| Irregular | Group 3 | Only する and 来る (くる) | する: します, しない, して |

## How to use the conjugator

1. **Type the dictionary form.** Enter the plain form that ends in u, such as 食べる, たべる, 書く or 勉強する.
2. **Read the verb type.** The page names the group and shows a note when the verb is an exception.
3. **Check the forms.** Polite, negative, past, te, potential, passive, causative, causative passive, imperative, volitional, ba, tara and tai are listed.
4. **Override if needed.** If a verb with kanji is classified wrongly, use the override toggle to force godan or ichidan.

## Telling godan from ichidan

A verb that ends in a kana other than る is always godan: 書く, 話す, 読む, 買う. A verb ending in る is ichidan if the sound before it is in the い row or え row, such as 食べる and 見る, and godan if it is not, such as 取る and 分かる. Most verbs follow this.

## The exceptions that look ichidan but are godan

A set of verbs end in iru or eru but conjugate as godan. Common ones are 帰る (かえる), 入る (はいる), 知る (しる), 走る (はしる), 切る (きる), 要る (いる), 減る (へる), 滑る (すべる) and 喋る (しゃべる). So 帰る becomes 帰ります and 帰らない, not 帰ない. Compounds follow the base verb, so 思い切る is godan too. Many words that sound the same are ichidan, for example 着る (きる, to wear) and 居る (いる, to be), so the written form matters. If the tool guesses wrong for a word, use the override.

## Irregular and honorific verbs

する and 来る change in their own way, and a noun plus する, such as 勉強する, follows する. 行く is regular except in the te and ta forms: 行って and 行った, not 行いて. ある has the negative ない, not あらない. The honorific verbs いらっしゃる, くださる, なさる, おっしゃる and ござる are godan but take い instead of り before ます: いらっしゃいます and ください (the imperative of くださる).

## Readings for kanji input

If you enter a verb with kanji, the page uses the dictionary to find its reading so it can decide the group. That downloads about 17 MB once and caches it. Kana input and the built-in list of common verbs work without it. After conjugating, you can put the forms into the [sentence breakdown](https://furiganascan.com/sentence-breakdown/) or add readings with the [furigana converter](https://furiganascan.com/furigana-converter/).

## Frequently asked questions

**What forms does it show?**

Polite, negative, past, past negative, te, potential, passive, causative, causative passive, imperative, volitional, ba conditional, tara conditional and tai. Each is shown in the same script you entered.

**Why was my verb classed as godan when I expected ichidan?**

Some verbs ending in iru or eru are godan, such as 帰る and 知る. The tool uses a list of those exceptions. If it still gets one wrong, use the override toggle.

**Does it handle する verbs?**

Yes. する and nouns plus する, such as 勉強する, conjugate like する. 来る and its kana form くる are supported too. Enter the noun and する together, for example 勉強する, and the page lists polite 勉強します, negative 勉強しない and the other forms.

**Is it free?**

Yes. The conjugator runs in your browser, has no limit, and does not use scan quota. There is no sign-up and no ads in the tool, and the verb list and rules are built into the page, so nothing is sent to a server.

**Does it include keigo?**

It covers honorific verbs with irregular forms: いらっしゃる, くださる, なさる, おっしゃる and ござる. It does not generate full keigo phrases from plain verbs. Polite and humble forms made from ordinary verbs, such as お書きになる, are not generated here.

Last updated: October 23, 2026

# Kanji to hiragana converter

https://furiganascan.com/kanji-to-hiragana/

Paste Japanese text and get it back written entirely in hiragana.

## Japanese text

Text with kanji

Also convert katakana to hiragana

## Hiragana

Hiragana output

## Short answer

To convert kanji to hiragana, paste your Japanese text into the box above. The converter replaces every kanji with its reading and turns katakana into hiragana, so the whole text is in kana. It runs in your browser, so it is free and has no daily limit. This also works for Japanese to hiragana.

## Kanji to hiragana or furigana?

Both show readings. They answer different needs.

| Need | Kanji to hiragana | Furigana converter |
| --- | --- | --- |
| Result | Only kana, kanji removed | Kanji kept with small kana above |
| Good for | Pronunciation, typing practice, text for young readers | Reading real text while learning kanji |
| Katakana words | Converted to hiragana | Left as written |
| Output | Plain text | HTML ruby, brackets or hiragana |

## How to convert Japanese text to hiragana

1. **Paste your text.** Put the Japanese text in the input box. Mixed kanji, hiragana and katakana are all fine.
2. **Let the dictionary load.** The first run downloads about 17 MB of dictionary data and caches it in your browser.
3. **Read the output.** The converted text appears as you type. Latin letters, numbers and punctuation stay as they were.
4. **Copy.** Press Copy hiragana to put the result on the clipboard.

