Live Dictation (mic)
Convert speech to text online free with voice typing and dictation in your browser. Multiple languages, copy, and download. Runs in your browser — live data comes from a public API.
Speech to Text uses your browser’s built-in Web Speech API to transcribe your voice in real time. Everything runs locally on your device — no audio is uploaded or stored. Keep the tab in focus and your microphone enabled while dictating.
This tool turns your voice into text as you speak. Choose a language, press Start dictation and talk; the words appear in the transcript box, with interim guesses shown live while the browser is still working out a phrase. When you stop, you can edit the transcript by hand, copy it, or download it as transcript.txt.
It is a front end for the Web Speech API built into your browser, and it does not do the recognition itself. There are 14 language options: English (US and UK), Arabic, Spanish, French, German, Italian, Portuguese (Brazil), Russian, Hindi, Japanese, Korean, Chinese (Simplified) and Turkish. Continuous mode keeps listening and restarts the recognizer when the browser stops it, and you can switch off the live interim results. It listens to your microphone only; it does not transcribe an audio or video file.
Why this runs on your device
This tool is not on-device, and your voice goes off your device while you dictate, so know that before you say anything private. The recognition is done by your browser's Web Speech API, and in Chrome and Edge that means your microphone audio is sent to the browser vendor's servers (Google for Chrome, Microsoft for Edge) to be turned into text. The page itself does not record or store your audio, and the transcript is only held in the page until you close it, but the audio does leave your device while you speak, under your browser vendor's privacy terms. If your audio is confidential, use the Audio Transcriber tool instead: it runs a Whisper model on your own device, and the file you give it is not uploaded.
- Open the page in Chrome or Edge and choose your language.
- Press Start dictation and allow microphone access when asked.
- Speak clearly. Watch the transcript fill in, then press Stop.
- Edit the text if needed, then copy it or download it as a .txt file.
- It needs a browser that supports speech recognition, in practice Chrome or Edge. Firefox support is limited, and where it is missing the page shows a notice and the Start button is disabled.
- Recognition needs an internet connection in Chrome and Edge because the work is done on the vendor's servers. It does not work offline there.
- Accuracy, supported accents, and any punctuation depend entirely on the browser's service. The tool does not add punctuation or capital letters itself, and it cannot be trained on names or jargon.
- It only listens to a live microphone. You cannot give it a recording, and the transcript has no timestamps or speaker labels. For files, use the Audio Transcriber.
- Keep the tab open and in focus with microphone permission allowed. If you deny the permission or the browser reports no speech, the tool shows an error and you have to start again.