How Glyphin works
Most apps drop you in front of an alphabet chart and ask you to memorize it. Glyphin starts with words you'd actually see on a street sign, a menu, or a packet — and teaches the letters as you go.
The usual approach
Traditional courses march you through the alphabet in its native order — letters you rarely see first, common ones buried deep, and no real words for weeks. You burn through your willpower long before anything clicks.
Reading is more than letter recognition. It's pattern matching against text you care about. Start with the patterns; the letters follow.
How we do it
Each one is simple. Together they make the script feel useful from day one.
Every lesson is built around one word you'd actually meet outside — a place, a food, a sign. You learn the letters by reading something real.
Learning researchers call automatic word recognition orthographic mapping. Words become instant after you've sounded them out a handful of times — but only if the letter-sound link is real, not memorized as a shape. Anchoring every lesson to a meaningful word gives your brain that link from the start.
A small handful of letters does most of the work. We sequence them so the first few hours of study unlock the biggest share of everyday text.
The usage-based view of language learning says you acquire what you see often, faster. In most writing systems, a small fraction of the letters covers the majority of everyday text. Teaching those first means every minute of study returns more text you can actually read.
Each lesson introduces only the new letters that word needs — usually two or three. Old ones come back as review so they actually stick.
Working memory only holds about four new chunks at a time. Cognitive load research shows that overloading it stalls learning — you remember nothing instead of something. Smaller, cumulative sets let each piece settle before the next arrives.
Every lesson ends with short drills — pick the letter, match the meaning, spot the look-alike. Retrieving an answer is what locks it in.
This is the testing effect, plus what Robert Bjork calls desirable difficulties. Pulling information out of your head is harder than re-reading it, and that extra effort is exactly what builds durable memory.
Inside a lesson
The same shape for every lesson so you always know what's next.
Meet the anchor word and where you'll see it.
See how the word splits into readable pieces.
Learn just the new letters this word needs.
Spot the pattern that makes the word readable.
Short checks to lock in the new material.
Score, summary, next lesson unlocked.
What we're still working on
The research that backs our approach also points to places we can get better. Here's what's next.
Multiple-choice drills train recognition really well, but production lags. We're adding type-the-word and tap-to-build drills so reading turns into writing too.
Old letters come back inside new lessons today, but we don't yet schedule them on the intervals the memory research recommends. A standalone review surface is coming.
Some scripts encode features — tones, stress, length — that are fundamentally a sound, not a shape. Native-speaker audio for every anchor word will close that gap.
Vocabulary sticks after eight to twenty meaningful encounters. We're adding short, decodable bits of real text — a menu line, a sign, a transit display — that you can read between lessons.
Eight minutes. One real word. A handful of letters you'll see everywhere after.