Free Speech to Text

What Is Speech to Text?

Speech to text — also called voice to text, dictation or voice typing — converts spoken words into written text. A microphone captures your voice, a speech recognition engine identifies the words, and the result appears on screen as text you can copy, edit or save.

This tool uses the browser's native Web Speech API. That means no account, no API key and no audio ever passing through XToolBoost's servers. Everything runs through your own browser.

Browser support

Unlike most of our tools, speech recognition depends on the browser vendor rather than on XToolBoost. Here's the current state of the art:

How the recognition works

Interim vs final results

As you speak, the engine produces two kinds of results:

• Interim — a provisional guess that may change as more audio arrives. Shown in grey italic.
• Final — the engine's settled answer once it has enough context. Shown in solid black.

Interim results make the tool feel responsive, but they can shift mid-sentence. If you turn off "Show live text", only final results appear — less flicker, but more latency.

Continuous mode

When Continuous is on, the tool keeps listening until you press the mic button again. Browsers normally stop after a few seconds of silence to save resources — our tool automatically restarts recognition on `onend` when continuous mode is enabled, so you can dictate long passages without interruption.

Automatic punctuation

Modern engines add commas, periods and question marks as you speak, based on intonation and pauses. You can also spell punctuation out loud:

• Say "comma" → inserts ,
• Say "period" or "full stop" → inserts .
• Say "question mark" → inserts ?
• Say "exclamation point" → inserts !
• Say "new line" or "new paragraph" → inserts a break

Tips for better accuracy

• Use a good microphone. A headset mic beats a laptop mic beats a phone speakerphone.
• Reduce background noise. Fans, music and traffic confuse the engine.
• Speak at a steady pace. Rushing causes dropped words; too slow causes fragmentation.
• Pick the right language. Choosing "English (UK)" when you speak American English reduces accuracy.
• Pause between sentences. It gives the engine a natural boundary to finalise the previous sentence.
• Check the interim text. If it's wildly wrong, stop and restart — the engine has drifted.

Common use cases

Speech to text vs voice typing vs dictation

These three terms all describe the same underlying technology, but the context differs:

• Speech to text — the general term, covers any system that converts voice to written text.
• Voice typing — usually refers to typing into a document or text field using your voice instead of a keyboard.
• Dictation — traditionally meant speaking words for someone else to write down; today it means speaking into a microphone for automatic transcription.

They all do the same thing — the terminology is largely interchangeable.

Privacy and security

XToolBoost does not record, store or transmit your audio. The microphone stream is handled directly by your browser's speech recognition engine and never touches our servers.

However, it's important to be honest about how browser speech recognition works: Chrome, Edge and Safari may send your audio to their vendor's servers (Google, Microsoft and Apple respectively) to perform the transcription. This happens between your browser and the vendor — XToolBoost is not involved and cannot see or intercept your audio.

For highly confidential content, check your browser's privacy documentation, or use a locally-installed dictation tool.

Frequently Asked Questions

Related Text Tools