## Why a conversion can be wrong

Kanji often have more than one reading, and the correct one depends on the word. 今日 is usually きょう, but it is read こんにち in 今日は (こんにちは), and 人気 can be にんき or ひとけ. The converter uses a dictionary and word boundaries to guess, and it is right most of the time, but it cannot see context beyond the word.

Names are the weakest spot. A surname written 小林 may be こばやし or おばやし, and the dictionary only knows common ones. Check names by hand.

## Sound changes you will see

Compound words often voice their second part, a change called rendaku. 手紙 is てがみ, not てかみ, and 花火 is はなび, not はなひ. The dictionary stores these readings for known words. If a new compound shows an unvoiced reading, the dictionary probably did not know it.

Particles are written as they are spelled. は stays は even though it is pronounced wa. For pronunciation in Latin letters, use the [Japanese to romaji](https://furiganascan.com/japanese-to-romaji/) page, which handles the particle sounds.

## Keeping the kanji

If you want to see both, use the [furigana converter](https://furiganascan.com/furigana-converter/). To get text out of a picture first, use the [screenshot scan](https://furiganascan.com/).

## Frequently asked questions

**Does it convert katakana too?**

Yes. Katakana is converted to hiragana, so the output is all hiragana. If you want katakana left alone, use the furigana converter instead. Loanwords such as コーヒー become こーひー, with the long vowel mark kept as it is.

**Is my text sent anywhere?**

No. This tool runs in your browser with a dictionary stored there. The text you type is not uploaded. The only download is the dictionary, which your browser caches after the first visit, and your text never leaves the page.

**Can I convert Japanese to hiragana for a whole article?**

Yes, paste it in sections if it is very long. There is no daily limit, though very large pastes can feel slow on older phones.

**Why are some readings wrong?**

The tool picks readings word by word and cannot read context or know every name. Wrong readings are most common with names, rare words and kanji with several readings.

**What happens to numbers and English words?**

They are left unchanged. Only Japanese characters are converted. Punctuation, spaces and line breaks are also kept, so the layout of your text stays the same after conversion.

Last updated: October 23, 2026

# Furiganascan pricing

https://furiganascan.com/pricing/

## Short answer

Furiganascan is free to use. Every visitor gets 3 scans per day, shared by the screenshot scan, the OCR page and sentence explanations. The converters run in your browser and are free without a limit. Two one-time scan packs are planned, but paid packs are not on sale yet.

## Free

$0 3 scans a day per visitor

Screenshot scans, OCR and sentence explanations share the 3 daily scans. The converters and the verb conjugator are free with no limit.

[Scan an image](https://furiganascan.com/)

## Scan packs

Paid packs are not on sale yet. Nothing here can be bought and no payment details are collected.

### 300 scans

Coming soon

$4.99 one time, not a subscription

### 800 scans

Coming soon

$9.99 one time, not a subscription

## What is free today

- 3 scans per day for each visitor. The count resets at 00:00 UTC.
- A screenshot scan, an OCR run and an Explain meanings and grammar request each use one scan.
- The [furigana converter](https://furiganascan.com/furigana-converter/), [kanji to hiragana](https://furiganascan.com/kanji-to-hiragana/), [Japanese to romaji](https://furiganascan.com/japanese-to-romaji/), [romaji to hiragana](https://furiganascan.com/romaji-to-hiragana/), [hiragana to romaji](https://furiganascan.com/hiragana-to-romaji/), [verb conjugator](https://furiganascan.com/japanese-verb-conjugator/) and the instant part of the [sentence breakdown](https://furiganascan.com/sentence-breakdown/). These run in your browser, cost nothing and have no limit.

A scan that fails because the model could not produce a result is not counted.

## Planned scan packs

Two packs are planned: $4.99 for 300 scans and $9.99 for 800 scans. Both would be one-time purchases, not subscriptions. They are shown here as Coming soon. Paid packs are not on sale yet, nothing can be bought, and no payment details are collected on this site.

Plans and prices may change before anything is sold. This page will say so on the day packs go on sale.

