About – vocab dash

vocab dash is a multilingual dictionary and spaced repetition system modeled on jpdb.io. It teaches Japanese, Hebrew, Telugu, Spanish, French, Korean and Chinese so far, with more languages coming soon. Reviews are scheduled with FSRS-6.

Data sources and licenses

Every language

Example sentences from the Tatoeba project (CC BY 2.0 FR).
Word frequencies from wordfreq (CC BY-SA 4.0), and for Telugu and Korean from FrequencyWords' subtitle counts (CC BY-SA 4.0).
Dictionaries, inflections and usage examples for every language but Japanese and Chinese from English Wiktionary (CC BY-SA 4.0), as extracted by kaikki.org. Some common Hebrew words Wiktionary lacks were added by hand.
Script charts, guides and the Everyday course decks: written for vocab dash.
Picture-book decks from Storybooks Canada (CC BY 4.0, with 11 of the 40 stories CC BY-NC), and first-words decks from Wordbank, Stanford (CC BY 4.0).
How often a word stands for each of the dictionary words spelled like it, and forms the dictionaries lack, from the Universal Dependencies treebanks Hebrew IAHLTwiki and IAHLTknesset, Spanish GSD and French GSD (CC BY-SA 4.0) and Spanish AnCora (CC BY 4.0), and for French from Lexique 4.1 (CC BY-SA 4.0).

Japanese

Vocabulary and kanji data from JMdict and KANJIDIC2, property of the Electronic Dictionary Research and Development Group, used in conformance with the Group's licence (CC BY-SA 4.0), via jmdict-simplified. Example sentences include the ones JMdict links to its senses.
Kanji and kana stroke data and kanji decompositions from KanjiVG by Ulrich Apel (CC BY-SA 3.0).
Kanji keywords from jpdb.io, as collected by KanjiCut (MPL-2.0) from Jo Liss's heisig-jpdb tables; the keywords remain jpdb.io's.
Furigana from JmdictFurigana (CC BY-SA 4.0); pitch accents from Kanjium by Uros O. (CC BY-SA 4.0).
Example sentence readings generated with Lindera and IPADIC.
Irodori decks from the Japan Foundation's Irodori word lists, and JLPT decks from Jonathan Waller's tanos.co.uk lists (CC BY), via JLPT_Vocabulary (MIT).

Chinese

The dictionary from CC-CEDICT (CC BY-SA 4.0).
Character meanings, parts and origins (LGPL-3.0), and stroke order drawn from Arphic Technology's fonts (Arphic Public License), from Make Me a Hanzi.
How text is split into words, measured against the Universal Dependencies Chinese GSDSimp treebank (CC BY-SA 4.0).
Graded reader decks from Alvaro Serrano's Pinyora (CC BY 4.0); HSK 3.0 decks from the 2025 syllabus via hsk30 (CC BY 4.0), and HSK 2.0 decks via complete-hsk-vocabulary (MIT).

Korean

Words found in their dictionary forms with Lindera and mecab-ko-dic (Apache License 2.0).
Beginner, intermediate and advanced decks from the National Institute of Korean Language's Basic Korean Dictionary (CC BY-SA 2.0 KR).

Spanish

CEFR decks from ELELex, CENTAL, UCLouvain (CC BY-NC-SA 4.0); school reader decks from Project Gutenberg (public domain).

French

Course decks from Français interactif, University of Texas at Austin (CC BY 3.0); CEFR decks from FLELex, CENTAL, UCLouvain (CC BY-NC-SA 4.0); Le Tour de la France par deux enfants from Project Gutenberg (public domain).

Telugu

Story decks from Pratham Books' StoryWeaver (CC BY 4.0), and the Swadesh list from Wiktionary (CC BY-SA 4.0).
Further example sentences from AI4Bharat's Samanantar corpus (CC0), via OPUS.

Japanese recordings are synthesized with AivisSpeech (voice: にせ); other languages use Google's Gemini text-to-speech.

The interface font is Nunito Sans, Japanese is set in Noto Sans JP, and the language marks are drawn from Noto Sans (all SIL Open Font License).

The app is built on open-source software; see the third-party licences.

About vocab dash