-
Notifications
You must be signed in to change notification settings - Fork 0
Home
Luc Côté edited this page Jul 6, 2026
·
3 revisions
Prebuilt Tesseract 5.5.0 + Leptonica 1.84.1 AAR for Android. On-device text recognition, no NDK, no rebuild, English bundled.
implementation("dev.ffmpegkit-maintained:tesseract-android:5.5.0")→ Beginner Tutorial — from zero to OCR, step by step, with code for every image source (file, gallery, camera, assets, network).
- Beginner Tutorial — step-by-step guide with copy-pasteable code.
- Quick Start — install, first OCR pass in 5 lines.
- How Image Decoding Works — why there are no image codecs in the AAR (and why JPEG/PNG/WebP/HEIC all work anyway).
- Languages — bundled English, adding more traineddata, model quality.
-
API Reference —
TesseractOCR,TesseractConfig,TesseractResult. - Performance — image prep, PSM modes, release vs debug, OpenMP.
- FAQ — common issues (empty text, low confidence, missing language).
- Pro Integration — 12 languages, best models, OpenMP, batch API.
- Building From Source — submodules, NDK, CMake.
| Free | Pro | |
|---|---|---|
| Engine | Tesseract 5.x + Leptonica | Tesseract 5.x + Leptonica |
| Bundled languages | English (1) | 12 (eng, fra, spa, deu, ita, por, chi_sim, jpn, kor, ara, rus, hin) |
| Add more languages | ✅ 100+ downloadable | ✅ 100+ downloadable |
| Model quality |
tessdata_fast (~4 MB) |
tessdata_best (~15 MB, max accuracy) |
| OpenMP | ✗ single-thread | ✅ multi-thread (2–3× faster) |
| Core API | ✅ recognize, recognizeFile, confidence, bounding boxes |
✅ same |
| Advanced API | ✗ | ✅ hOCR, batch OCR, per-character confidence |
| ABI | arm64-v8a | arm64-v8a + x86_64 |
| Image formats | JPEG, PNG, WebP, HEIC (via Android) | same |
| Channel | Maven Central / JitPack / GitHub Release | Gumroad |
| Price | $0 | $24 · $62 team |
→ Get Pro
Maintained by Jokobee.