Cue · Use Cases · Voice coding with Cue

Voice coding with Cue

Hold a hotkey. Describe what you want. Cue writes code directly into Cursor, VS Code, or any editor on your Mac or Windows. No translation layer between thought and shipped feature.

Typing was a latency layer between your thoughts and the code. Voice removes the layer.

3 real scenarios

Mid-feature
You say: "Refactor this React component to use useReducer for the form state"
Cue does: Cue reads your selected code, rewrites it, replaces in place in Cursor.
Debug at 2am
You say: "Translate this stack trace into human language and tell me where it broke"
Cue does: Cue reads Sentry/console error + related source file, gives you a clear diagnosis.
New file
You say: "Create a new utility file that exports a debounce function and a throttle function with TypeScript types"
Cue does: Cue creates the file, fills it, opens it in your editor.

How Cue does it

Try Cue free

Download for Mac or Windows. Free tier: 20 agent credits/day, unlimited dictation.

Get Cue →

FAQ

Which code editors does voice coding work in?

Cue works in any macOS or Windows app that supports text input — Cursor, VS Code, JetBrains IDEs, Vim/Neovim (in any terminal), Xcode, Android Studio. The 5-layer context capture works across all of them.

How does Cue handle the keyboard interruption when I voice code?

Cue uses Fn long-press as the hotkey, separate from typing. You can type and voice-code in the same session without interrupting each other.

Can I voice code in 20+ languages?

Yes. Cue supports voice input in 20+ languages (English, Chinese, Japanese, Korean, Spanish, French, German, etc). The voice command can be in any of these languages — Cue translates intent into code in the syntax you want.

How is voice coding faster than typing?

For complex refactors, voice + agent execution is 2-5x faster than typing. For simple one-line edits, typing wins. Most Cue users alternate fluidly between both.

They hear what you said.
Cue sees what you're doing.
And does the thing, in any app.