Credits

Numlore builds on the following third-party data sources and libraries.

CMU Pronouncing Dictionary

ARPAbet phonetic transcriptions for approximately 134,000 English words.

Princeton WordNet 3.1

Lexical database providing parts of speech and semantic hierarchies.

wordfreq

Used at build time to compute Zipf word frequency. Development dependency only.

Brysbaert, Warriner, & Kuperman (2014)

Concreteness ratings for 40,000 generally known English word lemmas. Behavior Research Methods, 46(3), 904–911.

Glyphex

Aggregate page analytics. The only third-party script Numlore loads — used to count visits, not to identify visitors. See /privacy for the contract.