Linguistic data

Sabes is built on open-licensed linguistic data. We gratefully acknowledge the following sources, and we publish our derived datasets under matching licences.

Frequency rankings

Derived from the OpenSubtitles 2018 corpus via the FrequencyWords project by HermitDave. Used and adapted under Creative Commons Attribution-ShareAlike 4.0 (CC BY-SA 4.0).

Lemmas and English translations

Wiktionary-derived data from the spanish_data project by Jeff Doozan. Used and adapted under Creative Commons Attribution-ShareAlike 4.0 (CC BY-SA 4.0).

Example sentences

Community-contributed sentences from Tatoeba.org. Used under Creative Commons Attribution 2.0 France (CC BY 2.0 FR).

Derived datasets

The compiled vocabulary and verb datasets bundled with the App (words.json and verbs.json) adapt the sources above: we filter the frequency list, merge it with the lemma and translation tables, select example sentences, and reformat the result for the App. These datasets are derivative works and are released under Creative Commons Attribution-ShareAlike 4.0 (CC BY-SA 4.0).

A copy of the current dataset build is available free of charge under that licence: email hello@sabes.app and we will send it to you.

Conjugation generation

Conjugation tables are generated at build time using verbecc by Brett Tolbert, licensed under the GNU Lesser General Public License v3 (LGPLv3). Only the generated JSON output ships in the App; the verbecc library itself is not bundled or distributed.

Self-authored content

Grammar lessons, verb deep dives, etymology notes, business vocabulary, regional variants, cultural notes and teaching insights are written by Sabes and are not derived from third-party sources. They remain the property of Helvellyn Ventures Ltd.

System resources

Sabes uses Apple SF Symbols and Apple system fonts under Apple's SF Symbols and system font licences.

Contact

Questions about licensing or attributions:

Helvellyn Ventures Ltd
hello@sabes.app