Phonics Pattern Toolkit

Build word lists for onset/rime families, vowel teams, silent-e pairs, and CVC-style structures; check pasted text against the patterns you have taught; and generate pronounceable pseudowords for practice. Matching is heuristic and spelling-first, cross-checked against our pronunciation data where a conventional mapping exists.

About this toolkit & its limits

English spelling is irregular, and this toolkit is a heuristic helper, not a grapheme-to-phoneme aligner. A word joins a pattern list because of how it is spelled — for example, any word ending in the letters “at” after consonant letters joins the -at family. Where the pattern has one conventional sound and our pronunciation data covers the word, we cross-check and mark each word verified or check pronunciation; words without pronunciation data stay unverified. Every list shows these coverage numbers — read them before using a list.

Data sources and bounds

  • Word lists come from this site’s compiled lexicon (see the data page for the pinned source lists and their limits). Membership in a list here never implies validity in any commercial word game or official curriculum.
  • Pronunciation checks use the pronunciation entries retained by our data build (derived from CMUdict), which primarily reflect North American English; dialects differ, so a mismatch flag is a prompt to check, not a verdict.
  • “Commonness” ordering uses this build’s internal frequency bands, not universal usage claims.

Known approximations

  • Vowels are a/e/i/o/u; the letter y is always treated as a consonant (wrong for words like “gym” or “fly”).
  • The decodable checker splits contractions at the apostrophe (“don’t” becomes “don” + “t”) and checks spelling patterns only.
  • Ambiguous vowel teams (ea, ow, oo) are never marked mismatched; their lists show the observed sound split instead.
  • Pseudowords are machine-built letter strings filtered against our lexicon and screened against our offensive-word list; preview any list before classroom use.

Results computed from data version v1/2026.07-lexical.6.