Credits

Our vocabulary database is powered by JMdict, the comprehensive Japanese–English dictionary maintained by the Electronic Dictionary Research and Development Group. Every entry in our dictionary keeps its JMdict sequence number, and word-frequency ordering derives from JMdict’s own priority data.

Author / holder

Electronic Dictionary Research and Development Group (EDRDG)

Source

github.com/scriptin/jmdict-simplified

Our changes

Content filter: 38 entries excluded and 137 senses removed (documented in data/README.md) Romaji, JLPT levels, frequency ranks and learn ordering are our own additions

This publication has included material from the JMdict (EDICT, etc.) dictionary files in accordance with the licence provisions of the Electronic Dictionaries Research Group.

Our adaptations of this data are available under the same license: if you want the JMdict (via jmdict-simplified) portions of our database in their adapted form, contact us and we will provide them under CC BY-SA 4.0.

Japanese-language definitions and etymological notes (the 日本語 definition panels) come from Japanese Wiktionary, extracted with the wiktextract tool by Tatu Ylonen. Each panel links back to the source page it came from.

Author / holder

Wiktionary contributors; extraction tool © Tatu Ylonen (MIT)

Source

ja.wiktionary.org

Our changes

Filtered to lang_code === "ja" entries, readings normalized, redirects dropped, reshaped into our sense schema

Academic use of the extracted data should cite: Tatu Ylonen, "Wiktextract: Wiktionary as Machine-Readable Structured Data", Proceedings of LREC 2022, pp. 1317–1325, Marseille.

Our adaptations of this data are available under the same license: if you want the Japanese Wiktionary (via wiktextract) portions of our database in their adapted form, contact us and we will provide them under CC BY-SA 4.0.

On and kun readings for our 5,015 kanji come from the KANJIDIC2 project, accessed via the kanjiapi.dev API.

Author / holder

Electronic Dictionary Research and Development Group (EDRDG)

Source

www.edrdg.org/wiki/index.php/KANJIDIC_Project

This publication has included material from the KANJIDIC dictionary files in accordance with the licence provisions of the Electronic Dictionaries Research Group.

Our adaptations of this data are available under the same license: if you want the KANJIDIC2 portions of our database in their adapted form, contact us and we will provide them under CC BY-SA 4.0.

Stroke-order diagrams and writing practice for kanji and kana are drawn from KanjiVG stroke data by Ulrich Apel.

Author / holder

Ulrich Apel

Source

kanjivg.tagaini.net

Our changes

XML prologue, DOCTYPE and per-file copyright comment stripped from each SVG on import; the kvg: element structure is kept intact

Our adaptations of this data are available under the same license: if you want the KanjiVG portions of our database in their adapted form, contact us and we will provide them under CC BY-SA 3.0.

The kanji character set, stroke counts, English descriptions, and JLPT level assignments originate from the jlpt-kanji-dictionary project by Egor Kiselev.

Author / holder

Egor Kiselev (AnchorI)

Source

github.com/AnchorI/jlpt-kanji-dictionary

Our changes

5 descriptions blanked by our content filter

JLPT level assignments for vocabulary come from the JLPT resource lists compiled by Jonathan Waller at tanos.co.uk, which have levelled study vocabulary for learners since the test changed format in 2010.

Author / holder

Jonathan Waller (tanos.co.uk), via jamsinclair/open-anki-jlpt-decks (MIT)

Source

www.tanos.co.uk/jlpt

Our changes

Cross-level duplicates deduplicated (easiest level wins); okurigana headword spellings corrected

Example sentences from real Japanese literature come from Aozora Bunko, the digital archive of works whose copyrights have expired. We import only works flagged public-domain in the official catalog, and every displayed sentence names the work it came from.

Author / holder

Aozora Bunko and the original authors

Source

www.aozora.gr.jp

Our changes

Sentence-segmented; ruby markup normalized

Furigana and word-boundary analysis use the kuromoji.js morphological analyzer by Takuya Asano, which bundles the mecab-ipadic dictionary from the Nara Institute of Science and Technology. The full NAIST notice is reproduced in our third-party notices.

Author / holder

Takuya Asano (kuromoji.js); Nara Institute of Science and Technology (IPADIC)

Source

github.com/takuyaa/kuromoji.js

All audio on the site is speech synthesized from our own sentences using Google Cloud Text-to-Speech (and OpenAI for some voices). No recorded human speech is used.

Author / holder

Google LLC / OpenAI

Source

cloud.google.com/text-to-speech

Everything else — the example sentences, grammar explanations, kana guide, word relations, chapter curricula, and all teaching notes — was written for RT Nihongo Club and is our own work.

Author / holder

RT Nihongo Club

Source

/credits