Hands-free typing software: which kind do you actually need?

People land on this page for very different reasons. Wrists that ache after a day of typing. A condition that rules out the mouse and keyboard entirely. Or just a preference for thinking out loud. Those are different problems with different right answers, and most roundups blur them into one ranked list. I build a dictation app for a living, so I will be specific about where my kind of tool fits — and where it does not.

What is hands-free typing software?

The label covers any software that puts words on screen without keypresses: you place the cursor in a text field, speak, and text appears there. The same mechanic gets sold under several names — automatic dictation, audio typing software, voice typing, speech-to-text. Same thing, different marketing.

The distinction hiding under the label is typing versus controlling, and it decides everything else in this guide. Dictation software writes for you and does nothing else. Voice control software operates the computer itself — it clicks, scrolls, opens apps, switches windows. If you cannot use a mouse, only the second category solves your problem; no dictation app will, mine included. If your hands work but typing wears them down, dictation is the lighter tool and usually the better one. Plenty of people end up running both. Confusing the two categories is how you buy the wrong software, so I keep them separate throughout.

What are the three types of hands-free typing software?

1. Built-in OS dictation. Windows has voice typing on Win + H and, on Windows 11, Voice Access; macOS has Dictation and Voice Control. Free, already installed, and genuinely fine for the occasional message in English — if that describes your use, stop reading and press Win + H. Daily use is where the cracks show: accuracy plateaus on names and jargon, an accent makes it worse, you switch languages by hand, and everything you say lands verbatim — ums, false starts, the sentence you abandoned halfway through.

2. AI dictation apps. Lazytype, Wispr Flow, and Superwhisper run Whisper-class speech models — a different tier of accuracy from the built-in engines, and the gap widens if you have an accent (I am Dutch; for me it is not subtle). They type into whatever app has focus, from Word to Slack to a code editor, punctuate from how you actually speak, and can strip the filler words before the text lands. Lazytype's whole interaction: hold the left Ctrl key, talk, release. Clean punctuated text appears at your cursor about a second later, in whichever of 100+ auto-detected languages you spoke.

3. Specialist accessibility suites. Dragon Professional, Windows Voice Access, and macOS Voice Control are their own category because typing is only part of what they do: they click, scroll, correct text by voice, and run the whole machine. If you cannot use a mouse or keyboard at all, this is your category, full stop. A dictation app alone will not get you there — it has no concept of a button.

What is the best hands-free typing software?

There is no single best, only a best for your reason. The short version:

Picking comes down to matching the category to why you are going hands-free in the first place.

• You type all day and it hurts, or you just hate it. An AI dictation app. You keep the mouse for navigation and move the writing — the bulk of your keystrokes — to your voice.
• You occasionally want to speak a note or a message. The built-in options. They are free and already installed; start there and upgrade only if the accuracy or the workflow starts to grate.
• You cannot (or should not) use your hands at the computer at all. An accessibility suite, possibly combined with a dictation app: Voice Access or Voice Control to click and navigate, a Whisper-based app for faster and more accurate writing.

Can I control my whole computer by voice?

Yes — but not with a dictation app, and I would rather say that plainly here than have you find out after paying for the wrong tool. Dictation apps, Lazytype included, do one thing: convert speech to text at your cursor. They do not click buttons, do not open programs, do not scroll. For actual control you need:

• Voice Access on Windows 11 — free, built in, runs on-device. You can say "open Edge", "click Send", or show a numbered grid overlay and speak the number of anything on screen. Microsoft's Voice Access guide on support.microsoft.com documents the full command set.
• Voice Control on macOS — Apple's equivalent, also free and on-device, with grid and item-number overlays. See Apple's Voice Control guide on support.apple.com.
• Dragon Professional — the long-standing commercial suite, still the deepest option for voice-driven correction, custom commands, and specialised vocabularies, at $699 one-time.

People who need everything hands-free usually run a combination: the OS accessibility tool drives the computer, a Whisper-based app does the writing. The split works because the two jobs are opposites. Navigation is a small, fixed vocabulary that must be caught perfectly every time ("click Send"), which is exactly what on-device command engines are built for. Prose is an open vocabulary spoken naturally, which is where Whisper-class models pull far ahead. If you only need to write hands-free, skip the control layer entirely — it is a genuine learning curve, and you will not use it.