## How the daily limit works

The limit is counted per visitor using a salted hash of the network address and the date, which is deleted after two days. Changing browsers on the same connection does not give more scans. Trying to avoid the limit is against the [terms](https://furiganascan.com/terms/). There is also a daily cap for the whole site, so scanning can occasionally be unavailable. If that happens, the converters still work. See the [privacy page](https://furiganascan.com/privacy/) for details and the [refund page](https://furiganascan.com/refund/) for the planned policy.

## Frequently asked questions

**Do I need an account?**

No. There are no accounts. The free scans are counted per visitor without a sign-up, and nothing is collected for payment. Your browser only keeps a small first-visit record in local storage, and the privacy page explains what it contains.

**Are the converters limited?**

No. The furigana, hiragana, romaji, conjugation and instant sentence breakdown tools run in your browser and have no limit. They also do not count toward the three daily scans, and they keep working when scanning is switched off or the daily cap has been reached.

**Can I buy a pack now?**

No. Paid packs are not on sale yet. The prices of $4.99 for 300 scans and $9.99 for 800 scans are planned, one-time and subject to change.

**What happens when I use all 3 scans?**

Scanning pauses until the count resets at 00:00 UTC. The page tells you when. Converters keep working in the meantime. You can come back the next day for three more, or use the converters, which never run out for any visitor.

**Is there a subscription?**

No subscription is planned for the packs. They are meant as one-time purchases of a set number of scans. Nothing is on sale yet, and there is no recurring billing or automatic renewal, because packs are the only paid option under consideration.

Last updated: October 23, 2026

# Furiganascan privacy policy

https://furiganascan.com/privacy/

## Short answer

Images and text you scan are processed in memory and not stored by this site. They are sent to an AI model via OpenRouter. We log basic facts about each scan, never its content. Only Google Analytics 4 sets cookies. The converters run in your browser and send nothing.

## What happens to an image or text you scan

When you press a scan button, your browser sends the image (after shrinking it) or the text to this site's server. The server forwards it to an AI model through OpenRouter to produce the result, then sends the result back to you. The image and the text are processed in memory and are not stored, written to disk or put in the database.

The models are hosted by third parties. Once the data reaches OpenRouter and the model provider, how they handle it is under their own terms and policies, which we do not control. Do not scan anything you do not want to be sent to such a service, such as private messages or documents with personal data.

## What we log for each scan

One row per scan, containing only these items:

- the date and time
- the mode (scan, OCR or text)
- the page it was used from
- the referrer host, if any (the site name only, not the full address)
- the utm parameters from your first visit, if any
- the path of the page you first landed on
- the model used
- whether it succeeded, and an error code if not
- how long it took
- token counts for the request and the reply

The row does not contain the image, the text, the result or your IP address.

## Rate limiting

To enforce the free daily limit, the server computes a salted hash of your IP address and the date and keeps a counter against that hash. It is used only for rate limiting and is deleted after 2 days. The IP address itself is not stored.

## Cookies and local storage

The site itself sets no cookies; Google Analytics 4 sets its own (see Analytics below). Your browser's local storage keeps one small record: first-touch information, made up of the referrer host, utm_source, utm_medium, utm_campaign, the landing path and the date you first visited. The first four items and the landing path are sent with each scan so we can see which pages bring useful traffic; the first-visit date stays in your browser and is never sent. Nothing else is kept in local storage. You can clear it from your browser settings at any time.

The dictionary used by the converters is cached by your browser as ordinary downloaded files.

## Cloudflare and Turnstile

The site is served through Cloudflare. Pages with a scan button load Cloudflare Turnstile, a check that tells people from bots. Turnstile may process information from your browser under Cloudflare's own terms. It loads only on pages with a scan button, so the converter pages do not use it.

## Analytics

We use Google Analytics 4 to count page views and see which pages and referrers bring visitors. It sets its own cookies and processes your IP address and browser details under Google's terms; we do not send it the images or text you scan, or any account or contact details. You can block it with a browser extension or Google's opt-out add-on, and the tools keep working.

## Your choices and contact

Because scans are not tied to your identity, there is nothing to look up or delete about you beyond what is described above. For any privacy question, write to [support@furiganascan.com](mailto:support@furiganascan.com). See also the [terms](https://furiganascan.com/terms/) and the [about page](https://furiganascan.com/about/).

Last updated: October 23, 2026

# Furiganascan refund policy

https://furiganascan.com/refund/

## Short answer

Paid scan packs are not on sale yet, so nothing has been charged and there is nothing to refund. When packs go on sale, the planned policy is a refund within 14 days of purchase if none of the scans have been used. This page will be confirmed on the day packs launch.

