macOS · Apple Silicon · Local Transcription · Actively maintained

Your voice.
Instant text.

Native macOS push-to-talk app for anyone who dictates on a Mac: it turns speech into text in under a second and can live in the menu bar, Dock, or both. Push to talk, release to paste. Local transcription with optional cleanup via Apple Intelligence, GitHub Copilot, or an OpenAI-compatible API.

<1s
Local transcription latency
190×
Realtime factor
25
Languages
On-device
Apple AI cleanup

Built for speed
and nothing else

No browser. No Electron. No Python runtime. Just native Swift talking directly to the Apple Neural Engine. Every millisecond matters.

Push to Talk

Hold the Fn key to record, release to transcribe. Natural as a walkie-talkie. No clicking, no UI.

Auto-Paste

Text appears at your cursor in whatever app is focused. TextEdit, browser, terminal, Slack—anywhere.

Local Transcription

Audio is always transcribed on-device via CoreML. Apple Intelligence cleanup also stays local; cloud providers receive transcript text only when selected.

Neural Engine

The encoder runs on Apple’s dedicated Neural Engine via CoreML. 190× realtime on M-series chips.

Live Waveform

A compact borderless waveform follows your voice and clearly distinguishes Listening, Enhanced, Processing, and Done states.

25 Languages

Powered by NVIDIA Parakeet TDT v3 supporting English, German, French, Spanish, and 21 more European languages.

AI-powered
transcript cleanup

Optionally clean up transcripts using Apple Intelligence, GitHub Copilot, or an OpenAI-compatible API—fix grammar, remove filler words, and polish punctuation while preserving meaning, tone, and language.

Apple Intelligence

Local & Private

Fast on-device cleanup on compatible Macs running macOS 26. No transcript text leaves your machine and there is no token cost.

GPT-5.6 Luna

Cloud Default

The evaluated default for strong cleanup quality at low token cost through GitHub Copilot CLI.

Gemini · MAI · Claude

Choose Your Model

Switch among evaluated models, see current price tiers, and use a model-tuned default prompt or your own update-safe customization.

GitHub Copilot enhancement requires the GitHub Copilot CLI installed and authenticated; OpenAI-compatible providers use the configured endpoint. Enhancement can run only with Shift or for every recording, with Shift as a one-time bypass.

Three steps.
Under one second.

The local pipeline—from microphone to pasted text—completes before you finish lifting your finger. Enhanced Mode adds provider processing before paste.

01

Hold Fn

Press and hold your configured hotkey. The mic activates instantly and the bottom waveform follows your voice. Purple Enhanced feedback confirms when cleanup will run.

02

Release

Let go of Fn. The audio buffer is passed to the Parakeet TDT model running on the Neural Engine via CoreML. Transcription finishes in milliseconds.

03

Pasted

The local transcript—or the provider-cleaned text when Enhanced Mode is enabled—is placed on the clipboard and a Cmd+V is simulated. Text appears at your cursor. Your original clipboard is silently restored.

Built on proven
foundations

OpenWritr stands on the shoulders of exceptional open-source work.

FluidAudio

CoreML-optimized speech recognition framework by FluidInference. Provides the Parakeet TDT 0.6b v3 model compiled for Apple Neural Engine with ~190× realtime performance.

NVIDIA Parakeet TDT v3

State-of-the-art 600M parameter ASR model using Token-and-Duration Transducer architecture. Non-autoregressive decoding for minimum latency. 25 European languages.

Apple CoreML & Neural Engine

Hardware-accelerated inference on Apple Silicon’s dedicated Neural Engine. The encoder model runs entirely on the ANE for maximum throughput and minimal power draw.

Swift 6 & SwiftUI

Pure native implementation with strict concurrency safety. MenuBarExtra for the system tray, AVAudioEngine for low-latency capture, CGEventTap for global hotkeys.

Up and running
in 30 seconds

Download the latest release, or build it from source. Requires macOS 14+ and Apple Silicon. The ASR model (~460 MB) downloads automatically on first launch.

Terminal
# Download, move OpenWritr.app to /Applications, then open it
# Grant Microphone + Accessibility permissions when prompted
# Hold Fn to speak, release to paste
# Build from source: see the README