Glyfo

Name: Glyfo — Download free. Text recognition
Availability: InStock
Rating: 5 (1 reviews)
Author: LVLE

Glyfo extracts editable text from images, screenshots, PDFs and scanned documents on Windows. It captures screen regions with a keyboard shortcut, opens files or folders for batch processing, reads QR codes and barcodes from the same picture, keeps a searchable history, reads results aloud with installed voices, and restores paragraph breaks or removes unwanted spaces from CJK recognition output. Recognition runs on the OCR engine included with Windows, and the program makes no network connections.

★★★★★

5.0(1 ratings)

Download Glyfo (Official links)

File size: 42.7 MB

The latest version of Glyfo is: 1.3.0

Operating system: Windows

Languages: English, Spanish, Portuguese, German, French, Chinese

Developer: LVLE

Price: $0.00 USD (Open Source (MIT))

Screen region capture. Activates a selection overlay that lets the operator drag a rectangular box around any visible part of the display. The marked area is captured as a bitmap and passed to the recognition engine. A separate shortcut captures the entire screen in one action. Both shortcuts can be reassigned by the operator through the settings interface.
Clipboard image paste. Reads image data directly from the Windows clipboard when the operator presses the paste shortcut. Plain text already on the clipboard is also accepted and inserted into the result area without recognition. This removes the need to save a temporary file before processing a copied screenshot.
File and PDF opening. Opens common image formats and PDF documents through a standard file dialog or by dragging a file onto the main window. A PDF is displayed page by page, and the current page is recognised on demand. The operator can move through the document with navigation controls and process each page separately.
Whole-document PDF recognition. Sends every page of an open PDF through the recognition engine in a single operation rather than one page at a time. The resulting text is collected in page order so the operator obtains a complete transcript of the document without repeating the recognition step manually for each page.
Folder batch processing. Accepts a folder or a group of image files dropped onto the window and recognises each item in sequence during one run. This is intended for stacks of scans or photographed pages that would otherwise require opening and processing one file at a time. Progress is shown while the batch is running.
Text export to .txt and .md. Saves recognised output as plain text or Markdown. The operator chooses between a single combined file for the whole document or one file per page, which suits multi-page PDFs and folder batches. The file name and destination are selected through the standard save dialog.
QR code and barcode decoding. Scans the same image that is submitted for text recognition and decodes any QR codes or barcodes it contains. The decoded payload is presented alongside the recognised text, so no separate scanning program has to be opened for images that contain both machine-readable codes and printed words.
Searchable capture history. Stores every recognition result in a local database that persists after the program is closed and restarted. The operator can search previous captures by their text content and reopen them. The history holds text only, can be emptied at any time, and can be disabled entirely in the settings.
Read-aloud output. Passes the recognised text to the Windows speech synthesis system and reads it aloud using any voice installed on the computer. This provides an audible version of the captured content without exporting it to another application or converting it to an audio file first.
Paragraph reconstruction. Detects hard line breaks produced by the recognition engine and joins the affected lines back into continuous paragraphs. The operator can toggle this behaviour, which is relevant for prose documents where each visual line would otherwise appear as a separate paragraph in the exported text.
CJK space removal. Strips the artificial spaces that recognition engines insert between Chinese, Japanese and Korean characters. Those scripts do not use spaces between words, and the inserted gaps make the output unsuitable for reuse. The option can be switched on for documents in those languages and left off otherwise.
Recognition language selection. Chooses the recognition language from the OCR language packs installed on the computer, including Chinese, Japanese, Korean, Cyrillic and Greek among others. The operator can select a specific language or let the program determine it automatically from the image content.
Clipboard watching mode. Monitors the clipboard for new screen captures when enabled. Every snip taken with the Windows snipping shortcut is recognised automatically, and the resulting text is placed back on the clipboard ready to paste into another application. The mode is off by default and must be switched on by the operator.
Notification area operation. Keeps the program running in the notification area after the main window is closed, with the capture shortcut still active. A capture can therefore be started without opening the full interface. The interface is available in 33 languages and follows the Windows language setting, with light and dark display modes.

Glyfo is developed by an independent author and distributed through the Microsoft Store and the project site. The program was created in the 2020s and is written in C# for the Windows platform, using the OCR engine that ships with the operating system as its recognition backend. It is released as free software under the MIT licence, with no account requirement, no telemetry and no advertising. Development continues with updates that follow changes in the Windows OCR components and the Microsoft Store distribution channel.

Alternatives to Glyfo:

Greenshot — Free Download. Screenshot capture

Glyfo

Alternatives to Glyfo:

Greenshot

HyperSnap

SnipFor

GClaude Indexer