SigXor — File preview, voice input, screenshot and OCR for Windows

What is SigXor?

SigXor is a desktop productivity tool for Windows 10 and 11, developed by FireNeb and released as open source. It brings the macOS Finder Quick Look workflow to Windows: select a file in File Explorer, press Space, and a preview window appears instantly. Press Space or Esc again to close it.

Beyond file preview, SigXor includes offline voice input, region screenshots, local OCR and a pixel color picker. Speech recognition and OCR run entirely on your machine, file preview needs no cloud service, and the app lives quietly in the system tray.

Read the documentation View source

Developer: FireNeb
Platform: Windows 10 / 11 (64-bit)
Current version: 1.2.0
Price: Free
License: MIT License
Stack: .NET 10 · WPF · WebView2
Data: Local only (no collection, no upload)
Languages: Simplified Chinese / English

The core experience

Spacebar preview

Select any file in File Explorer and press Space. A quick-look window opens instantly; press Space or Esc to close it, or use ↑ / ↓ to move through a multi-selection. No full app launch, no waiting.

Lives in the tray without cluttering the taskbar
Text, code, Office, images, audio and video in one viewer
Move between selected files with ↑ / ↓
Edit and save text or hex content directly

Program.cs — SigXor

using System;

namespace Demo
{
    public class Program
    {
        static void Main()
        {
            Console.WriteLine("Hello, SigXor!");
        }
    }
}

More than a previewer

One lightweight tray app for preview, editing, voice and screenshots

Offline voice input

Tap or hold Right Alt and speak. SenseVoice recognizes Chinese, English, Japanese, Korean and Cantonese offline, then types the result into the focused window.

Region screenshot & OCR

Select any part of the virtual desktop, double-click for full screen, and preview at 1:1. PaddleOCR PP-OCRv6 / v5 runs locally; copy text or save a PNG.

Color picker

A frozen screen and a 15×15 pixel magnifier. Switch HEX / RGB, click or press C to copy, and nudge by one pixel with WASD or the arrow keys.

Text & code

Syntax highlighting, encoding detection (UTF-8 / UTF-16 / GB18030 and more), edit mode, undo / redo, atomic saves and configurable fonts.

Image preview

Preview PNG, JPG, BMP, GIF, WebP and SVG directly, with fit-to-window, actual size and free zoom.

Office documents

Dual-engine preview for docx / xlsx / pptx: a fast structured view and a high-fidelity web view, plus RTF and ODF support.

Hex editor

A custom virtualized control with offset / hex / ASCII columns, byte editing, offset jumps, hex search and multiple copy formats.

Markdown & web

GitHub-flavored Markdown with a one-click source view, and local HTML rendered by WebView2 so relative assets work naturally.

Audio & PDF

Play mp3 / wav / flac / ogg / m4a with seeking, speed and loop controls; PDFs use the built-in Chromium viewer.

Safe and dependable

No lost edits, no large-file stalls, no data leaving your machine

Atomic saves

Every save goes through a temporary file and an atomic replace, with a backup in your Documents folder first. A crash cannot corrupt the original.

Large files

The hex view reads through a virtualized window, so memory use is independent of file size. Files up to 1 GB stay smoothly editable.

Local first

Voice and OCR models run on-device. Audio, screenshots and text are never uploaded. The website has no trackers and the app has no telemetry.

Supported formats

The file types you meet in everyday development and office work

Text & code

.txt .cs .py .js .ts .html .css .json .xml .yaml .toml .md .log .sql .sh .bat .ini .cfg

Images

.png .jpg .jpeg .bmp .gif .webp .svg

Office

.docx .xlsx .pptx .docm .xlsm .pptm .rtf .odt .ods .odp .xls

Media & more

.mp3 .wav .flac .ogg .m4a .aac .pdf .html .htm any binary

See the documentation for the full list and per-format limits. Format details

Shortcuts

Stay on the keyboard

Preview

Space Open / close preview
Esc Close preview
↑↓ Previous / next file

Voice & screenshot

Right Alt Voice input (default)
Alt` Region screenshot (default)
Enter Confirm screenshot selection
Esc Cancel screenshot / picker

Hex editor

0-9A-F Enter hex bytes
Tab Switch hex / ASCII pane
Insert Overwrite / insert mode
CtrlC Copy hex string

Full shortcut reference

Built on modern .NET

Native performance with a carefully chosen stack

.NET 10

modern runtime

WPF

native Windows UI

MVVM

CommunityToolkit

AvalonEdit

text editor

WebView2

Chromium rendering

Open XML

Office parsing

Markdig

Markdown engine

NAudio

microphone capture

SenseVoice

offline speech

PaddleOCR

local OCR

xUnit

unit tests

Implementation details

Download and install

A portable ZIP with built-in updates and rollback

Portable ZIP

Unzip into a writable folder and run. Updates are built in.

Windows 10 / 11 (64-bit)
About 60 MB; voice / OCR models download on demand
Check, download and restart to update; automatic rollback on failure
Requires the .NET 10 Desktop Runtime and WebView2

Download from GitHub Releases

Build from source

Compile it yourself

# Clone the repository
git clone https://github.com/pengcunfu/SigXor.git
cd SigXor

# Build
dotnet build SigXor.sln -c Release

# Run
dotnet run --project SigXor

Build documentation

System requirements

OS Windows 10 or later (64-bit)

Runtime .NET 10 Desktop Runtime, WebView2

Memory 4 GB or more; 8 GB recommended with voice models

Disk About 1 GB; voice / OCR models add about 260 MB

Microphone Required for voice input (optional)

App folder Must be writable for automatic updates

Voice and OCR models are not bundled. They download on first use and then run fully offline.
Upgrading from the old Velopack installer? Uninstall the old version first, then unzip the new build into a fresh writable folder.

Frequently asked questions

Installation, offline capability and updates

Is SigXor free?

Yes. SigXor is completely free and released under the MIT license. The source code is on GitHub, and there is no cost to download or use it.

Do I need a .NET runtime?

Yes. The release is a framework-dependent single file, so the target machine needs the .NET 10 Desktop Runtime and WebView2. The ZIP already contains the webassets and other runtime resources, so it runs as soon as it is unzipped. Voice and OCR models download on first use.

Does voice recognition or OCR upload my data?

No. SenseVoice speech recognition and PaddleOCR text recognition both run locally. Models are downloaded on first use and then work offline. The app collects no telemetry.

How do automatic updates work?

The client reads latest.json from GitHub Releases, downloads the full ZIP and verifies its SHA256. After the app exits, the embedded updater replaces the installation. If the new version is not ready within 30 seconds, it rolls back automatically, so the app folder must be writable.

Which file formats are supported?

More than 30 formats, including text and code, images, Word / Excel / PowerPoint, RTF, ODF, Markdown, HTML, common audio formats and PDF. Any binary file can be opened in the hex view.

How do I uninstall it?

Quit the tray app and delete the program folder. User data stays in Documents/FNSoftware/SigXor; delete that folder too for a complete cleanup.

More troubleshooting

Free, open source, offline first

Keep preview, voice, screenshots and color picking one tray icon away.

Download SigXor Browse all documentation