Take a screenshot and copy text out of it directly from a quick settings tile.
Share a photo or screenshot from another app to TextGrab to make its text selectable.
Use on-device text recognition on a de-googled Android phone with no Google Play services.
Verify a downloaded APK's signing certificate before installing an update.
| notune/textgrab | blindman81/snippets | figurophobia/milk-filter-mobile | |
|---|---|---|---|
| Stars | 4 | 4 | 4 |
| Language | Kotlin | Kotlin | Kotlin |
| Setup difficulty | easy | moderate | easy |
| Complexity | 2/5 | 2/5 | 1/5 |
| Audience | general | general | general |
Figures from each repo's GitHub metadata at analysis time.
Requires manually adding the OCR screenshot tile to quick settings and granting screenshot accessibility for the instant capture feature.
TextGrab is an Android app that lets you select and copy text out of any image, working like the text selection features built into iOS or Pixel phones, except it runs entirely on your device with no internet connection needed. It was made specifically for de-googled Android systems such as GrapheneOS, CalyxOS, and LineageOS, phones that do not have Google Play services installed, so users of those systems can still get this kind of text recognition. There are two main ways to use it. The first is a quick settings tile you add once: swipe down from the top of your screen, tap the OCR screenshot tile, and it takes a screenshot and lets you tap or drag over any text on it to select and copy it. You can also turn on instant capture so the tile takes the screenshot automatically without an extra step. The second way is sharing an existing image to TextGrab from any app, such as your gallery, a browser, or a messaging app, or opening an image file directly with it from a file manager. Under the hood, the app includes Google's ML Kit text recognition model bundled directly inside the app itself, the same type of model that powers Pixel's Live Text feature. Because the model ships inside the app, recognition happens fully on the device, the app requests no internet permission at all, and it works without any Google Play services. It typically recognizes text in under a second. Out of the box it supports English and most other languages that use Latin script, along with digits and common punctuation. Chinese, Japanese, Korean, and Devanagari script are not included yet, since each needs its own separate downloadable model pack of around 20 megabytes, though the developer says adding them is easy on request. The app's own source code is released under the MIT license and can be built from source using Gradle. The bundled Google text recognition model itself is proprietary software from Google, included under Google's own terms, and is downloaded separately at build time rather than stored in this repository. Release builds can be verified against a published signing certificate, and the app can also be installed and kept updated through a tool called Obtainium.
An offline Android app that lets you select and copy text from any image, built for de-googled phones without Google Play services.
Mainly Kotlin. The stack also includes Kotlin, Android, ML Kit.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly general.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.