DE version is available. Content is displayed in original English for accuracy.
Advertisement
Advertisement
⚡ Community Insights
Discussion Sentiment
93% Positive
Analyzed from 741 words in the discussion.
Trending Topics
#japanese#dictionary#thanks#context#especially#cases#furigana#useful#appreciate#words
Discussion (17 Comments)Read Original on HackerNews
Regardless, I'm impressed with the tool!
如何 is context-dependent, and I hadn’t come across this case yet. I’ll add it to the model soon. Really appreciate the report and the kind words.
Also interested to hear if you plan to eventually support an option to add pitch accent; I've never seen what training material exists for that or how that is supported in unicode.
No signup was a deliberate choice, to keep the barrier to trying it as low as possible.
I’ve thought about pitch accent, but it feels like a whole separate beast. The datasets are less comprehensive, and pitch can be even more context-dependent. I’d like to look into it eventually though.
Also, it’s disappointing that Japanese does not appear even when I select it.
Please let me know if there’s anything I can do to help.
今日 is a tradeoff I made intentionally: I disabled the fallback model for it because most cases are きょう, while こんにち is much rarer. But yes, this is one of the cases that gets lost with that choice.
And agreed on Japanese dictionary support. I plan to add Japanese soon. Thanks again.
By the way, I tried testing it further while thinking back to the kinds of tests I had when I was in school. The accuracy is still excellent. My guess is that “一日” and “分別” are being handled in a similar way to “今日.” “分別” is very rare, but I don’t think “一日” is all that uncommon.
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:
* 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 or さなか or もなか
* 方: かた or ほう
The engine is a hybrid system:
* Sudachi for tokenization, base forms, POS, and candidate readings
* Expanded dictionary coverage for compounds and fixed expressions
* Custom rules for counters, suffixes, rendaku patterns, and phrase overrides
* ModernBERT fallback for 144 especially context-dependent target words
I have been testing it against an LLM-assisted benchmark of 7,500 Japanese lines. On the current benchmark, it gets about 12 wrong readings per 1,000 tokens. I treat that as a practical regression benchmark rather than a formal academic evaluation, but it has been useful for comparing versions and catching regressions.
The hardest remaining cases are personal names, place names, rendaku, rare vocabulary, and domain-specific terms.
I would especially appreciate examples where it gets the reading wrong, since those are the most useful for improving the system.
Since January, I’ve been having Claude build a static Japanese-English dictionary in which all of the kanji and jukugo can be displayed either with or without furigana:
https://www.tkgje.jp/index.html
I haven’t spotted any mistakes in the furigana myself, though there must be some. I have a scheduled routine running multiple times a day to have Claude check and polish existing entries; it should be correcting most of whatever furigana mistakes might be in the data. At some point, I will set up an agent to use a different LLM to run a similar set of checks to try to reduce the error rate even more.
As you note, the readings of Japanese words depend on the context, so producing accurate furigana cannot be done entirely programmatically. Sentences must be interpreted semantically.
I am releasing all of the dictionary data into the public domain, and anyone is free to fork it or adapt it however they like:
https://github.com/tkgally/je-dict-1
I especially like the dictionary + example sentence format. I haven’t found a really good Japanese-English dictionary for learners, and yours looks promising.
I’m curious how token-intensive the repeated Claude polishing runs are.
(Also: vouched, your comment was dead FYI)