Use Voice Input to Talk to Claude Code Instead of Typing
Some prompts are easier to say than type. Describing a bug you just saw, explaining how a feature should work, or walking through your reasoning — these flow naturally in speech but feel clunky to type out. Voice input lets you just talk.
Press Option+V (Mac) or Alt+V (Windows/Linux) to start voice input
Speak your prompt naturally, and Claude Code transcribes it into text. You can review and edit the transcription before sending, or just let it go if it looks right.
This is especially useful for:
- Complex descriptions — "the dropdown component in the sidebar is rendering behind the modal overlay when you click the settings gear, and it only happens when the sidebar is in collapsed mode"
- Thinking out loud — "I think the issue is that we're fetching the user's preferences before the auth middleware runs, so the preferences query happens with no user context and returns defaults"
- Long instructions — multi-step prompts that would take a minute to type take ten seconds to say
- Accessibility — hands-free coding for developers who prefer or need voice interaction
Voice works well with Claude Code's conversational style. You don't need to speak in formal commands — just talk the way you'd explain the problem to a colleague:
"Hey, the checkout page is showing the wrong total when there's
a discount code applied. I think the discount is being calculated
before tax instead of after. Can you check the order calculation
logic and fix it?"
Claude processes the transcribed text the same way it handles typed prompts — reading your code, making edits, running commands. The input method changes but the capabilities stay the same.
You can mix voice and typing freely in the same session — voice for explanations and descriptions, typing for short commands and file paths.
When your fingers can't keep up with your brain, switch to voice — the fastest prompt is the one you just say out loud.
via Claude Code
Log in to leave a comment.
When Claude is heading down the wrong path — editing the wrong file, writing code you don't want, or giving a long explanation you don't need — press Escape to stop it immediately. You keep everything it did up to that point and can redirect with a new prompt.
Closed a session and realized you weren't done? Pass --continue (or -c) when launching Claude Code to pick up exactly where you left off — same context, same files, same conversation history — without re-explaining what you were working on.
The -p (print) flag runs Claude Code as a one-shot command — ask a question, get an answer, and return to your shell. Perfect for quick lookups, scripting, piping output, and integrating Claude into shell workflows without starting a full interactive session.