Sources
Attribution & Licences
Where the words, tables and sentences on this site come from, and which parts were written by AI. The word list and the verb tables are published under CC BY-SA 4.0, inherited from Wiktionary. The example sentences keep Tatoeba's licence, CC BY 2.0 FR. If you reuse any of it, keep this attribution reachable.
What was written by AI
- The grammar point list. The 156 grammar points' titles and model phrases, and the patterns that find their example sentences, were written by Claude Opus 5.5, made by Anthropic. Each point's level is the earliest at which the Eaquals/CIEP inventory lists it; the imperfect subjunctive and the passé antérieur, which it does not name, were placed at C1 beside its passé simple by the same author.
- The grammar explanations. Original explanations authored with OpenAI Codex, not yet reviewed by a language teacher. The explanations, including short illustrations within the prose, are offered under CC BY 4.0.
- This site's code, and the code that builds its data, written with Claude Code and OpenAI Codex.
Vocabulary meanings are Wiktionary's, cut to a short form by rule rather than rewritten. The example sentences in the vocabulary and grammar Examples sections and their translations are Tatoeba's and Wiktionary's; recordings are Lingua Libre volunteers'; and levels, frequencies, usage counts and the irregular verb forms are computed.
Graded reading
Stories and nonfiction use African Storybook / Global ASP texts under CC BY 3.0 or CC BY 4.0 and Wikipedia excerpts under CC BY-SA 4.0. Selected complete Tatoeba dialogues, where present, retain CC BY 2.0 FR. Each reading credits its authors and translators and links its source and licence. Noncommercial and no-derivatives sources are excluded.
Original beginner stories and their English versions were written with OpenAI Codex and are labelled AI-assisted, not yet reviewed by a language teacher. They are offered under CC BY 4.0. Source stories and source translations are retained rather than rewritten. Vocabulary percentages are computed against this site's cumulative word lists; they are estimates of reading difficulty, not official exam ratings.
Immersion videos
The immersion videos and native captions come from the publishers credited on each video page. Each page links to the original video and channel. Video and caption rights remain with those publishers. The lexicon's CC BY-SA licence does not apply to them.
Short English titles and listening-level estimates are AI-assisted editorial work. English caption gaps are machine translations, labelled on the video page; the channel's English is used wherever available. Native captions are supplied by the channel.
Where the data comes from
| Source | Provides | Licence |
|---|---|---|
| English Wiktionary via wiktextract | The words, their parts of speech and meanings, gender, plurals and feminine forms, the masculine before a vowel (bel, nouvel), aspirated h, IPA with Quebec's where it differs, synonyms and antonyms, and every verb table with its auxiliary | CC BY-SA 4.0 |
| Tatoeba | The example sentences for words and grammar points with their English translations, and the corpus the usage counts are taken from | CC BY 2.0 FR (some CC0) |
| Lingua Libre on Wikimedia Commons | Recordings of 11,691 words, played from Commons. Each play button names its speaker and licence | Per file: CC BY-SA 4.0, CC BY-SA 3.0, CC BY 4.0 or CC0 |
| wordfreq (Robyn Speer) and FrequencyWords (Hermit Dave) | How common each word is, in general text and in film subtitles, which orders every list and sets every word's level | CC BY-SA 4.0 |
| Lexique 3.83 (New, Pallier et al.) | How common each word is in books, and the written syllables where Wiktionary has none | CC BY-SA 4.0 |
| Inventaire linguistique des contenus clés des niveaux du CECRL (Eaquals and CIEP, 2015) | The level of every grammar point. Words are graded by how common they are, raised for everyday words that Tatoeba's sentences use often | Used for levels only, not republished |