Skip to main content
AdOpenFree logoPromote your productReach more potential users and drive product growth and revenue.Advertise
Favicon of Hex

Hex

Free Listing

Hex is local-first voice dictation for macOS, Windows, and Linux. Hold a shortcut, speak, and Hex transcribes on-device and pastes the text into the app you are using, with optional modes and CLI tools.

Visit anomalyco/hex

Overview

Hex is a local-first voice dictation application built in Rust. It lets users hold a shortcut, speak, and paste transcribed text into the foreground app. The macOS app is available now, with Windows and Linux betas. Ordinary dictation uses on-device speech recognition; optional OpenCode-backed features can rewrite or process transcripts when an OpenCode service is already running.

Key Features

  • On-device speech transcription using downloadable local models; ordinary dictation does not require OpenCode or a cloud transcription service.
  • Configurable shortcut, microphone, language, local model, recording behavior, feedback volume, launch-at-login behavior, and Dock visibility.
  • Modes provide corrections, application-specific behavior, and optional text processing or rewriting.
  • Annotate records screen drawing and speech, then offers paste or send-to-OpenCode actions on macOS and Linux, not Windows.
  • CLI queries expose dictation history and meeting transcripts, while custom TypeScript or JavaScript transformations run in an embedded engine.
  • The TypeScript SDK lets other applications use Hex for local transcription when supplied a compatible Hex executable.

Use Cases

  • Dictating text into any focused application with a hold-to-talk shortcut.
  • Fixing names and preferred spellings before paste with Modes corrections.
  • Capturing annotated screen recordings with voice explanation for review or sending to an agent.
  • Reading recent dictation history or meeting transcripts from the terminal on supported builds.

Getting Started

  • macOS: install with Homebrew using brew install --cask anomalyco/tap/hex, or download the signed DMG, drag Hex.app into Applications, launch it, and complete microphone, Accessibility, and transcription model setup.
  • Windows: download the x64 or ARM64 installer and run it; it installs for the current user without administrator rights.
  • Linux: use the Linux install script, Debian or Fedora packages, or a binary, then follow the Linux guide for shortcuts and compositor setup.
  • Optional OpenCode features require an OpenCode service that is already running and a configured model provider.

Deployment & Requirements

  • macOS requires Apple silicon and macOS 15 or later.
  • Windows beta runs on Windows 10 or 11 on x64 or ARM64; x64 needs a 2013 or newer processor with AVX2 (Intel Haswell / AMD Excavator or later).
  • Linux beta targets x86_64 Linux on i3/X11 and compatible wlroots Wayland compositors; x86_64 needs a 2013 or newer processor with AVX2.
  • Native Wayland requires compositor support for clipboard and layer-shell protocols; paste uses wtype with the virtual keyboard protocol or the desktop's RemoteDesktop portal.
  • Network access is required for model installation and software updates; local model setup also needs huggingface.co and its model download CDN.

Before You Adopt

  • License: MIT. Review its terms before using, modifying, or distributing the project.
  • Windows support is a beta and runs on Windows 10 and 11 for x64 and ARM64.
  • Linux support is a beta with processor and compositor requirements.
  • History retention stores pasted dictation text and limited metadata for seven days by default, but does not store audio, full browser URLs, or window titles.
  • Model installation and software updates require network access.

Comments

Sign in to leave a comment.

More like Hex

Favicon of whisper.cpp

whisper.cpp

Free ListingStars: 53.9K

High-performance C/C++ inference for OpenAI Whisper

Transcription & Dictation

whisper.cpp is a dependency-light C/C++ port of OpenAI's Whisper automatic speech recognition model, optimized for high-performance inference across CPUs, Apple Silicon, GPUs, NPUs, and edge platforms.

Favicon of Handy

Handy

Free ListingStars: 34.7K

Free, open source, offline speech-to-text for your desktop

Transcription & Dictation

Handy is a free, open source, cross-platform desktop speech-to-text app that runs fully offline and pastes transcribed speech into any text field without sending audio to the cloud.

Favicon of VidBee

VidBee

Free ListingStars: 10.8K

Download videos and create searchable transcripts locally.

Assistants & ProductivityTranscription & Dictation

VidBee downloads video and audio from YouTube, TikTok, Instagram, X, and 1000+ sites, or local files. It creates searchable, timestamped transcripts on your computer, then summarizes, translates, or answers questions with your AI provider.