goodroot/hyprwhspr

Native speech-to-text for Linux - Fast, accurate, private, and hackable system-wide dictation

What it solves

hyprwhspr provides a native, system-wide speech-to-text (STT) dictation tool for Linux users. It eliminates the need for cloud-dependent dictation by offering high-performance local transcription that is private, fast, and integrates directly into any active text buffer on the OS.

How it works

The tool captures audio via the microphone and processes it using a variety of configurable backends. It supports local in-memory models for offline privacy and speed, as well as cloud APIs for those who prefer them. It utilizes ONNX-ASR for CPU optimization and supports CUDA or Vulkan acceleration for GPU-powered transcription. Once transcribed, the text is automatically pasted into the active window using clipboard and window tools like wl-clipboard or xclip.

Who it’s for

Linux users (Arch, Debian, Ubuntu, Fedora, openSUSE) who want a fast, private, and hackable dictation system that works across both Wayland and X11 sessions.

Highlights

  • Multiple Backend Support: Compatible with Cohere, Parakeet, Whisper, Qwen3-ASR, Gemini, and ElevenLabs.
  • Privacy First: Offers strictly offline transcription so data never leaves the machine.
  • Linux Native: Built specifically for Linux with native AUR packages and systemd integration.
  • High Performance: Optimized for both high-end NVIDIA GPUs and CPU-only hardware via ONNX.
  • User Experience: Includes a themed voice visualizer, audio ducking to reduce background noise during recording, and auto-paste functionality.
  • Multilingual: Strong support for CJK (Chinese, Japanese, Korean) and built-in translation capabilities.

Related

  • Project
    OpenWhispr/openwhispr

    OpenWhispr is a cross‑platform desktop app for privacy‑first voice dictation, meeting transcription, and AI‑assistant interaction. It supports offline speech‑to‑text with Whisper‑based models or cloud services, offers speaker diarisation, note‑taking with semantic search, team collaboration, and a public API for developers.

  • Project
    VocaHQ/vocalinux

    A privacy-focused voice dictation app for Linux that converts speech to typed text locally on-device using engines like Whisper and VOSK.

  • Project
    QuentinFuxa/WhisperLiveKit

    An ultra-low-latency, self-hosted speech-to-text pipeline that enables real-time transcription and translation using advanced simultaneous speech research and multiple model backends.

  • Project
    homelab-00/TranscriptionSuite

    A free, open-source transcription app that converts speech to text locally on your computer using various AI models for privacy and speed.

  • Project
    TypeWhisper/typewhisper-mac

    A system-wide speech-to-text and AI text processing utility for macOS that supports both local on-device models and cloud APIs for transcription and transformation.