Corpus & data files

These are the same paths the app loads under data/. Large .txt files may take a moment in the browser.

words_alpha.txtMain dictionary (~370k when present); app falls back to a remote mirror if missing.
ovilus5_bank.txtPackaged Ovilus-style bank (built via build_ovilus_bank.mjs).
english_10k.txtFrequency / pool building.
google_20k_freq.txtFrequency n-grams.
norvig_count_1w.txtNorvig 1-gram counts.
ovilus5_dd_tokens.txtOvilus DD token list.
names_common.txtBundled names for pools / phrases.
us_places_common.txtUS states & cities.
spoken_phrases_us.txtSpoken phrase lines.
phrase_buckets.jsonPOS buckets for structured phrases.
phrase_slots.jsonLegacy phrase slots.
phrase_pools.jsonLegacy phrase pools.
build_ovilus_bank.mjsNode script to build the Ovilus bank.
gen_phrase_buckets.mjsGenerator script.
gen_phrase_pools.mjsGenerator script.
gen_phrase_slots.mjsGenerator script.

Extra names and word lists are also embedded inside index.html (e.g. curated arrays) — not everything lives under data/.