Speech to text costs nothing to start. Every Mac, iPhone, Windows PC, Android phone and Chromebook already has dictation built in, Google Docs has it in the browser, and a growing group of free apps run open-source speech models on your own computer with no word limit at all. What differs is the catch: some need an internet connection, some work in only one app, and some free plans run out after a few thousand words a week.
The short answer: the best free speech to text is the dictation already on your device: Apple Dictation on a Mac or iPhone, voice typing (Win + H) on Windows, and Gboard voice typing on Android. For a free app that goes further, Handy and Spokenly run offline with no word limits, and Buzz transcribes recordings at no cost.
We checked every free plan below on each developer's own site, app listing or help pages on November 9, 2026. Free tiers change often, so read them as "at the time of writing." We make Aqua Voice, which has a free plan of its own. It gets its own section after the picks, with its limits stated plainly.
What "free" means in speech to text
Free tools fall into three groups, and knowing which one you are looking at saves a lot of disappointment:
- Built in. Free with the device, with no download and no account. Apple, Microsoft and Google all ship one.
- Free and open source. Apps such as Handy and Buzz that run a speech model (usually OpenAI's open-source Whisper or NVIDIA's Parakeet) on your own computer. No limits and no subscription, but your hardware sets the speed.
- Free tiers. Commercial apps with a free plan that caps your words or holds back features. Good for a trial, and sometimes enough for light use.
Two more questions sort the field. Does it work offline, which matters for privacy and patchy connections? And does it type into any app, or only into one?
The best free speech to text tools at a glance
Tool | Platforms | What is free | The catch |
|---|---|---|---|
Apple Dictation | Mac, iPhone, iPad | All of it | The newest on-device model needs recent hardware |
Windows voice typing | Windows 10 and 11 | All of it | Needs an internet connection |
Gboard voice typing | Android | All of it | Not every language supports talk-to-text |
Chromebook Dictation | ChromeOS | All of it | You must speak your device's language |
Google Docs voice typing | Chrome, Edge, Safari | All of it | Works only inside Docs and Slides |
Handy | Windows, Mac, Linux | All of it, open source | No phone app |
Spokenly | Mac, Windows, Linux, iPhone | Unlimited local dictation | Cloud models are paid |
Superwhisper | Mac, Windows, iPhone, Android | Unlimited Whisper models | Offline models want an Apple silicon Mac |
Google AI Edge Eloquent | iPhone, iPad | All of it | A standalone app, not a keyboard |
Wispr Flow | Mac, Windows, iPhone, Android | A weekly word allowance | 2,000 words a week on desktop |
Buzz | Windows, Mac, Linux | All of it, open source | Transcribes files, does not type into apps |
MacWhisper | Mac | File transcription in 100 languages | Pro features cost €64 once |
Aqua Voice | Mac, Windows, iPhone, Android | 1,000 words | Then a paid plan |
Free speech to text already on your device
Start here. These cost nothing, need no account, and for short messages and quick notes they may be all you need.
Mac: Apple Dictation
- Open the Apple menu, choose System Settings, then Keyboard, and turn on Dictation.
- Click in any text field, then press the Microphone key (if your keyboard has one), use the Dictation keyboard shortcut, or choose Edit, then Start Dictation.
- Speak. In supported languages Dictation adds commas, periods and question marks for you; say "new line" or "new paragraph" to break the text.
Apple says there is no timeout: dictation stops when you press Escape, the Microphone key or the shortcut, or after 30 seconds with no speech. On a Mac with Apple silicon you can keep typing while you dictate. With Apple Intelligence in macOS 27, an on-device model improves spelling, punctuation and capitalization, in English on Macs with M3 or later and at least 12GB of memory. Our guide to dictation on a Mac covers shortcuts and commands in full.
iPhone: keyboard dictation
- Open Settings, tap General, then Keyboard, and turn on Enable Dictation.
- Tap in a text field, then tap the microphone on the keyboard and start talking.
