utter — local voice input

$ utter --listen

Voice input that lands
exactly where you work.

Hold a key, speak naturally, and Utter turns messy speech into polished text in any Mac app.

Download for macOS View source

macOS 26+ · Apple Silicon · MIT licensed

● 00:07Live ●
Transcript (Smart Format) Press play to see it transcribe.
Screen contextFake email thumbnail showing flight confirmation
  • Local ASR
  • LLM cleanup
  • Screen context
  • Any app

$ utter pipeline --verbose

From voice to final text in one key release.

  1. 01

    Hold fn

    fn

    A global hotkey starts capture without leaving the current app.

  2. 02

    Speak naturally

    ··|·|··

    Talk like you would to a person. Utter captures audio on your Mac.

  3. 03

    Local ASR

    </>

    Speech recognition can run locally on Apple Silicon.

  4. 04

    LLM cleanup

    ≡

    The selected model removes filler, fixes grammar, and formats for context.

  5. 05

    Insert anywhere

    ↵

    Accessibility insertion places the final text at the cursor.

Raw transcript

so basically we need to ship the update this week and test it before friday

⟶
Final text

We need to ship the update this week and test it before Friday.

$ utter features --list

The features are the pipeline.

  • ⌗

    Screen context OCR

    Reads nearby on-screen text to better understand names, project terms, and homophones.

  • ⌘

    Voice command mode

    Edit, summarize, translate, or reply without switching away from the current app.

  • ◇

    Local and remote models

    Run fully on-device, or choose a remote LLM provider when it suits your workflow.

  • ◎

    Bilingual UI

    Switch the interface between English and Simplified Chinese.

$ utter privacy --local

Private by default.
Local when you want it.

WhisperKit transcribes on-device. MLX-Swift-LM can refine text locally. Remote providers are optional.

Audio upload
0 required
Cloud calls
0 required
Remote providers
optional

$ utter stack

WhisperKit
Offline speech recognition
MLX-Swift-LM
On-device LLM inference
Swift 6
Native macOS app
ScreenCaptureKit
Screen context capture
AVAudioEngine
Low-latency mic capture
Vision
On-device OCR

$ utter install

Get Utter from GitHub Releases.

Download for macOSView source

Run once if macOS shows a Gatekeeper prompt.

$ utter faq

Questions, answered.

What is Utter?

Utter is a native macOS menu bar app that turns your voice into polished text in any app. Hold a hotkey, speak, and the result is typed directly at your cursor.

Does Utter work offline?

Yes. With WhisperKit or Apple Speech plus a local MLX model, Utter can transcribe and refine text entirely on-device. Remote LLM providers are optional.

Which macOS version and hardware does Utter require?

macOS 26 or later on Apple Silicon. Intel Macs are not supported.

Which apps does Utter work in?

Any macOS app that accepts text input. Utter uses the Accessibility API, with a clipboard-paste fallback.

Is Utter free and open source?

Yes. Utter is MIT-licensed and the full source is on GitHub.