Speech-to-Text
Speak naturally and convert your voice to text instantly. Supports 17 languages with real-time transcription.
Click the microphone button and start speaking...
Clear Speech
Speak clearly and at a moderate pace for best results.
Privacy
All processing happens in your browser. No data is sent to servers.
17 Languages
Supports major languages with automatic speech detection.
About the Speech to Text
This speech to text tool transcribes what you say using your browser's recognition engine, across more than seventeen languages. There is no account, no per-minute charge and no audio file to upload.
Speech to text through the browser is worth using because dictation is roughly three times faster than typing for most people. Notes, first drafts and long messages all come out quicker spoken than typed, even accounting for the corrections.
What makes it work is punctuation. Recognition engines will not infer it, so saying "comma" and "full stop" out loud is the difference between usable text and one enormous run-on sentence. Speaking at a normal pace in a quiet room also helps considerably more than speaking slowly and carefully.
Accuracy depends on your microphone, background noise and accent, and it improves noticeably with a headset over a laptop microphone. Note that browser speech recognition may send audio to the browser vendor's service for processing depending on your browser and platform — so it is not the tool for confidential dictation.
How to use the Speech to Text
- Choose your language. Pick from the languages your browser supports.
- Allow microphone access. Grant permission when the browser asks.
- Dictate. Speak at a normal pace, saying punctuation aloud.
- Correct and copy. Fix any misheard words, then take the text.
Speech to Text features
- Recognition across 17+ languages
- Live transcription as you speak
- No account or per-minute charge
- No audio file to upload
- Punctuation by voice command
- Uses the browser speech recognition engine
Frequently asked questions
How do I get punctuation?
Say it aloud — "comma", "full stop", "new paragraph". Recognition engines will not infer punctuation.
How accurate is it?
It depends on microphone, background noise and accent. A headset improves accuracy noticeably over a laptop microphone.
Is my audio private?
Not necessarily. Browser speech recognition may send audio to the browser vendor for processing, so avoid confidential dictation.
Which languages are supported?
More than seventeen, depending on what your browser provides.
Is there a time limit?
No per-minute charge or quota, though browsers may stop a session after a period of silence.