Yes, Windows voice typing is still officially supported in 2026. It is built into Windows 11, you start it with Win + H, Microsoft ships and updates it as part of the operating system, and it costs nothing. No download, no account, no subscription. If that was the only thing you came to check, you can stop here and press Win + H.

The next question is the one most people actually have. If the built-in one is free and supported, is it the best voice to text app for Windows? For occasional use, yes, and nothing else is worth installing. It stops being enough when you dictate for hours a day, when you need punctuation and formatting handled while you speak, when you switch languages mid-sentence, or when you need a record of the time you spent talking. This post puts Win + H next to the five third-party Windows apps worth knowing about in 2026, and says plainly where each one wins. That includes the places where Superscribe, which is the app I build, is the wrong choice.

The Windows options at a glance

App	Price	Text appears live	Runs offline	Time tracking
Windows voice typing (Win + H)	Free, in the OS	Yes	No, uses online recognition	No
Superscribe	Free tier, paid plan	Yes	No	Yes, automatic
Wispr Flow	Paid subscription	Yes	No	No
SuperWhisper	Free tier, paid Pro	No, pastes after you stop	Yes, with local models	No
Buzz	Free, open source	No, file and recording based	Yes	No
Dragon Professional	Paid, one-time licence	Yes	Yes	No

Prices move. Check each vendor’s own page before you buy.

Is Windows voice typing officially supported in 2026?

It is. Voice typing is a shipped Windows 11 feature, not a preview, and it arrives with the OS. Win + H opens it in any text field: browser inputs, native apps, Office, chat windows.

Most of the confusion behind this question comes from a different feature. Windows Speech Recognition, the older command-and-control system that let you drive the whole desktop by voice, has been wound down in favour of Voice Access. Voice typing is not that feature and was not affected. If you read somewhere that Microsoft discontinued Windows dictation, that is what the writer was thinking of.

You get live text straight into the focused field, an automatic punctuation toggle, and language selection through your keyboard language in Settings. Zero setup, no per-seat cost.

What you do not get:

• Custom vocabulary, so client and product names come out however the model guesses
• Any cleanup pass, so filler words and false starts land exactly as you said them
• Different behaviour per app. A Slack message and a client email get the same raw output
• Two languages in one sentence. Voice typing follows one active input language at a time
• An offline mode. Recognition runs in the cloud, so no connection means no dictation
• Any record of what you dictated or how long it took

That is a fair deal for free. For a few sentences a day, Win + H is the correct answer and the rest of this post is unnecessary spending.

Everything below sells you one of three things on top of that: better accuracy, a cleanup pass so speech comes out as writing, or something useful left behind. On plain accuracy the gap is smaller than the marketing suggests. On the other two it is large.

SuperWhisper as a Windows dictation app in 2026

SuperWhisper started on Mac and iOS and now has a Windows app. The workflow is hold a shortcut, speak, release, and the transcript is pasted into whatever field you had focused.

Two things it does better than most of this list. First, local models. Transcription can run on your own machine, so dictation keeps working on a plane and audio never leaves the laptop. Almost nothing else here with a real dictation workflow offers that. Second, its modes, which are saved prompts that reshape the transcript into an email, a bug report, a set of notes, whatever you set up once and reuse.

The trade-off on Windows is the paste-after model. Text arrives when you stop, not while you speak, so you find out afterwards whether it heard “Kubernetes” or “cube or netties”. Across a long session that means more re-reading. There is also no time tracking.

If you already run SuperWhisper on a Mac, the Windows app is the obvious pick for continuity. Longer side-by-side in Superscribe vs SuperWhisper. The short version: SuperWhisper wins on offline and on prompt customisation.

Superscribe on Windows

Superscribe streams text live, character by character, into the field you have focused. Cursor in the compose box, hold Shift + F9, and the words appear while you talk. F9 pastes after you stop instead. Esc cancels. It works in browser fields, native Windows apps, your CRM, Notion, Slack.

What is different from everything else on this list is automatic time tracking. Every dictation session is logged by project and duration without you starting a timer. Twenty five minutes across a proposal, two emails and a Slack thread gets captured as a matter of course. For anyone who bills hourly, that is the whole reason to install it. Mechanics in Voice to Text With Time Tracking for Freelancers.