- To stop, tap the microphone again, say "Stop dictation," or stay silent for 30 seconds.
Apple says dictation is processed on your iPhone in many languages with no internet connection required, and the keyboard stays open so you can switch between typing and talking. Our guide to voice to text on iPhone has the commands and fixes.
Windows: voice typing (Win + H)
- Click in a text box.
- Press Windows logo key + H, or tap the microphone on the touch keyboard.
- Wait for "Listening...", then speak. Say "Stop listening" when you are done.
Microsoft's support page says voice typing uses online speech recognition powered by Azure Speech services, so it needs an internet connection and a working microphone. The gear icon in the voice typing menu turns on automatic punctuation and the voice typing launcher. Windows 11 also includes Voice access, which controls the whole PC by voice. Our guide to voice to text on Windows explains when to use each.
Android: Gboard voice typing
- Open any app you can type in, such as Gmail or Keep, and tap a text field.
- At the top of the Gboard keyboard, tap Microphone.
- When you see "Speak now," say what you want written. Say "period," "comma," "new line" or "new paragraph" for punctuation.
To fix a word, touch and hold it, tap the microphone and say the replacement. Google notes that talk-to-text does not work in every language, and that some Pixel phones offer Advanced voice typing instead. Our guide to talk to text on Android walks through settings and fixes. On a Galaxy phone, Samsung Keyboard has its own voice input, covered in our Samsung voice to text guide.
Chromebook: Dictation
- Select the time at the bottom right (or press Alt + Shift + s), then choose Settings and Accessibility.
- Under "Keyboard and text input," turn on Dictation.
- Click where you want to type, then select Speak, or press Search + d (or Launcher + d).
Google warns that speech will not work if the language you speak does not match the device language. Once dictation is on, commands like "Select all," "Copy" and "Cut" edit your text by voice.
In the browser: Google Docs voice typing
Open a document in the latest Chrome, Edge or Safari, click Tools, then Voice typing, and click the microphone. Say "Stop listening" to pause. Google says your browser controls the speech-to-text service and sends the text to Docs. It only types into Google Docs (and Slides speaker notes, under Tools, then Dictate speaker notes), so it will not help in email or chat. Our Google Docs voice typing guide covers commands and fixes, and our roundup of speech to text Chrome extensions covers dictation in every tab.
The best free speech to text apps
When the built-in tool is not enough, these apps are free to use for as long as you like, or offer a free plan worth knowing about.
1. Handy: best free app for Windows, Mac and Linux
Handy is the cleanest answer to "free speech to text on a computer." It is free and open source, and its developer says it works completely offline: your voice stays on your computer. Press a keyboard shortcut, speak, release, and Handy pastes what you said into the text field you are typing in. Push-to-talk is on by default, and you can switch to press once to start and again to stop.
It runs Whisper models (small, medium, turbo and large) with GPU acceleration when your computer has it, plus NVIDIA's Parakeet V3, which is tuned for ordinary processors and detects the language automatically.
- Free plan: everything. Sponsors and donations fund the project.
- Platforms: Windows, Mac and Linux. On Windows you can also install it with
winget install cjpais.Handy. - The catch: it does one job, transcribe and paste, and there is no phone app. See Aqua Voice vs Handy.
2. Spokenly: free local dictation on the most platforms
Spokenly's pricing page puts it bluntly: on-device Whisper and Parakeet models are free forever, with no word caps, no trial clock and no credit card, and you can download it without an account. It also lets you connect your own API keys from cloud providers such as OpenAI, Deepgram and Groq at no extra charge; you pay the provider directly.
- Free plan: unlimited offline dictation in 100+ languages.
- Platforms: Mac, Windows, Linux and iPhone.
- The catch: Spokenly's own cloud models and AI cleanup are in Pro, at $12/mo billed annually ($144 a year). A $29 one-time Local+ purchase adds extras like settings sync. See Aqua Voice vs Spokenly.
