How to Voice Type in Windows Terminal
The short answer: click into Windows Terminal, press a dictation hotkey, say the command or prompt, and press the hotkey again. The words are pasted at the prompt and nothing runs until you press Enter. Windows has a built-in option (Win+H) that uses Microsoft’s online speech service. HyperVoice is a free app that does the same thing with one hotkey, transcribes on your own PC, and can press Enter for you when you say “send it”.
Option 1: Windows’ built-in voice typing (Win+H)
Windows 10 and 11 come with voice typing. To use it in Windows Terminal:
- Click into the terminal so the cursor is at the prompt.
- Press Win+H. The voice typing bar opens and starts listening.
- Speak. The words appear as you talk.
- Press Win+H again (or click the microphone) to stop.
It costs nothing and there is nothing to install, so it is worth trying first. Three things to know before you rely on it for terminal work:
- It needs the internet. Microsoft’s own support page says voice typing “uses online speech recognition, which is powered by Azure Speech services”, so your audio goes to Microsoft to be transcribed.
- Punctuation is either spoken or guessed. You say “comma” and “period”, or turn on automatic punctuation in the voice typing settings.
- Some terminal programs don’t present a text box Windows recognises. At a plain PowerShell prompt it generally works, but full-screen programs running inside the terminal can answer Win+H with “select a text box and try again”. Claude Code users have reported exactly that.
Option 2: A dictation hotkey (HyperVoice)
HyperVoice is a dictation hotkey for Windows: press it in any app, talk, and the text lands where your cursor is. In Windows Terminal:
- Install it from the Microsoft Store. It’s signed by Microsoft, so there is no install warning. Setup ends with a test dictation.
- Click into the terminal tab. PowerShell, Command Prompt, a WSL shell, an SSH session: anything running in Windows Terminal.
- Press Ctrl+Shift+Space and talk. Say the command, or the whole prompt. You can change the hotkey in Settings.
- Press Ctrl+Shift+Space again. The words are pasted at the prompt. Read them, then press Enter.
Speech-to-text runs on your PC (Whisper or NVIDIA Parakeet models), so your audio isn’t sent anywhere to be transcribed. Punctuation comes from how you speak; you don’t say “comma”. The optional clean-up styles, which rewrite what you said as a tidy message or bullet points, are off by default. When you turn one on, the text (never the audio) is sent to an AI provider.
It’s free at 500 words a day with a free account, no card.
What to dictate in a terminal, and what to type
Be realistic about shell syntax. “git space commit space dash m” is slower to say than to type, and flags, paths and pipes don’t survive speech well. The keyboard stays the right tool for short commands.
Voice wins when what you’re entering reads like a sentence:
- Prompts to AI tools that run in the terminal, such as Claude Code. These are long, plain-English and full of context (what to change, what to leave alone, why), which is exactly what people cut short when they have to type it. There’s more on this in How to Use Voice Dictation in Claude Code and Voice Typing in VS Code, Cursor, and Claude Code.
- Commit messages. Type
git commit -m ", dictate the message, type the closing quote. - Anything wordy you’d otherwise paste in: an issue title, a note to a script, a message to a chat tool on the command line.
A useful split: type the command, speak the words.
Let it press Enter: “send it”
For prompts you’d submit straight away, HyperVoice can press Enter for you. Turn it on in Settings → Dictation → Voice commands, then end a dictation with the words “send it”. Only the very end of a dictation counts, so saying “send it” mid-sentence doesn’t submit anything.
It suits AI prompts well. For shell commands that change or delete things, leave it off and read what landed at the prompt before you press Enter yourself.
If the text doesn’t show up
- Check focus. The terminal tab has to be the active window when you press the hotkey. HyperVoice won’t paste into a different window from the one you were in.
- Several lines at once. If a dictation comes out as more than one line (a bullet-point clean-up style, for example), Windows Terminal may ask you to confirm the paste. That’s the terminal’s own safety check.
- A program that still loses the text. Open Settings → Dictation → Delivery overrides, add the program’s .exe name (
WindowsTerminal.exefor Windows Terminal) and choose Type. The words are then typed in as keystrokes instead of pasted.
Which one to use
If you dictate into the terminal now and then and you’re always online, Win+H is free and already on your PC. If you talk to AI tools in the terminal all day, want punctuation without saying it, or would rather the audio stayed on your PC, HyperVoice is built for that. Try it free from the Microsoft Store.
Frequently asked questions
Does Windows Terminal have voice typing built in?
Windows Terminal itself has no dictation feature. Windows' system-wide voice typing (press Win+H) can type at an ordinary PowerShell or Command Prompt prompt, but it uses Microsoft's online speech service, so it needs an internet connection. A dictation app such as HyperVoice adds its own hotkey that works in any terminal tab and transcribes on your PC.
Can I voice type in PowerShell and Command Prompt?
Yes. On Windows 11 both normally open inside Windows Terminal, so the same steps work: click into the tab, press the dictation hotkey, speak, press it again. The words are pasted at the prompt.
Will dictating into the terminal run a command by accident?
No. Dictated text is pasted at the prompt and waits there until you press Enter. The only exception is HyperVoice's optional 'send it' voice command, which is off by default: when you turn it on and end a dictation with the words 'send it', HyperVoice presses Enter for you.
Does it work in WSL and SSH sessions?
Yes. The text is pasted into whatever the Windows Terminal tab is running, including a WSL shell or a remote server over SSH. The speech recognition still happens on your Windows PC; only the finished text reaches the session.
Can I talk to Claude Code in Windows Terminal?
Yes. Put your cursor in the Claude Code prompt, press the hotkey and explain the task the way you would to a colleague. Prompts to AI tools are where voice typing in a terminal pays off most, because they are long and written in plain sentences.
What if the dictated text doesn't appear in the terminal?
First check the terminal tab had focus when you pressed the hotkey: HyperVoice will not paste into a different window from the one you were in. If a particular terminal program still loses the text, open HyperVoice Settings, Dictation, Delivery overrides, add the program's .exe name (for Windows Terminal it is WindowsTerminal.exe) and choose Type, so the words are typed in as keystrokes instead of pasted.
Related posts
Best Voice-to-Text Apps for Windows in 2026
A detailed comparison of the top Windows dictation tools in 2026 — HyperVoice, Wispr Flow, Voicy, Dragon NaturallySpeaking, WhisperTyping, and Windows built-in dictation.
Filing Bug Reports and Tickets with Your Voice
Stop context-switching to write up tickets. HyperVoice's Ticket / Issue mode turns your spoken description into a structured bug report or task — ready to paste into Jira, Linear, or GitHub.
HyperVoice for Developers: Voice Dictation in VS Code, Cursor, and the Terminal
How developers use HyperVoice to dictate code comments, commit messages, documentation, and chat with AI coding agents — all without leaving their editor.