About

Version 4.1.0

Patch notes

Dictionary data

The in-app dictionary uses the JMdict dictionary file, which is the property of the Electronic Dictionary Research and Development Group, and is used in conformance with the Group's licence.

Kanji information — readings, meanings, stroke counts, and radicals — is from the KANJIDIC2 file, also the property of the Electronic Dictionary Research and Development Group, used in conformance with the Group's licence.

JLPT level tags are from Jonathan Waller's JLPT Resources and the yomitan-jlpt-vocab project, used with credit under their Creative Commons licences (CC BY / CC BY-SA 4.0). Official JLPT word lists are no longer published, so levels are community estimates.

Word-frequency estimates are derived from data in the wordfreq project by Robyn Speer, used under CC BY-SA 4.0.

English word forms are adapted from MorphyNet by Khuyagbaatar Batsuren, Gábor Bella and Fausto Giunchiglia (2021). The adapted index retains English inflections and suffix derivations, excludes prefixes and deduplicates bidirectional links. It is licensed under CC BY-SA 3.0.

English semantic search uses compressed vectors adapted from Model2Vec by Stephan Tulkens and Thomas van Dongen, under the MIT licence. The derived dictionary search indexes are licensed under CC BY-SA 4.0.

English synonym links include adapted data from Princeton University's WordNet 3.0 (copyright 2006 Princeton University), used under its licence with the original notices retained.

Additional English synonym links are derived from glosses within the same JMdict sense, under CC BY-SA 4.0.

Pitch-accent marks are derived from UniDic, the electronic dictionary of the National Institute for Japanese Language and Linguistics (NINJAL), used under its BSD licence.

Word formality histograms are the developer's own estimates, computed from register frequencies in 『現代日本語書き言葉均衡コーパス』短単位語彙表 ver.1.0 and 『日本語日常会話コーパス』短単位語彙表 ver.202209 (NINJAL). Each bar shows how much of a word's usage falls at one formality level, estimated from where the word appears:

The displayed levels and tiers are learning guidelines derived by this app — NINJAL does not guarantee or endorse the classification.

Example sentences

Example sentences shown on dictionary entries come from these openly-licensed collections:

Word audio

Spoken word readings are synthesised with VOICEVOX:東北きりたん・雀松朱司.

Stroke order

Kana and kanji stroke-order diagrams are derived from KanjiVG by Ulrich Apel (CC BY-SA 3.0).

The typeset forms shown beside them are outlines from Noto Sans JP, Noto Serif JP and Klee One (SIL Open Font License 1.1).

Made by Danbing Chen.