Credits & attribution
Čechovač is built on freely available linguistic data and open-source software. This page lists the sources as required by their licenses. The non-commercial (NC) clause of several data licenses means this site is offered free of charge, without advertising.
Linguistic data
- MorfFlex CZ 2.1 & PDT-C 2.0 — morphological dictionary and tagger model for Czech, used for all inflection lookup. Czech Models 250909, Straka & Straková, ÚFAL MFF UK, LINDAT handle 11234/1-5985. License: CC BY-NC-SA 4.0.
- VALLEX 4.5 — valency lexicon of Czech verbs, used for reflexivity notes and as a source for aspect counterparts. Lopatková, Markéta; Kettnerová, Václava; Mírovský, Jiří; Bejček, Eduard; Barančíková, Petra; Straka, Milan; Hajič, Jan, 2022, VALLEX 4.5, LINDAT/CLARIAH-CZ, handle 11234/1-4756. License: CC BY-NC-SA 4.0.
- DeriNet 2.3 — Czech derivational network, used for adjective → adverb links. Olbrich, Michal; et al., 2025, DeriNet 2.3, LINDAT/CLARIAH-CZ, handle 11234/1-5846. License: CC BY-NC-SA 4.0.
- English Wiktionary via kaikki.org's Czech extract, used for aspect counterparts. License: CC BY-SA 4.0 and GFDL.
Built with Django, MorphoDiTa, htmx and Tailwind CSS, and typeset in
IBM Plex Serif and Inter. All open source; the full list with
licenses lives in LICENSES.md.
The linguistic data files themselves are not part of the repository;
they are built on the deployment server (see
LICENSES.md).
Privacy
This site sets no cookies and stores no personal data about you.
Page views are counted with a self-hosted, cookieless analytics
instance (GoatCounter): IP addresses are hashed and used only for
the daily unique-visitor count, and the counted page addresses
include the searched word (e.g. /lookup?q=dobrý). No
third party receives any data about your visit.