## Current status

The free service costs nothing, and no one can buy anything on this site today. We collect no payment details. If you see a charge that you believe relates to this site, write to [support@furiganascan.com](mailto:support@furiganascan.com) and we will look into it.

## Planned policy once packs are on sale

- Packs are one-time purchases, not subscriptions, so there is nothing to cancel.
- You can ask for a refund within 14 days of purchase if you have not used any scans from the pack.
- Used scans cannot be refunded, as each one has a real cost to run.
- If a scan fails because of an error on our side, it is not counted against your pack.

Details such as the payment provider and how to request a refund will be added here before packs are sold. See the [pricing page](https://furiganascan.com/pricing/) for the planned packs and the [terms](https://furiganascan.com/terms/) for the rest.

Last updated: October 23, 2026

# Romaji to hiragana converter

https://furiganascan.com/romaji-to-hiragana/

Type Japanese in Latin letters and watch hiragana appear. Switch to katakana when you need it.

## Romaji

Type romaji

Output katakana

## Kana

Kana output

## Short answer

To convert romaji to hiragana, type Japanese words in Latin letters into the box above, such as konnichiwa. Hiragana appears as you type. A switch changes the output to katakana. The converter accepts Hepburn spellings like shi and tsu, and Kunrei spellings like si and tu. It is free.

## Common romaji and the kana they make

A few spellings cause most of the confusion.

| You type | You get | Note |
| --- | --- | --- |
| shi / si | し | Both accepted |
| chi / ti | ち | Both accepted |
| tsu / tu | つ | Both accepted |
| nn or n' | ん | Use before a vowel, as in kin'en |
| kka | っか | A doubled consonant makes small っ |
| xtu or ltu | っ | Types a small tsu on its own |
| - | ー | Long vowel mark, mostly for katakana |

## How to type hiragana with romaji

1. **Type in Latin letters.** Use lowercase letters. Type nihongo and you see にほんご.
2. **Double a consonant for small tsu.** Typing kitte gives きって. Type nn when ん comes before a vowel.
3. **Pick hiragana or katakana.** Use the switch to change the script. The text updates straight away.
4. **Copy the kana.** Press Copy to put it on the clipboard.

## Why the last letter may stay as a letter

While you type, a letter like k or ky cannot become kana until the next letter arrives. The converter leaves that last piece as Latin text until it is complete, so you may see a trailing k. Once you type a vowel it turns into the kana.

## Hepburn and Kunrei input

Hepburn writes sounds as an English speaker would read them: shi, chi, tsu, fu. Kunrei-shiki follows the kana table: si, ti, tu, hu. This tool accepts both, so you do not have to remember which one a source used. See the [Japanese to romaji page](https://furiganascan.com/japanese-to-romaji/) for the reverse direction with kanji.

## Typing particles

Type the kana, not the sound. To write the particle を, type wo. For the topic particle は, type ha, even though it is said wa. For へ type he. Mixing these up is the most common mistake in romaji typing.

Need to convert kana back to Latin letters? Use [hiragana to romaji](https://furiganascan.com/hiragana-to-romaji/).

## Frequently asked questions

**Does it convert as I type?**

Yes. The kana updates after every keystroke. A trailing consonant stays as a letter until you type the vowel that completes it. Nothing needs to be pressed first, and you can edit the text at any point to correct a spelling.

**How do I type a small tsu?**

Double the next consonant, for example kitte for きって, or type xtu or ltu to get っ alone. Both methods give the same result, so use whichever is easier on your keyboard.

**How do I type ん before a vowel?**

Type nn or n' so the converter knows the n is a full ん. For example, kin'en gives きんえん, while kinen gives きねん. A single n before a consonant, such as in kantan, already becomes ん without any extra letter.

**Can it make katakana?**

Yes. Turn on the katakana option and the same input gives katakana. Use the dash for the long vowel mark. This suits loanwords and names, for example typing koohii with the katakana option gives コオヒイ, while ko-hi- gives コーヒー.

**Is anything sent to a server?**

No. The conversion runs in your browser and needs no dictionary download. Your typing stays on the page, nothing is stored, and the tool keeps working once the page has loaded, even offline.

Last updated: October 23, 2026

# Japanese sentence breakdown and analyzer

https://furiganascan.com/sentence-breakdown/

Paste a sentence to split it into words, readings and parts of speech. Ask for meanings and grammar when you want them.

## Your sentence

Japanese sentence

3 free scans a day

Breaking a sentence into words is free and runs in your browser. Explanations use one of your 3 free scans a day.

