Features
Happy Hanzi reads Chinese wherever you meet it — a menu, a sign, a page of homework, a character you can only half-remember — and everything happens on your device, offline.
Point your camera at it
Photograph Chinese text or pick a photo from your library. Lines are detected on-device, grouped into paragraphs, translated, and split into dictionary words you can tap — each with its pinyin and meaning underneath. Pinch to zoom the photo; tilted or angled text is traced correctly, and the picture tucks away as you read.
Write it, find it
Draw with a finger or Apple Pencil on an ink-brush canvas that responds to speed and pressure — rest the pen and the ink pools and bleeds. Candidates appear as you write: two recognisers work together, Apple’s on-device text recogniser and a stroke matcher built from stroke-order data, so even one-stroke characters like 一 and 十 are found. Zoom out and pan across a grid of guide cells to write whole sentences.
Translate by typing
Type English or Chinese and translate on-device with Apple’s translation models (a one-time language-pack download). Every Chinese result carries a pinyin line, and offline dictionary matches from CC-CEDICT appear alongside — tap any of them to explore the characters.
Pinyin and meanings
Every character shows its readings and English senses from CC-CEDICT, together with the words it appears in.
Radicals and components
See the Kangxi radical a character is filed under and every component it’s built from — each one a tap away, right down to the smallest part. Browse all 214 radicals by stroke count, with their variant forms and example characters.
Stroke by stroke
Watch the stroke order animate, then read each stroke’s name — 横 héng, 竖 shù, 撇 piě, 捺 nà and the rest — with a little diagram highlighting where it sits.
Hear it
Tap the speaker to hear the character pronounced using the Mandarin voice built into iOS. No download, no connection.
Completely offline
The dictionary, stroke data and recognition all live on your device. It works on a plane, on the underground and in a classroom with no Wi-Fi. There are no accounts, no ads and nothing is collected.
Open data, credited
Dictionary entries come from CC-CEDICT (CC BY-SA 4.0). Character decomposition and stroke data come from makemeahanzi (LGPL / Arphic Public License).