Skip to content

About

Public release artifacts for the Echophrase desktop app

Resources

Stars

1 star

Watchers

0 watching

Forks

Latest commit

 

History

21 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 

Repository files navigation

Echophrase

Privacy-first, GPU-accelerated local AI dictation

Your voice never leaves your computer.

Latest release Homebrew winget Website Discord


Echophrase is a desktop dictation app that turns your speech into text in any application — fully offline after a one-time model download, accelerated by your GPU, with multilingual models (Whisper, Parakeet, Qwen3-ASR). Push-to-talk, no account, no cloud - a polished, featherweight native app (Tauri + Rust, not Electron).

This repository holds release artifacts only — no source code. Signed, notarized macOS .dmg builds are published as GitHub Releases, tagged v<version> to match the app. It exists so the Homebrew tap has a public, versioned, checksummable URL to install from.

🍺 Install (macOS, Apple Silicon)

brew install --cask imperium42/tap/echophrase

Requires macOS 13.3+ (Ventura) on Apple Silicon (arm64). Prefer a manual install? Grab the .dmg from the latest release.

🪟 Install (Windows)

Download the signed installer (Echophrase_<version>_x64-setup.exe) from the latest release. GPU acceleration works out of the box via DirectML; an optional CUDA plugin for NVIDIA GPUs is available in-app.

winget install Imperium42.Echophrase

🐧 Install (Linux, Alpha)

Grab the AppImage (x86_64, NVIDIA CUDA required), then chmod +x and run.

On Debian/Ubuntu you can install the .deb package instead:

sudo dpkg -i echophrase.deb

Linux support is Alpha — see echophrase.com/download for requirements and known limitations.

✨ Why Echophrase?

🔒 Unlimited Offline Transcriptions Your voice is processed on-device — recordings are never uploaded (models are a one-time download on first run). Optional online features (account, cloud sync, anonymous telemetry) can be turned off
⚡ GPU-accelerated Native DirectML on Windows, Metal on macOS — nothing extra to install
🌍 Multilingual Pick the model that fits: Whisper, Parakeet, or Qwen3-ASR — from fast English-first dictation to strong Chinese & international accuracy
✍️ Grammar cleanup An optional post-transcription transform model polishes grammar and punctuation, so your dictation reads the way you meant it
🎙️ Push-to-talk Dictate into any text field with a hotkey
🪶 Featherweight Native Tauri + Rust — not an Electron memory hog

🔗 Links


© Imperium42 LLC · Privacy · Terms

About

Public release artifacts for the Echophrase desktop app

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors