What is Speech to Text?

Speech to Text converts your spoken words into written text using your device's built-in speech recognition. Use it to dictate notes, draft messages, or capture ideas hands-free. Everything is processed privately on your device.

Recognition uses the Web Speech API that ships with Chrome, Edge, and Safari, so accuracy depends on your device and microphone, not on a remote server. Thirteen recognition languages are supported including English variants, Mandarin, Arabic, and Hindi. Punctuation isn't inserted automatically — say 'comma' or 'period' to add it.

How to use

1. Step 1 — Click the microphone button and grant microphone access when prompted.
2. Step 2 — Speak clearly into your microphone. Your words appear as text in real time.
3. Step 3 — Click stop when finished, then copy or download your transcribed text.

When to use

• Drafting messages, emails, or notes hands-free during a commute or while cooking.
• Capturing rough meeting notes when typing would be too slow or distracting.
• Reducing wrist strain by speaking instead of typing during long writing sessions.

Result

You're in a meeting and need quick notes. Press the mic button, speak naturally about the discussion points, and get a clean text transcript you can paste into your note-taking app.

FAQ

Does my voice get sent to a server?

Recognition routes the audio through your device's built-in speech service — Google on Chrome, Apple on Safari, on-device on a few platforms. Once the text appears it stays right here on your device. We also autosave it locally so a reload doesn't lose your work; the Clear button wipes the saved copy.

Why doesn't the mic button do anything?

Either the browser doesn't expose the Web Speech API (older Firefox, some private windows) or microphone permission was denied. Check the address bar's site-settings icon. Chrome and Edge on desktop or Android tend to work best.

It keeps stopping after a few seconds, what's wrong?

Most browsers auto-stop after a silence gap to save resources. Talking continuously usually keeps it going. If it cuts off mid-sentence, click the mic again — the existing transcript is preserved and new speech appends.

Can I add commas, periods, and line breaks?

Yes, say the punctuation aloud: 'comma', 'period', 'question mark', 'new line'. Recognition engines vary in how reliably they catch this, so a quick pass through the transcript afterwards is normal.

It transcribed an accent wrong, can I fix the dictionary?

The recognition model is the browser's, not editable here. Picking the closest regional variant (en-GB vs en-US, es-MX vs es-ES) helps most accent issues. For names or jargon, edit the transcript directly before copying it out.

Related Tools