The Intelligent Reading Specialist
LUCADictionary is a comprehensive database of 763,000+ grapheme-phoneme mappings that powers LUCA's phoneme-level speech recognition. Identification at the sound level, not just the word level.
.png&w=256&q=75&dpl=dpl_FwNuK7RWgbrkS4Pu7k1mSMLJPziw)
A letter-sound mapping database (also called a grapheme-phoneme mapping database) connects written letters (graphemes) to their spoken sounds (phonemes). LUCADictionary connects written symbols to their spoken sounds across the full complexity of English spelling, from simple consonants to complex vowel teams and Latin/Greek roots.
Word-level tools detect errors but can't identify them. They know a word is wrong but not why, leaving teachers guessing about which skill gaps to address.
Marks the word as incorrect. Full stop. No data on whether it was a speed error or a systematic digraph confusion.
Identifies the specific /sh/ → /s/ substitution. Signals a clear skill gap in consonant digraph recognition that triggers targeted instruction.
Every letter and combination encountered, simple consonants to complex vowel teams.
Positional variations: the "c" in "cat" differs from "c" in "city". Exact phonetic modeling.
All Dolch and Fry sight words are fully mapped and integrated into the skill-identification layer.
Prefixes, suffixes, and Latin/Greek roots mapped systematically for advanced word study.
| Credential | Impact |
|---|---|
| Carnegie Mellon Partnership | Architecture developed with CMU's AI program for technical rigor. |
| NSF Validation | NSF SBIR funding (3% acceptance) validates innovation and impact. |
| Patent Protection | U.S. Patent No. 12,394,332 B2 protects LUCA's mapping technology. |
| Ehri (2014) Alignment | Built to support orthographic mapping, mapping graphemes to phonemes. |
A grapheme is a letter combination that represents a sound; a phoneme is the sound itself. Mapping connects written symbols to spoken sounds. LUCADictionary contains 763,000+ of these, ensuring precision for any word a student reads.
Standard dictionaries define meanings. LUCADictionary maps the sound structure, connecting letters to sounds at the phoneme level. This is linguistic intelligence for skill identification, not just vocabulary lookup.
Word-level tools tell you a student missed a word but not why. Phoneme-level assessment reveals the pattern (e.g., /sh/ digraph confusion) that should guide intervention. NSF-funded research shows this detects 40% more errors.
Dyslexic readers often have specific sound substitutions that word-level tools miss. LUCADictionary identifies these patterns, ensuring intervention targets the actual skill gaps aligned with IDA best practices.
Yes. It supports orthographic mapping (Ehri, 2014) and was validated through NSF SBIR funding and built in collaboration with Carnegie Mellon researchers's AI program.
Witness the foundation that makes LUCA's skill-identification precision possible. See our evidence or book a demo.