Sources & licenses
Every prompt comes from an openly licensed corpus. None were written for this project. Source, ID, author and license are stored with each prompt and included in dataset exports.
Tatoeba
CC BY 2.0 FRTagalog sentences. The contributor's username is kept with every prompt.
Mozilla Common Voice
CC0 1.0Tagalog sentences from the Common Voice sentence collector.
Wiktionary (via kaikki.org)
CC BY-SA 4.0Tagalog headwords. Proper nouns, affixes and inflected forms are excluded.