OpenDMG
Utter icon

Utter for Mac

Voice dictation for Mac that runs on-device: hold a key, speak, words appear at the cursor.

By h3x-levi. Free and open source, Apache-2.0 licence.

Get it with OpenDMG View source on GitHub

Utter at a glance

Version
1.2.3
Released
15 October 2026
Download size
193 MB
Works on
Apple silicon and Intel
Licence
Apache-2.0
Category
Audio and Music
Developer
h3x-levi
GitHub stars
3

About Utter

Utter-Whisper

How to install Utter on a Mac

  1. Get OpenDMG. It is a free app store for open source Mac apps. Download OpenDMG.
  2. Search for Utter and press Get. OpenDMG downloads version 1.2.3 from h3x-levi's own GitHub release.
  3. Open it. The download is checked against its SHA-256 fingerprint and its signature is checked on your Mac before it lands in Applications.

Updates to Utter show up in OpenDMG and install with one button. No account is needed.

Prefer to do it by hand? The developer's file is on GitHub: Utter-1.2.3.dmg (193 MB).

What is new in Utter 1.2.3

## 🎙️ Utter v1.2.3

On-device voice dictation for macOS, Windows, and Linux that never sends your audio to the cloud.

This release introduces the interactive picture-book Busytown website preview and design system, over 100 new companion moments with customizable outfits and finishes, and low-latency speech detection powered by Silero VAD.

### ✨ What's New

- **🌐 Busytown Picture-Book Website Preview**:
 - Interactive picture-book cutaway town showcasing how audio stays local.
 - Interactive robot player preview and integrated product video reel.
 - Complete design specification (`website/DESIGN.md`) covering typography, colors, and motion tokens.

- **🤖 Companion & Robot Choreography**:
 - Over 100 new moments for dictation, meetings, reminders, time-of-day, and play.
 - Customization options for companion finish, outfit/accessories, and quiet mode behavior.
 - Synchronized state machine and choreography across macOS, Windows, and Linux.

- **⚡ Fast, Quiet Speech Detection (Silero VAD)**:
 - Bundled Silero VAD model skips stretches of silence and background noise before inference.
 - Significant CPU savings and reduced latency during dictation pauses.

- **🔤 Phonetic Sound-Alike Correction**:
 - Enhanced dictionary corrections for sound-alike words (e.g., "cube or needs" -> "Kubernetes").

### 📦 Downloads & Binaries
…

More Audio and Music apps

All Audio and Music apps