Sources and credits

• Wiktionary

  Pronunciations (IPA), word forms and usage labels come in part from Wiktionary (English and Catalan editions, through the Kaikki.org extracts), written by its contributors and licensed under CC BY-SA 4.0. Where we show them, they keep that licence. CC BY-SA 4.0

• TERMCAT

  Terminology in the Catalan word sets was checked against TERMCAT's open vocabularies (Centre de Terminologia de la Llengua Catalana), used under their CC BY licences. The selection, grouping, explanations and examples are our own.

• Softcatalà

  The Catalan word-form dictionary (catalan-dict-tools) and the English-Catalan dictionary, licensed under CC BY 4.0, by Softcatalà and its contributors. CC BY 4.0

• Universal Dependencies (AnCora)

  The Catalan and Spanish AnCora treebanks (CC BY 4.0) help us tell which word a form belongs to.

• Unicode CLDR

  Numbers written out in words, under the Unicode licence.

These references decide what is correct and at which level it is taught. We consult them and write our own lessons; we do not reproduce them, and their authors do not endorse parleu.

• Institut d'Estudis Catalans

  The Catalan standard: the dictionary (DIEC2), the grammar (GIEC and its basic edition), the spelling rules (OIEC) and the oral standard.

• Generalitat de Catalunya, Consorci per a la Normalització Lingüística

  The official Catalan programmes for each level, Vocabulari en imatges and Nivell llindar.

• Instituto Cervantes, Plan Curricular (PCIC)

  The inventories of Spanish grammar, functions and vocabulary for each level.

• Real Academia Española

  The dictionary (DLE), the spelling rules and the glossary of grammatical terms.

• CEFRLex (ELELex)

  Word frequencies by level in Spanish learning materials (UCLouvain).

• SUBTLEX-CAT, SUBTLEX-ESP, OpenSubtitles (OPUS)

  Word frequencies from film and television subtitles, used as counts only.

The story pictures are made with AI image models (Google Gemini and xAI Grok) from our own descriptions.

The voices in lessons, Review and Speak are synthetic voices from Google's Gemini, not recordings of real people.

Lessons, stories and explanations are written with Anthropic's Claude and translated with OpenAI models, then checked against the references above. Speak partners are AI conversation partners played live by Google's Gemini; word explanations and Speak feedback are made by Gemini too.

Podcasts belong to their publishers: their names, artwork and episodes are theirs, and the audio plays from their own servers. Songs in lessons are YouTube videos of their rights holders, played in YouTube's own player. A rights holder who wants something changed or removed can write to legal@parleu.com.

The fonts Geist, Bricolage Grotesque, Fraunces and Caveat are used under the SIL Open Font License 1.1.

parleu is built on open-source software. Every package the app ships, with its licence and the copyright and licence notices it requires, is listed in a file made from our dependencies at every build: third-party notices.