## Words

Type or paste a sentence, then press Break down sentence. The words, readings and parts of speech show up here.

## Short answer

A Japanese sentence breakdown splits a sentence into words and labels each one. Paste a sentence above and the page shows every word with its reading, part of speech and dictionary form, instantly and for free in your browser. The Explain meanings and grammar button adds English meanings and notes, and uses one scan.

## Free segmentation or AI explanation

The page has two layers. The first costs nothing, the second uses your daily quota.

| Feature | Instant breakdown | Explain meanings and grammar |
| --- | --- | --- |
| Where it runs | In your browser | On an AI model through the scan endpoint |
| Gives you | Words, readings, part of speech, dictionary form | English meanings, grammar notes, translation |
| Cost | Free, unlimited | 1 of your 3 free daily scans |
| Needs | A dictionary of about 17 MB, downloaded once | A network connection and a passed check |

## How to break down a Japanese sentence

1. **Paste a sentence.** Enter one sentence or a short passage, up to 500 characters.
2. **Read the table.** Each row shows the word as written, its reading, its part of speech and its dictionary form.
3. **Ask for an explanation.** Press Explain meanings and grammar to get English meanings, grammar notes and a translation.
4. **Look words up.** Use the dictionary form column when you search a dictionary. Conjugated words do not appear under their surface form.

## What the table shows

The analyzer returns the surface form, the reading in hiragana, the part of speech and the base form of each word. For 食べました the base form is 食べる and the part of speech is a verb. Particles such as は and を show up as separate rows, which is useful when you are learning how a sentence is built. Auxiliary pieces like ました appear as their own rows too.

## Why segmentation can be wrong

Japanese has no spaces, so the analyzer uses a dictionary and statistics to place word boundaries. It does well on ordinary written text. Slang, dialect, names, game terms and text written all in hiragana are harder. If a split looks odd, try adding the kanji or check the word in a dictionary.

## When to use the explanation button

The breakdown tells you what the pieces are. It does not tell you what they mean together. For that, press the explanation button. It sends the text to an AI model, which returns word meanings, one to three grammar notes and a translation, and counts as one scan. Read [how your text is handled](https://furiganascan.com/privacy/) before sending anything private. For pictures, use the [screenshot scan](https://furiganascan.com/).

## Frequently asked questions

**Is the sentence breakdown free?**

The instant breakdown is free and unlimited because it runs in your browser. The Explain meanings and grammar button uses one of your 3 free daily scans.

**Is my sentence uploaded?**

Not for the instant breakdown. Only when you press the explanation button is the text sent to an AI model through OpenRouter. It is not stored here.

**How long can the sentence be?**

The explanation accepts up to 500 characters. The instant breakdown works on longer text but is easier to read in short pieces. Anything longer is cut off for the explanation, so split long passages into sentences and explain the ones that you find difficult.

**Why are there so many small pieces?**

Verb endings, auxiliaries and particles are counted as separate words. That is how the analyzer models the grammar, and it is helpful for learning conjugation.

**What is the dictionary form column?**

It is the base form of the word, for example 食べる for 食べました. Use it when you look the word up in a dictionary. Verbs and adjectives change their endings in a sentence, so searching for the surface form usually finds nothing.

Last updated: October 23, 2026

# Furiganascan terms of use

https://furiganascan.com/terms/

## Short answer

Use Furiganascan lawfully and for personal study. Do not upload illegal content or child sexual abuse material, do not use scripts or bots, and do not try to get around the daily limit. Results can be wrong and the service comes with no warranty. By using the site you accept these terms.

## The service

Furiganascan offers browser-based Japanese reading tools and a scan feature that uses AI models to read and explain Japanese text in images or pasted text. It is an independent personal project, provided as is. See the [about page](https://furiganascan.com/about/) and the [pricing page](https://furiganascan.com/pricing/).

## Acceptable use

- Use the service only for lawful content.
- Do not upload child sexual abuse material or any content that exploits minors. Such use is prohibited and may be reported.
- Do not use scripts, bots, scrapers or other automated means to call the scan feature.
- Do not try to bypass or reset the daily limit, the Turnstile check or the site-wide cap.
- Do not interfere with the service or attempt to access it in ways it is not meant to be used.
- Use it for personal study and reading. Do not resell results or the service.

## Your content

You are responsible for what you upload or paste. Images from manga, games, books and video are often copyrighted. Scanning them for your own private study is your responsibility, and you must make sure you are allowed to do so where you live. We claim no rights to your content, and we do not store it. See the [privacy policy](https://furiganascan.com/privacy/).

