How common a word is, the hedgehog stages and our estimate of how many of the words in a typical film you already know are based on wordfreq by Robyn Speer. Its data is licensed under CC BY-SA 4.0 and is built from:
What we changed: we take the 40,000 most common forms of each language, remove names, abbreviations and fragments, group forms under their dictionary word and rank words by the summed frequency of their forms.
Links between word forms, alternative spellings and the "Other translations" list in the word card come from Wiktionary, written by Wiktionary contributors and licensed under CC BY-SA 4.0 and the GNU Free Documentation License. We use the machine-readable dumps of kaikki.org (Wiktextract by Tatu Ylonen). What we changed: we keep only word forms, spellings and translations and store them in our own index; the definitions shown in the card are written for Lalea and do not come from Wiktionary.
Dictionary forms are also found with LemmInflect by Brad Jascob (English, MIT license) and Simplemma by Adrien Barbaresi (MIT license).
When we decide which very common words are too simple to show as new, three research word lists help us keep real vocabulary in:
These lists are used only to check our own list of skipped words; they are not shown in Lalea or shared.
If you are an author of any of these resources and see a mistake in how we credit your work, write to a contact email will be published here before public launch and we will fix it.