3. Superwhisper: the most polished free tier
Superwhisper's free plan includes voice to text in any app, support for 100+ languages and unlimited use of Whisper models, which run offline. New users also get 3,000 words of Pro features to try, after which, the company says, the free tier stays available for good.
- Free plan: dictation in any app with local Whisper models, with no word limit.
- Platforms: Mac, Windows, iPhone and Android.
- The catch: Superwhisper's FAQ says offline models only run really well on Apple silicon Macs and suggests cloud models, a Pro feature, on Intel Macs. Pro is $8.49/mo month to month at the time of writing. See Aqua Voice vs Superwhisper.
4. Google AI Edge Eloquent: best free app on iPhone
Google's experimental Eloquent app is free with no usage limits, according to its App Store listing, and it works offline. It goes a step past plain dictation: it removes "ums," "uhs" and mid-sentence self-corrections, so what you get reads like written text. It supports 15 dictation languages and needs iOS 16 or later.
- Free plan: everything.
- Platforms: iPhone and iPad.
- The catch: it is a standalone app, not a keyboard (the listing says a keyboard is "coming soon"), so you dictate in Eloquent and then paste the text where it needs to go. Our iPhone speech to text roundup compares it with the paid options.
5. Wispr Flow: a free plan with a weekly allowance
Wispr Flow's free plan includes speech to text in any app, 100+ languages, and a dictionary that learns your names and jargon, on Mac, Windows, iPhone and Android. The limit is the word count. Wispr's help center says the free plan gives Mac and Windows a 2,000-word weekly allowance, after which Flow slows down, with a hard cap of 5,000 words a week; iPhone and Android share a separate weekly pool shown in the app. The count resets every Sunday.
- Free plan: dictation everywhere, up to the weekly allowance.
- Platforms: Mac, Windows, iPhone and Android.
- The catch: heavy users will hit the cap. Pro is $15 per user per month, or $12 billed annually. See Aqua Voice vs Wispr Flow and our Wispr Flow alternatives.
Free speech to text for recordings
Dictation types as you talk. If you already have audio (an interview, a lecture, a voice note), you want a transcriber instead.
6. Buzz: free transcription on any computer
Buzz transcribes and translates audio and video files, and YouTube links, offline on your own computer. It can also transcribe live from your microphone, identify speakers, watch a folder for new files, and export TXT, SRT and VTT. It is free under the MIT license.
- Platforms: Windows, Linux and Apple silicon Macs (the last version for Intel Macs is 1.4.5).
- The catch: it is for files and live captions, not for typing into other apps, and the developer notes the Windows installer is not signed, so expect a warning when you install it.
7. MacWhisper: free file transcription on a Mac
MacWhisper's free version records from your Mac's microphone, transcribes mp3, wav, m4a, ogg, opus, mov and mp4 files in 100 languages, and exports subtitles (SRT, VTT) or plain text. MacWhisper Pro is a €64 one-time license for heavier work. Our Whisper apps roundup compares it with Buzz in more detail.
For developers
OpenAI's Whisper model is open source under the MIT license, so you can run it yourself for free. Hosted speech APIs charge by the hour; our own Avalon API gives you the first $1.00 of usage free with no card, then costs $0.39 per audio hour on its OpenAI-compatible transcription routes. Our speech to text API roundup compares the hosted options.
Where free speech to text falls short
Free tools have improved a lot, and for texts, notes and quick replies they are often enough. The gaps show up in longer, more demanding work:
- Names and jargon. General-purpose models slip on product names, technical terms and unusual spellings, which is exactly what work writing is full of.
- One app or one device. Google Docs voice typing only types into Docs, and Gboard only lives on your phone. Each device's built-in tool has its own shortcut, settings and language list, so nothing carries over when you switch.
- Your hardware sets the ceiling. Free local apps are only as fast and accurate as the model your computer can run. On an older laptop you trade accuracy for speed.
- Fixing mistakes. Voice commands in free tools cover the basics, like punctuation, selecting and deleting. Rewording a sentence usually means going back to the keyboard, which eats into the time you saved by talking.