## Results may be wrong

Readings, meanings, grammar notes and translations are generated by AI models or by dictionary software and can contain errors. They are learning aids, not professional translation, and should not be relied on for legal, medical, safety or business decisions.

## No warranty and limits of liability

The service is provided as is and as available, with no promise that it will be accurate, uninterrupted or fit for a particular purpose. To the extent the law allows, we are not liable for any loss or damage arising from your use of the service or from errors in its results. Nothing here limits rights that cannot be limited by law.

## Changes and suspension

We may change the service, the limits or these terms, and may suspend or stop the service or an individual's access, for example for misuse, at any time. The date at the bottom shows the last update. Continued use means you accept the current terms. Questions go to [support@furiganascan.com](mailto:support@furiganascan.com).

Last updated: October 23, 2026

# What is furigana?

https://furiganascan.com/what-is-furigana/

## Short answer

Furigana are small hiragana or katakana printed above or beside a kanji to show how it is read. For example, 漢字 with かんじ above it. They are used in children's books, manga, learner material and for rare kanji. On the web they are written with the HTML ruby element.

## Furigana compared with related terms

Several words describe similar things.

| Term | Meaning | Example |
| --- | --- | --- |
| Furigana (振り仮名) | Kana that show the reading of kanji | 漢字 with かんじ above |
| Ruby (ルビ) | Another word for furigana, from the printing term | The same as furigana |
| Okurigana (送り仮名) | Kana that follow a kanji stem as part of the word | The べる in 食べる |
| Romaji | Japanese written in Latin letters | kanji |
| Pinyin or zhuyin | Reading aids for Chinese, not used for Japanese | hànzì |

## How to get furigana for Japanese text

1. **Have the text ready.** Copy the Japanese text, or take a screenshot if it is in a picture.
2. **Use a converter.** Paste text into the furigana converter. For pictures, use the screenshot scan, which reads the text and adds furigana in one step.
3. **Check unusual readings.** Names and rare words are where converters err. Confirm those in a dictionary.
4. **Copy in the format you need.** Use HTML ruby for web pages or the bracket format for plain notes.

## How furigana work

Each kanji can be read several ways, and a reader cannot always tell which is right. Furigana remove the doubt. In horizontal text they sit above the kanji, in vertical text to its right. They are normally hiragana, but katakana is used for foreign words or for the reading of Chinese-origin terms in some books. The word furigana comes from 振り仮名, meaning kana attached to something, and the printing term is ruby.

## Where you will see furigana

- Children's books and school textbooks, where kanji have not been learned yet.
- Manga, especially titles for young readers. Many weekly shonen series put furigana on every kanji, while seinen series use them less.
- Learner readers and graded readers.
- Rare or difficult kanji, names and unusual readings in novels and newspapers.
- Video games with a furigana option, and subtitles for learners.

Ordinary adult text, signs and websites usually have none.

## Furigana on the web

HTML has the elements `ruby`, `rt` and `rp`. A reading goes in `rt` and fallback brackets in `rp`. The [furigana converter](https://furiganascan.com/furigana-converter/) produces this markup for you and also gives a bracket version for plain text.

## Using furigana to learn

Furigana are a crutch that is worth using at first. Read with them, then cover them and test yourself. Over time, drop them for words you know. To turn a whole text into kana with no kanji, use [kanji to hiragana](https://furiganascan.com/kanji-to-hiragana/). For stuck screenshots, see [how to read Japanese text in an image](https://furiganascan.com/how-to-read-japanese-text-in-an-image/).

## Frequently asked questions

**What is the difference between furigana and okurigana?**

Furigana are extra small kana that tell you how to read a kanji. Okurigana are normal-size kana that are part of the word's spelling, like the べる in 食べる. Furigana can be removed. Okurigana cannot.

**Is furigana the same as ruby?**

Yes, for practical purposes. Ruby is the typesetting term, taken from a small type size, and furigana is the Japanese word for the reading aid. Web developers use the ruby element.

**Do Japanese adults use furigana?**

Not for ordinary text, but they appear in manga, books for young readers, and for unusual kanji or names. Many adults rely on them for rare characters.

**Can I add furigana to any text?**

Yes. Paste text into a converter. For text in an image, a screenshot scan reads it first. Check names and uncommon words, because automatic readings can be wrong.

**Why do some manga have no furigana?**

Publishers choose based on the target age. Series aimed at older readers often assume they know the kanji. Learners may prefer editions that include furigana on all kanji.

Last updated: October 23, 2026