Hi everyone,
I’ve been working on a free Japanese reading tool called EZFurigana, and I’d really appreciate feedback from WaniKani users who are moving from kanji study into native reading material.
The main idea is simple: take Japanese text or files and add furigana in a way that tries to respect context, rather than just applying a dictionary reading everywhere.
The main problem I wanted to solve was that simple dictionary-based furigana works well for common cases, but breaks on words where the reading depends on context, for example:
- 市場: いちば or しじょう
- 大分: おおいた or だいぶ
- 人気: にんき or ひとけ
- 最中: さいちゅう, さなか, or もなか
- 方: かた or ほう
For the furigana engine, I’m using a context-aware BERT-based approach to choose readings from the surrounding sentence. It is not perfect, especially with names, unusual compounds, and ambiguous phrases, but I’ve been trying to make the accuracy measurable and transparent rather than just claiming it works.
I also recently added an EPUB reader and a PDF furigana tool. The EPUB reader is meant for native books with adjustable furigana, dictionary popups, and reading settings. The PDF tool is still more limited, but I’m polishing it too.
I’ve also been improving the popup dictionary experience. The English learner-oriented entries use TKG Japanese-English Learner’s Dictionary, which is designed more like a learner dictionary than a terse lookup database: clearer explanations, examples, usage notes, and furigana. I’m not trying to replace tools like Yomitan, but to make native reading feel smoother and less intimidating.
I’m posting here because I’d like feedback from people who know the pain points of reading native texts after learning kanji. I’d especially like to know:
- whether the furigana feels accurate enough to trust while reading
- whether the EPUB/PDF workflows feel useful for real reading practice
- what would make this more useful alongside WaniKani, Yomitan, or Anki
The tool is free and does not require signup.
Link: https://www.ezfurigana.com/
Thanks, and I’m open to criticism or examples where the readings are wrong.