Speech to Text
Speak into your microphone and see it transcribed live. WebToolGrid never receives your audio, but your browser's speech recognition typically sends it to the browser vendor's service (Google, in Chrome).
How it works
This tool uses your browser's built-in speech recognition engine (the Web Speech API) to transcribe your microphone input live. Your browser sends the audio to its own speech-recognition service to produce the transcript — this site never sees or stores the audio or the resulting text.
Browser support
Speech recognition support varies significantly by browser. Chrome and Edge have solid support; Safari has partial support; Firefox does not currently support the Web Speech Recognition API at all. You'll also need to grant microphone permission when prompted.
What the controls do
Pick one of ten languages before you start: English (US), English (UK), English (India), Spanish, French, German, Hindi, Portuguese (Brazil), Japanese or Chinese (Mandarin). The menu is locked while the microphone is on. Recognition runs in continuous mode with interim results, so words show up in the box as you speak and are replaced by the final version when the browser settles on it.
Editing and saving the transcript
The box is an ordinary text area, so you can correct it by hand. Each finished phrase is added with a space after it, and any capital letters or punctuation come from the browser's recognizer, not from this page. Clear appears once there is text, Copy copies the trimmed transcript, and Download .txt opens a download page offering it as transcript.txt, with a button to go back to the transcript. Words still shown as interim are not part of what is copied or saved until they are finalized.
What it cannot do
Only the microphone is supported, not an audio or video file. There are no timestamps and no speaker labels. A short silence is ignored without an error, while a blocked microphone shows a message explaining how to allow it for the site. When the browser ends the session, the button returns to Start listening and the text so far is kept.
Frequently asked questions
Is my audio sent anywhere?
Not to WebToolGrid — this site never receives or stores your audio or the transcript. However, your browser's built-in speech recognition typically sends the audio to the browser vendor's own service (for example, Google's in Chrome) to produce the transcript, so it is not processed only on your device.
Why doesn't this work in my browser?
Speech recognition support varies by browser. Chrome and Edge support it well; Safari has partial support; Firefox does not currently support the Web Speech Recognition API. Try a recent version of Chrome or Edge if it isn't working.
Why did it stop listening on its own?
Browsers automatically stop the microphone after a period of silence or a maximum session length as a privacy safeguard. Press Start listening again to continue — your existing transcript is preserved.
Can I transcribe a recording or an audio file?
No. The tool listens to your microphone only and has no file upload, so it cannot transcribe a recording that is already saved on your device.
Why does the text change while I am still talking?
Words first appear as interim guesses and are replaced when the browser finalizes a phrase. Only finalized phrases are saved when you copy or download.