Which browsers are supported? Why not Firefox?

The tool relies on the browser’s built-in Web Speech API for recognition. Chrome and Edge (desktop and Android) and Safari (macOS and iOS 14.5+) support it; Firefox does not offer the API yet, so the page shows an “unsupported” notice there. Some Chromium-based browsers also cannot reach a recognition service.

Is my audio uploaded?

This site never uploads or stores your audio, and the transcript stays in your browser. The recognition itself, however, is performed by your browser vendor’s service: Chrome sends audio to Google’s servers, Edge uses Microsoft’s online speech service, and Safari uses Apple’s speech recognition (on-device on some hardware). Keep that in mind for sensitive content.

How accurate is it, and what does auto punctuation do?

Accuracy depends on your browser, microphone, background noise and how clearly you speak — pausing briefly between sentences helps. Browsers often return Chinese without punctuation and English without a final period, so auto punctuation capitalizes the first letter and adds a period or a question mark (for sentences starting with words like what, how, is or can). It is rule-based, so review important text before exporting.

Can I use the SRT file for video subtitles?

Yes. Each segment starts when its first word was heard and ends when the sentence was finalized, with paused time excluded, so you can transcribe while a video plays and get roughly aligned captions. Every subtitle shows for at least one second and never overlaps the next; fine-tune the timing in your video editor if you need frame accuracy.

🗣️

🗣️Speech to Text Online - Free Live Transcription, TXT & SRT

Click Start, allow microphone access and start talking — your words appear as text in real time. What you are saying right now shows in grey italics, and each finished sentence becomes its own segment stamped with the time since you started. Pick from more than twenty recognition languages, including English (US, UK, Australia, India), Mandarin, Taiwanese Mandarin, Cantonese, Japanese, Korean, French, German, Spanish and more. Pause and resume whenever you like; paused time is not counted. Every segment can be edited or deleted in place, and the transcript is saved in your browser automatically. When you’re done, copy the full text or download it as plain TXT, TXT with timestamps, or an SRT subtitle file built from the segment times. Recognition uses the browser’s built-in Web Speech API — Chrome, Edge or Safari recommended.

Free · No sign-up · Runs in your browserLast updated: 2026-10-26

Loading tool...

How to use Speech to Text

1Choose the recognition language, click Start and allow microphone access
2Speak — text appears live, one segment per sentence with a timestamp; pause and resume at any time
3Fix any mistakes, then copy the text or download TXT or SRT

FAQ

Which browsers are supported? Why not Firefox?: The tool relies on the browser’s built-in Web Speech API for recognition. Chrome and Edge (desktop and Android) and Safari (macOS and iOS 14.5+) support it; Firefox does not offer the API yet, so the page shows an “unsupported” notice there. Some Chromium-based browsers also cannot reach a recognition service.
Is my audio uploaded?: This site never uploads or stores your audio, and the transcript stays in your browser. The recognition itself, however, is performed by your browser vendor’s service: Chrome sends audio to Google’s servers, Edge uses Microsoft’s online speech service, and Safari uses Apple’s speech recognition (on-device on some hardware). Keep that in mind for sensitive content.
How accurate is it, and what does auto punctuation do?: Accuracy depends on your browser, microphone, background noise and how clearly you speak — pausing briefly between sentences helps. Browsers often return Chinese without punctuation and English without a final period, so auto punctuation capitalizes the first letter and adds a period or a question mark (for sentences starting with words like what, how, is or can). It is rule-based, so review important text before exporting.
Can I use the SRT file for video subtitles?: Yes. Each segment starts when its first word was heard and ends when the sentence was finalized, with paused time excluded, so you can transcribe while a video plays and get roughly aligned captions. Every subtitle shows for at least one second and never overlaps the next; fine-tune the timing in your video editor if you need frame accuracy.