Licences
TypEvent’s engine and interface are its own work. Its dictionaries and emoji data are built on open sources — thank you to their authors. Their licences ask us to credit them; this page does that, with the full licence texts.
Dictionaries, by language
Every dictionary is a modified version: TypEvent converted the sources into its own binary format, filtered them and gave them its own frequency ranking. The dictionaries contain words and numbers, not sentences or text.
| language | word list source | licence |
|---|---|---|
| Hungarian | morphdb.hu — Eszter Simon, Péter Rebrus, András Rung, Viktor Trón, Péter Vajda | CC BY 2.5 |
| English (UK and US) | SCOWL en_GB-ise and en_US — Kevin Atkinson and contributors | SCOWL licence — en_GB-ise, en_US |
| Dutch | OpenTaal — Nederlandse woordenlijst | CC BY 3.0 — licence text |
| Danish | Det Centrale Ordregister (COR), Dansk Sprognævn | CC0 1.0 |
| Polish | SJP.PL | Apache 2.0 — licence text |
| Swedish | SALDO / SALDOM, Språkbanken Text | CC BY 4.0 |
| Spanish | RLA-ES Hunspell dictionary | MPL 1.1 — licence text |
| French | Grammalecte dictionary — Olivier R. and contributors | MPL 2.0 — licence text |
| Russian | Hunspell ru_RU — Alexander I. Lebedev | BSD 3-Clause — licence text |
| Slovak | sk-spell | MPL 2.0 |
| Serbian | dict-sr — Goran Rakić and contributors | MPL 1.1 — licence text |
| Lithuanian | ispell-lt — Albertas Agejevas and contributors | BSD 3-Clause — licence text |
| Estonian | LibreOffice et_EE: word list by the Eesti Keele Instituut, affix rules by Jaak Pruulmann | EKI licence; LGPL — licence text |
| Finnish | hunspell-fi — Filip Ginter, TurkuNLP | CC0 1.0 |
Frequency, inflection, filtering
Leipzig Corpora Collection
Word frequency rankings for most languages. Universität Leipzig. Licence: CC BY 4.0.
OPUS — OpenSubtitles
Everyday-speech frequency weights in the Hungarian, Dutch and Estonian dictionaries: this is how TypEvent knows which words we use most in everyday speech. Compiled by Jörg Tiedemann and the OPUS project; the corpus comes from the OpenSubtitles.org subtitle collection. Only word frequencies went into the dictionaries, no text. Licence: ODC-BY 1.0.
Tatoeba
Hungarian word-pair frequencies and a supplementary word list. The Tatoeba community. Licence: CC BY 2.0 FR.
Magyar Ispell (Hunspell hu_HU)
Used unmodified as a filter while building the Hungarian dictionary. © László Németh and Ferenc Godó. Licence: MPL 2.0 or LGPL 3+ — licence text. Source: github.com/LibreOffice/dictionaries.
morphdb.hu — about the changes
TypEvent converted the database into its own binary format and used only its inflection classes and word stems. That’s how it knows how to inflect Hungarian words such as bokor, ló or kéz.
LDNOOBW — List of Dirty, Naughty, Obscene and Otherwise Bad Words
These lists tell TypEvent which words not to offer on its own. © 2012–2020 Shutterstock, Inc. Licence: CC BY 4.0.
Emoji and accents
iamcal/emoji-data
Emoji categories and order. Licence: MIT. Source: github.com/iamcal/emoji-data.
Unicode CLDR
Emoji search keywords in Hungarian, English and German, and the set of accented letters on the International layout. © Unicode, Inc. Licence: Unicode License v3.
Publishing the dictionary files
The MPL-licensed sources (Spanish, French, Slovak, Serbian) and the LGPL-licensed Estonian affix rules ask for the modified files made from them to be made available. We’re publishing these files and the recipe used to make them in a public repository; until it’s ready, we’ll send them on request: legal@typevent.app.
The website
Inter
The website’s typeface. © The Inter Project Authors. Licence: SIL Open Font License 1.1 — full licence text. The TypEvent app doesn’t include any external typeface.
If you’re the author of one of these sources and see a mistake on this page, write to legal@typevent.app and we’ll correct it.