What helps with RSI or a hand injury?

If typing is painful, the useful measure is keystrokes avoided. A 50-word email is roughly 300 keypresses typed, or one key held for fifteen seconds spoken — move your writing to your voice and most of the day's mechanical load comes off your hands. To be clear about the limits: dictation reduces load, it is not treatment. If your wrists hurt, a doctor or physiotherapist belongs in the plan alongside any software. That said, a few mechanics matter more than any feature list.

Hold-to-talk beats always-listening for most RSI cases. With hold-to-talk, the microphone is only live while you physically hold a key; release it and the app is deaf. Always-listening sounds more hands-free, but it means the software decides when you are talking to it, and it gets that wrong — it happily transcribes the colleague at your desk, the phone call you answered, the sentence you were only trying out. You then spend keystrokes deleting text you never meant to write, which defeats the point. One lightly held key — left Ctrl under a pinky, in Lazytype's case — is a cheap price for deciding yourself. And if even one held key is too much for your hands, that is not a failure of technique; it is the signal you belong in the accessibility-suite category, where always-listening plus voice-driven correction is the intended design.

Dictate in short bursts, not monologues. A sentence or two, release, glance, continue. Short clips transcribe faster, and when the engine mishears you, you lose five words instead of a paragraph. Lazytype's realtime preview bar shows your words as you speak, so you catch a wrong word mid-sentence instead of after it has been typed.

Correct by voice too. Fixing dictation errors on the keyboard hands the keystrokes straight back. Say "scratch that" and the last dictation is deleted; re-speak the sentence and you have corrected it without touching a key. One honest limit: for single-character surgery — one wrong letter in a name — the keyboard is still faster, which is why "far fewer keystrokes" is a more realistic goal than "zero".

How do I format and correct text by voice?

With built-in dictation you either speak punctuation aloud ("period", "comma") or switch on auto-punctuation. Whisper-based apps punctuate from your pauses and intonation, so there is nothing to announce. On top of that, Lazytype understands a small set of spoken commands: "new line" and "new paragraph" for structure, "scratch that" to undo the last dictation. Voice macros go a step further — each hotkey can carry its own behaviour, so one key types clean text, another translates while you speak, and a third rewrites what you just said ("make this formal"). Context-aware tone handles the rest without being asked: casual in Slack, formal in Word, technical terms left alone in a code editor.

Does hands-free typing software work offline?

Partly, and it depends on the tool. Win + H voice typing needs the internet — it runs on Microsoft's cloud speech service. Voice Access and Apple Dictation run on-device and keep working with no connection. Among AI dictation apps, offline support is the exception: Lazytype ships a local whisper.cpp engine — set the tray menu to Engine → Local and audio is processed entirely on your machine, never sent anywhere. That matters beyond flaky wifi. If you work under confidentiality rules — legal, medical, client NDAs — on-device processing is often the difference between "may use dictation" and "may not". The trade-off is speed: local transcription is slower than the cloud engine on modest hardware, though accuracy stays at Whisper level.

Which one should you actually install today?

If you have never dictated anything: press Win + H (or enable Apple Dictation) and speak one email today. It costs nothing and tells you whether your voice and your writing get along — some people find they compose fine out loud, others freeze, and no paid app changes that. If you need the whole computer hands-free, set up Voice Access or Voice Control first; that is the foundation, and it is free. And if the writing itself is the problem — volume, pain, or plain dislike — a Whisper-based app is the upgrade you will notice every day. Lazytype (v1.8.2) costs €25 one-time with your own free Groq key, or €5/month for Pro with hosted transcription, runs on Windows 10/11 and macOS 12+, and the 14-day trial needs no card. Hold a key, speak, and your hands get the rest of the sentence off.

Written by Bas Niese

Founder of Lazytype. Bas is a Dutch developer who dictates most of what he types — in Dutch and English — and has shipped every Lazytype release since v1.0. More about Bas and Lazytype →