Where it loses: no offline mode. Transcription runs in the cloud, so no connection means no dictation, and if your rules say audio never leaves the machine, use Buzz or SuperWhisper’s local models instead. There is also no custom vocabulary training in the Dragon sense.

Wispr Flow on Windows

Wispr Flow also streams live into the focused field, so the core feel is close to Superscribe. Its differentiator is context awareness: it reads which app you are in and adjusts tone and formatting, so Gmail gets email register and Slack gets something looser.

It also has the strongest compliance story here, including SOC 2 Type II and HIPAA eligibility on the enterprise tier. If your employer’s security review is the actual gate, Wispr Flow clears it and most of this list does not. I am not going to talk around that.

Where it stops: no time tracking, and it is subscription-only. Longer comparison in Superscribe vs Wispr Flow.

Buzz and Dragon, the offline options

Buzz is free, open source, and runs Whisper models locally on Windows, Mac and Linux. No audio leaves your machine, ever. The catch is that it is a transcription tool, not a dictation tool: you record or feed it a file and get a transcript. No shortcut-to-type workflow, nothing appearing in your active app, and slow without a GPU. For sensitive material it is still the safest thing here.

Dragon Professional holds one real niche: years of trained custom vocabulary. A lawyer or clinician who has fed Dragon their firm’s terminology since 2022 gets accuracy on those terms that cloud tools do not match, plus full offline operation. The price and setup overhead make sense only if that describes you.

So which voice to text app for Windows should you use?

• Win + H if you dictate a few times a week. Free, supported, zero setup. Start here.
• Superscribe if you dictate daily and bill for your time.
• Wispr Flow if a security review is the gate, or you want per-app tone matching.
• SuperWhisper if you want local models, prompt customisation, or continuity with a Mac.
• Buzz if audio must not leave the machine and a transcript is all you need.
• Dragon if you have years of vocabulary training you cannot rebuild elsewhere.

What Superscribe does not do on Windows

Worth saying clearly, because people arrive expecting it. The Windows and Mac desktop apps do two things: live dictation into any input field, and automatic time tracking. They do not capture calls or meetings. No Windows meeting recorder, no bot that joins calls.

Call and meeting capture lives in the free Superscribe Phone app, and it is iPhone only. You keep your own number and answer calls as usual, and afterwards you get a transcript, a title, a summary and one-tap suggested actions like calling back, adding a calendar event or saving an unknown caller. It does not write or send your emails. You get the summary and the action items and write the follow-up from those. If that is the part you need, start with call notes or meeting minutes, or see how transcribing phone calls on iPhone works without a meeting bot.

If you are on Windows hoping for call recording on the same machine, none of the tools in this roundup do that.

---

Try Superscribe for Windows at superscribe.io Live text into any field, and your time logged without a timer.

Related reading

• Best Voice to Text Apps for Mac in 2026
• Dictation App for Mac That Types Where You Work
• Live Dictation Into Any Input Field
• Why Real-Time Dictation Beats Record-Then-Transcribe

Frequently asked questions

Is Windows voice typing officially supported in 2026? Yes. Voice typing (Win + H) is a standard Windows 11 feature that ships and updates with the OS. The feature Microsoft wound down was Windows Speech Recognition, the older whole-desktop voice control system, which has been replaced by Voice Access. Voice typing was not affected.

What is the best free voice to text app for Windows? Win + H, for almost everyone. Live, works in any field, nothing to install. Buzz is the free pick if you need offline transcription of recordings rather than dictation into apps.

Does SuperWhisper work on Windows in 2026? Yes. It pastes the transcript after you stop speaking rather than streaming live, and it supports local models so audio can stay on your machine. It does not track time.

Does Superscribe work on Windows? Yes. Shift + F9 streams text live into the focused field, F9 pastes after you stop, Esc cancels. It needs a connection, and it does not record calls or meetings on Windows.

Can I dictate in two languages in one sentence? Not with Win + H. It follows one active input language at a time, so mixing mid-sentence produces nonsense in whichever language is not selected. Cloud tools handle mixed speech better, none perfectly.

Which Windows dictation app tracks billable time automatically? Superscribe. Sessions are logged by project and duration with no timer to start or forget. Nothing else in this roundup does it, because it is not what they are built for.