How to Dictate Into Ghostty and Modern Terminals

Ghostty, Warp, WezTerm and other modern terminals are fast and beautiful, but they still expect you to type every character. If your hands are tired or you just think faster out loud, you can dictate straight into the prompt instead. Here is how to set it up on a Mac.

Can you dictate into Ghostty?

Yes. Despite how specialized a terminal feels, Ghostty is just another macOS window with an editable text surface. That is the key insight behind speech recognition for the terminal: if your cursor can blink there, a system-wide dictation tool can insert text there. The same is true for Warp, WezTerm, Kitty, Alacritty and iTerm2. You do not need a special plugin for each one, you need a dictation app that works everywhere your cursor goes.

This is where built-in options fall short. macOS ships with its own dictation feature, and it does technically type into a terminal, but it has no awareness of developer vocabulary and does little to clean up the raw stream of words. A purpose-built tool that runs entirely on your Mac handles both the privacy and the polish that terminal work demands.

The dictation workflow, end to end

Before the setup steps, it helps to picture what actually happens when you speak. Your microphone audio never leaves the machine: it flows through a local model, gets cleaned up on-device, and lands in the terminal as text.

Set up dictation for your terminal in five steps

The setup is the same whether you use Ghostty, Warp or WezTerm, because you are configuring the dictation app, not the terminal. Here is the full path.

Commands versus prose: use dictation where it wins

Honesty matters here. Dictation is not a magic replacement for typing a one-liner packed with pipes, flags and backslashes. Symbols are slow to say out loud and easy to mishear. Where voice genuinely shines is the prose that surrounds your commands: commit messages, code comments, pull request descriptions, and long prompts to an AI coding agent running inside the terminal. Most people speak around three to four times faster than they type, so those paragraphs land far quicker by voice.

A practical rhythm is to dictate the words and type the symbols. Speak your commit message or your prompt, then fix any exact flags by hand. The table below shows where each approach tends to fit.

Because the AI cleanup runs on-device, it removes your filler words, fixes punctuation and shapes grammar without any of your text touching a server. That is a real difference from cloud dictation tools, and it is the reason terminal users who handle tokens, keys and client code tend to prefer a local engine. If you are curious whether voice keeps working with the network off, we cover it in our guide on whether the best Mac dictation software can run fully offline.

Why on-device matters for terminal work

Terminals are where secrets live. You paste tokens, echo environment variables, and read log lines that should never leave your laptop. A cloud dictation tool records your microphone and sends the audio to a server for transcription, which is exactly the wrong trade-off for this context. With an on-device model, your audio and the resulting text stay on the Mac, full stop.

On-device also means dictation survives a flaky connection. On a train, on a plane, or behind a locked-down corporate network, the local model still turns your voice into text. This same portability is why voice typing has become a favorite for people managing repetitive strain: it is one reason many developers ask what people with RSI use to keep working. And if you switch between the terminal and your mail client all day, the same setup lets you dictate emails on your Mac without changing tools. You can compare tiers on the pricing page.