---
title: 'Voice Dictation for Cursor on macOS'
excerpt: 'Dictate prompts into Cursor chat and composer: hold a hotkey, speak, and Keebye types it — on-device speech-to-text, offline, private.'
date: '2026-07-14'
author: 'Teodor Deleanu'
tags: ['cursor', 'dictation', 'ai-coding']
keywords: ['voice input for cursor', 'dictation for cursor', 'cursor ai voice prompts']
faq:
  - question: "Does Keebye work in Cursor's chat panel and composer?"
    answer: 'Yes. Keebye inserts text into whatever field has focus, so the chat panel, the composer, and inline edit prompts all work the same way — no Cursor-specific setup needed.'
  - question: 'Does Cursor or its model provider ever see my audio?'
    answer: 'No — the audio is processed on-device and never leaves your Mac. Cursor sees nothing but the final text, indistinguishable from text you typed.'
  - question: 'Does it work offline?'
    answer: 'Yes, after the one-time model download. Dictation itself needs no network connection — though Cursor naturally needs one to reach its models.'
  - question: 'How do I start and stop dictating?'
    answer: 'Press and hold Right Command, speak your prompt, and let go — the text is inserted on release. A single tap toggles instead, for longer dictations, and Esc cancels one in progress. The hotkey can be switched to Fn or Right Option.'
  - question: 'Can it learn my project-specific terms?'
    answer: 'Yes. Add repo names, framework terms, and the jargon you say constantly to the custom dictionary so they transcribe correctly the first time.'
---

## Why dictate prompts to Cursor

Cursor turns coding into a [prompt-heavy workflow](/dictation-for/vibe-coders). Between the chat panel, the composer, and inline edits, most of what you produce in a session is not code — it's paragraphs describing the change you want, the files involved, and the constraints to respect. That's a typing workload, and multi-sentence instructions are quicker — and far less tiring — to say than to type.

Keebye doesn't integrate with Cursor specifically, and doesn't need to: it types into whatever field has focus. Chat, composer, an inline edit prompt, the rename dialog — if your cursor is in it, you can dictate into it.

## How it works

Click into Cursor's chat or composer input, hold Right ⌘, and describe the change. Letting go of the key drops the transcribed text at your cursor — send it as-is or touch it up first.

Transcription happens entirely on your machine: Parakeet, the English-tuned default model, needs a single download and then works with no connection at all. There's no telemetry, and your voice never leaves the Mac — Cursor only ever gets the finished text.

The same hotkey works in every other app on your Mac, so [the terminal running your agent](/dictation-for/terminal), your browser, and Slack are covered by the same muscle memory.

## Setup in two minutes

Getting started takes two permissions and one choice: allow Accessibility and microphone access when Keebye asks, then decide which key you'll hold — Right ⌘ by default, or Fn / Right ⌥ if you prefer. Flip on launch-at-login and it's there whenever Cursor is.

Then spend a minute on the custom dictionary. Add your repo names, package names, and the framework terms you say constantly, so they land in the prompt correctly instead of needing a fix-up pass.

## Limits, honestly

Words don't stream in as you speak — transcription is batch, triggered on key-release, so the text shows up only once you let go of the hotkey.

Keebye has no Cursor-specific integration — it doesn't read Cursor's state or trigger its actions. It types text where your cursor is; you stay in charge of sending it.

Keebye is macOS only.