peteonrails/voxtype

Voice-to-text with push-to-talk for Wayland compositors

What it solves

Voxtype is a local voice-to-text tool for Linux that allows users to dictate text directly into any application at the cursor position. It eliminates the need for cloud-based transcription services, providing a private, low-latency alternative for system-wide dictation.

How it works

The tool runs as a background daemon that listens for a configured hotkey (push-to-talk or toggle). When the hotkey is pressed, it records audio and, upon release, transcribes the audio using one of several local AI engines. It then simulates typing to insert the text into the active window. It integrates deeply with Linux desktop environments, including support for Wayland and X11, and can automatically pause media players via MPRIS.

Who it’s for

Linux users who want a fast, private, and highly customizable voice-to-text experience without relying on cloud subscriptions or telemetry.

Highlights

  • Multiple AI Engines: Supports Whisper, Parakeet, Moonshine, SenseVoice, Paraformer, Dolphin, and Omnilingual, covering over 1600 languages.
  • High Performance: Optimized for CPU and GPU (Vulkan, CUDA, Metal, ROCm) acceleration, achieving up to 11x real-time speed on modern CPUs.
  • Local-First: No cloud, no subscriptions, and no telemetry; audio stays on the machine.
  • Advanced Text Processing: Includes built-in word replacements, spoken punctuation conversion, and the ability to pipe transcriptions through external LLMs for grammar correction.
  • Meeting Mode: Provides continuous transcription with speaker attribution and exports to formats like Markdown, JSON, and SRT.
  • Native Linux Integration: Deep support for various compositors (Hyprland, Sway, GNOME, KDE) and a floating waveform OSD for visual feedback.

Related

  • Project
    goodroot/hyprwhspr

    A native Linux speech-to-text tool that provides fast, private, and system-wide dictation using local or cloud-based AI models.

  • Project
    moinulmoin/voicetypr

    An offline-first voice-to-text dictation app for macOS and Windows that transcribes speech and inserts it directly into any active application.

  • Project
    VocaHQ/vocalinux

    A privacy-focused, on-device voice-to-text application for Linux that allows system-wide dictation into any app using engines like whisper.cpp and VOSK.

  • Project
    akshayaggarwal99/jarvis-ai-assistant

    An open-source voice dictation tool that converts speech to text anywhere on your system, supporting local transcription and LLM post-processing for privacy and speed.

  • Project
    OpenWhispr/openwhispr

    OpenWhispr is a cross‑platform desktop app that converts speech to text, provides live meeting transcription with speaker diarisation, and lets you talk to AI assistants (GPT‑5, Claude, Gemini, etc.) directly from your keyboard. It works offline with Whisper‑based models or via cloud services, stores notes locally (with optional cloud sync), and offers an API for programmatic use.