Chrome speech to text extension for Google Docs

What it does differently

• One key, not a panel.

  Hold the right Option / Alt key inside the document and talk. Release and the text lands at your cursor. Two quick taps switch to hands-free until you tap again. There is also a Voice button on the page and a global shortcut, Alt+Shift+D — and the key is changeable in the settings if it collides with something else.

• Silence does not end the session.

  Stop to think, read a source, check a message in another window. It is still listening when you come back. This is the single biggest difference from the built-in Google Docs feature, which gives up after 10–15 seconds of quiet and also stops when the tab loses focus.

• Punctuation and capitals happen by themselves.

  You do not say "comma" out loud. Full stops, commas, question marks and sentence case are inserted for you. Say "new line" or "new paragraph" when you want a break in a specific place.

• Two ways to get the text.

  When I stop (the default) collects the whole take and inserts it once you finish, with fillers, false starts and stutters removed — the words stay yours, the stumbles do not. While I speak puts each phrase in as it is recognised, if you prefer to watch the page fill.

• Your own Dictionary.

  Add colleagues, clients, products and jargon once, one per line. Every transcription uses that list, so a surname stops coming back wrong twenty times a day.

• 49 recognition languages, with Auto-detect.

  Pick a primary language for accuracy, or leave auto-detect on if you switch languages inside one sentence. English terms inside another language are recognised either way.

• It works in comments.

  Review threads and suggestions get dictation too, not only the document body.

• Esc throws away a take.

  Changed your mind mid-sentence? Press Escape and nothing is inserted. History keeps your recent transcriptions if you need to put one back.

What it does not do

Being straight about the edges, because this is where generic extensions overpromise:

• It is a Google Docs tool right now. Dictating into Gmail, Sheets or any text box on the web is planned, not shipped. If you need dictation everywhere today, a general-purpose extension will serve you better.
• Chrome only. No Safari, no Firefox, no mobile browser.
• Recognition runs in the cloud, so it needs a connection. Audio is sent to be transcribed and is not stored afterwards; the document's contents are never read.
• One recording runs up to five minutes, then the text is inserted and you start the next one. The widget warns you 30 seconds before.

FAQ

Is there a free speech to text extension for Chrome?

This one has a free tier: one minute with no account, ten more after signing in with Google. Beyond that it is paid, because every minute of recognition costs us money to run. Extensions that are free without limit usually run recognition in the browser, which is why their accuracy drops on accents, background noise and long sentences.

What is the difference between speech to text and text to speech?

Speech to text turns what you say into written words — dictation, which is what this extension does. Text to speech is the opposite: it reads written text aloud. The two get confused constantly in search results.

Does it work outside Google Docs?

Not yet. Dictating into any field on the web is on the roadmap; today the extension works inside Google Docs, including comments.

Which languages does it support?

49 for recognition, with auto-detect for mixed-language sentences. The extension's own interface is in English.

Can it learn names and technical terms?

Yes — that is what the Dictionary is for. Add the words once and every transcription uses them.

Is my audio stored?

No. It goes to our server to be transcribed and is discarded after the text comes back. The extension does not read your document's contents; it only inserts the words you dictate.

Why does it need microphone access?

To hear you. Chrome asks once, from the extension's settings page. If you have previously blocked the microphone for Google Docs, see the permission guide.