Japanese Readings and Pitch Accent Add-on

This Anki add-on allows automatically generating readings (furigana) and pitch accent information for an entire field with a single button press. It can also convert any amount of cards in bulk.

<img src="https://i.imgur.com/8NFw0HB.png">

The add-on is currently experimental and you'll probably encounter bugs or crashes. Data corruption is extremely unlikely (probably even impossible), the specific notes being changed aside, but you should still make backups, especially when bulk converting notes, and enable Anki's built-in automatic backups if you haven't already.
See below for details on how to report issues.

Getting Started

1. Make sure you have a supported version of Anki installed. See the AnkiWeb page for further information.
2. On non-Windows platforms only, install MeCab since only a Windows executable is bundled with the add-on.
3. Install the full version of the add-on from AnkiWeb.
   Alternatively, download one of the .ankiaddon files from the latest release and manually install it from the add-on menu.
   • The smallest file contains only the add-on itself, which means you will need to supply your own pitch accent/variant data and install MeCab as well as a MeCab dictionary alongside a dictionary for it to use externally.
   • The full file has everything except a MeCab dictionary, but including a MeCab exe for Windows.
   • The ipadic file contains everything the add-on needs out of the box and is probably what most users will want to install initially.
     This is also the version shared on AnkiWeb.
4. Restart Anki, open the Manage Note Types menu (Ctrl+⇧+N) and set up your templates as described here.
5. Open the add-on's preferences under Tools in Anki's main menu bar, go the Note Types tab, then select and Add all note types you set up in the last step.
   Click Remove MIA/Migaku for any note types you previously used the Migaku Japanese Add-on with to remove the script and styling from that add-on, otherwise this add-on will not work properly.
6. Save your preferences. You should now be able to preview or review any cards that contain reading/accent syntax with furigana and accent coloring / indicators.

Adding Readings and Pitch Accent Information

Individual

To generate or remove readings and accent info for a single note, select it in the Anki browser, focus the field you want to change and click one of the conversion buttons at the right of the editor toolbar or press the associated shortcut.

Bulk Conversion

The Notes entry in the browser's menu bar contains an action for converting notes in bulk.

Select any notes you want to change, choose the bulk conversion action from the menu bar, adjust the configuration dialog to what you want and click Convert. All notes need to be of the same note type, since it wouldn't be possible to determine the target field otherwise.

With the Default and Migaku conversion types, any notes that already contain reading/accent syntax will be converted directly to the closest equivalent in the target syntax without regenerating (unless the corresponding option is checked), or left unchanged if the existing syntax type is the same as the target.

If you're converting large amounts of notes, such as sentence mining decks containing hundreds or thousands of cards, make sure to back everything up before running the conversion. Exporting the deck(s) with scheduling information (but without media, since the conversion only changes the notes themselves) should be sufficient.

Syntax

The add-on supports its own fully-featured syntax and is also compatible with the syntax from the old Migaku Japanese Add-on.

Default

In the default syntax, readings are written in square brackets with kanji on the left and furigana on the right separated by a |, like [振|ふ]り[仮名|がな].
Words with pitch accent information are enclosed in curly braces that contain the word (including reading tags) itself and accent information after a semicolon: {[受|う]け[入|い]れる;Y4,0}.

The accent information normally consists of a comma-separated list of numbers, each representing the accented mora, or 0 for unaccented (平板) words. The only special cases are unknown accents marked with a ?, which can only occur by converting from Migaku syntax, and split accents like 一目瞭然 (いち↓もく・りょ→うぜん) which are composed of several parts separated by dashes. Each part consists of the number of the accented mora, an @ sign, and the number of moras it applies to. For example: {[一目瞭然|いちもくりょうぜん];2@4-0@4,0}.

An ! and Y can be placed before the actual pitch accent information.
Exclamation marks indicate ambiguous accents, such as for many kana-only words or certain words with multiple accents. The add-on automatically inserts ambiguity marks if it found more than one possible set of accents with the same reading for a word. You should then manually look up the words in question and adjust the accent tags as necessary.
If present, a Y (from 用 as in 活用・用言) will cause all accents other than [0] to be displayed as the kifuku (起伏) pattern. Ys are added to all words identified as 動詞 (verbs) or 形容詞 (i-adjectives) by MeCab.

Base readings for conjugated words can be indicated in two ways, inline like {[行|い][って=く];Y0}, or after the accent(s) like {[来|き]た;Y1|くる}.

Migaku

For backwards compatibility, Migaku-style syntax is also supported, but it has some limitations compared to the new syntax:

• Only one reading per word is possible, leading to furigana frequently duplicating characters from the word, like <ruby>この先<rt>このさき</rt></ruby>, <ruby>付き合<rt>つきあ</rt></ruby>う or <ruby>変わり身<rt>かわりみ</rt></ruby>.
• The base reading of conjugable words must always be included in full, even the part identical to the reading of the conjugated form. This is also true if the base reading and actual reading are the same, so the Migaku version of what would be {[陥|おとしい]れる;Y5,0} in default syntax is 陥[おとしい,おとしいれる;k5,h]れる.
• Since words are space-separated, regular ASCII spaces can't be represented properly and are replaced with en spaces (U+2002).
• Ambiguous accents can't be marked and thus won't be highlighted.
• Split accents can't be accurately represented and will be displayed as unknown.

Migrating from the Migaku Japanese Add-on

Follow the guide to getting started above. You'll need to manually transfer your accent pattern colors to the note type style settings if you want to keep them. The add-ons have different default colors.

You can easily migrate existing notes to the new syntax without losing any manual changes by selecting them in the browser and bulk converting them with the Default conversion type as long as Regenerate contents is disabled. Make sure to back up all notes in case there are bugs in the conversion algorithm.

Reporting Issues

If you encounter a bug or crash while using the add-on please report it. You can either open a GitHub issue in this repository or contact me on Discord, where you can find me in the Refold community servers.

Explain what you were doing when the issue occurred as accurately as possible, ideally with a list of steps to reproduce it. If the problem is related to specific cards, consider including an .apkg export of those cards (Notes → Export Notes... in the Anki browser while having the cards in question selected). If there was an error message include its full contents as well, if possible as selectable text.