Universal Screen Reader

Provides a powerful interface to read text aloud from images or the clipboard.

Key features:

• Automatically extract text from images using Tesseract OCR
• Read text from any application by taking a screenshot, then opening it from the clipboard
• Keep the reader window in front or use it as an overlay
• Optionally, automatically read newly copied text and images from the clipboard - even when the reader is minimized
• Clean & simple interface which matches system transparency and dark mode settings
• Supports Windows, macOS, and Linux (*Linux is missing transparency)

This can be useful as:

• An accessibility feature for sighted people who have trouble reading,
• A tool for reviewing or revising text
• A tool for speed-reading

... or if you simply prefer listening to reading.

Special thanks to Alex Allen for the idea, testing and feedback.

Installation

Go to releases and download the appropriate installer for your platform. "Universal Screen Reader" can then be opened as you would open any other application.

Usage

Select a voice which matches the language of your text.

Open some text from the clipboard, or open an image from the clipboard or a file to automatically extract text.

Then, speak it aloud.

Screenshotting Text

You can open images from your clipboard with a single button press (or, optionally, automatically). This lets you conveniently read out text from any format by screenshotting part of your screen.

On Windows, use Win+Ctrl+S and then select a region. You can also screenshot a window, or the entire screen.

On macOS, use Ctrl+Shift+Cmd+4 and then select a region. This keyboard shortcut may be changed in macOS's System Preferences, under Keyboard -> Keyboard Shortcuts -> Screenshots

On Linux, you may be able to use Shift+Ctrl+PrintScreen. If not, look into gnome-screenshot

Downloading OCR Models

The OCR language is selected based on the current voice's locale. The first time you open an image in a new language, an internet connection is required to download the Tesseract OCR model for that language.

Alternatively, you can download models here and place them in a folder named TesseractData in the path from which the jar is executed.

More Info

For an in-depth explanation of all controls and shortcuts, check FUNCTIONS.md

For documentation on building or running from source, check BUILD.md