Wavetype

Say it instead of typing it.

You have a long reply to write. You already know what you want to say, and saying it out loud would be quicker than typing it.

So you try dictation. You press the key and start to talk, and the screen stays empty. Is it listening? Did it understand the name? You continue and hope. When you stop, a block of text arrives with every “um” still in it and the commas missing, and you spend the next few minutes fixing it. Sometimes a part of what you said is missing. The next day you type again.

Wavetype is built so that you can see it working. Press Win+Ctrl in any app on Windows and talk. A small card next to your cursor shows every word as you say it, so you know it heard you, and you continue. Press Win+Ctrl again and the text is pasted where you were typing, with the filler words removed and the punctuation added. If you listed things, you get a list.

A long dictation arrives complete. Every recording is saved first, so if something fails, Win+Ctrl+R recovers your words. And when a paragraph you already wrote needs changes, select it and say what to change: make it shorter.

It also changed how I work with AI. A typed prompt is short, because typing is slow. When I talk, I explain the whole thing: what I want, why I want it, what I already tried. The AI gets the full picture, and the answer is better. I almost never type a prompt now.

I dictate most of what I write. This is the tool I wanted to talk into.

LorenzoMilan · September 2026