Aqua Voice's free plan, honestly
Aqua Voice is built for the work where those gaps hurt. It runs on Avalon, our own proprietary speech model, in the cloud, and types into any app.
The Free plan gives you 1,000 words with no credit card, and it includes Avalon 1.5, our current model. That is enough for a real test on your own writing, not a monthly allowance: once you have used it, Pro is $10/mo billed annually or $12/mo month to month. See pricing.
Why people pay after the trial:
- Accuracy on hard words. On AISpeak-10, a challenging subset of AISpeak, our own benchmark of real-world clips where speakers use AI jargon, Avalon got the key term right in 97.4% of clips, against 65.1% for Whisper Large v3. That is self-reported and dates from Avalon's launch in August 2025, so weigh it as one data point. For checkable context, Avalon debuted as the top proprietary model on the public OpenASR leaderboard in October 2025 (sixth overall at the time), and in an August 2025 9to5Mac test the writer read the same passage into built-in Mac dictation and Aqua at once and counted 17 errors versus 1. That was one passage and one speaker, before macOS 27's new dictation model, so it is an anecdote rather than a benchmark.
- Speed. Hold Fn on a Mac and talk. By our own figures, text starts appearing in under 50ms and final text lands in about 450ms, and you can dictate at up to 230 words per minute.
- It learns your words. A personal dictionary teaches it names and terms, and Edit Mode lets you fix text by saying what to change.
- Every device. Aqua runs on Mac, Windows and iPhone in 49 languages, and on Android, where it is a floating button above the keyboard you already use, in dozens of languages (launch post, Google Play). One subscription covers all four.
Aqua has a SOC 2 Type II audit report and is ISO 27001 certified.
The honest tradeoffs: Aqua processes audio in the cloud and needs a connection, so if audio must never leave your computer, pick Handy, Spokenly or Superwhisper's local models. The free plan is 1,000 words, not unlimited like Handy or Eloquent. And it is built for live dictation, not for transcribing a folder of recordings, which Buzz and MacWhisper do well.
New to dictation? Our setup guide covers the first week, and Best Dictation Software compares the paid field.
How to pick in one minute
- You want free and have never tried dictation: use the one built into your device first.
- Free, offline and unlimited on a computer: Handy, or Spokenly if you also want an iPhone app.
- Free and polished on a Mac: Superwhisper's free tier with local Whisper models.
- Free on iPhone with cleaner text: Google AI Edge Eloquent.
- Free transcription of recordings: Buzz on any computer, MacWhisper on a Mac.
- Accuracy on names and jargon on every device, with a free trial: try Aqua Voice.
FAQ
Is there a completely free speech to text app?
Yes. The dictation built into Mac, iPhone, Windows, Android and Chromebook is free, and Handy (Windows, Mac and Linux), Spokenly's local models, Google AI Edge Eloquent (iPhone) and Buzz (file transcription) are free with no word limits at the time of writing. Free tiers from commercial apps, such as Wispr Flow's, cap how many words you can dictate each week.
Is Google speech to text free?
Google's consumer tools are free: Gboard voice typing on Android, Google Docs voice typing in Chrome, Edge and Safari, Dictation on Chromebooks, and the Eloquent app on iPhone. Google's Cloud Speech-to-Text service for developers is a separate, usage-priced API; our speech to text API roundup covers it.
Which free speech to text works offline?
Handy, Spokenly's local models, Superwhisper's Whisper models, Google AI Edge Eloquent, Buzz and MacWhisper all run on your device. Apple says iPhone dictation is processed on the device in many languages with no internet required. Windows voice typing (Win + H) needs an internet connection.
Is Aqua Voice free?
Aqua Voice has a Free plan with 1,000 words and no credit card required, so you can test it on your own writing. After that, Pro is $10/mo billed annually or $12/mo month to month, and one subscription covers Mac, Windows, iPhone and Android.