Live transcription
See interim results while you speak, with final text kept safely as you go — no destructive overwrites.
Works in your browser — no install
Free speech to text converter
Convert your voice into accurate, editable text instantly, in more than 100 languages. Edit, copy, or download the result as TXT, Word, or PDF.
Recording only starts when you choose to start it. Nothing is recorded in the background.
Speech to text workspace
How it works
- Choose your language. Search or pick from the dropdown; the tool remembers your last choice.
- Press start. Your browser will ask for microphone permission the first time — it's only requested after you press the button.
- Speak naturally. Text appears as you talk; you can pause and keep going.
- Edit, copy or download. Fix anything by hand, then copy the text or save it as TXT, Word, or PDF.
Built for real use, not just a demo
Right-to-left support
Arabic, Persian and Hebrew are shown with correct direction and punctuation handling.
Editable by design
The output is a real text editor: undo, redo, copy, cut and paste all work as expected.
Download anywhere
Export to plain text, a Word-compatible document, or PDF, with Unicode-safe encoding.
Light and dark themes
Matches your system preference automatically, and remembers your choice if you change it.
No account required
Use the tool immediately. There is nothing to sign up for in this version.
More tools are on the way
Speech to Text is the first tool on Speech to Text. Audio and video transcription, subtitle generation, translation and document tools are planned — this page will link to each one as it ships.
Language coverage
Speech to Text's interface lists languages from India, Asia, Europe, the Americas, Africa and Oceania. Being listed in the interface is different from being supported by your browser's speech engine — support depends on your browser and operating system, and is detected automatically when you pick a language.
Resources & guides
- What is speech to text, and how accurate is it?
- How speech recognition works in the browser
- Voice typing on Android and iPhone
- Speech to text for students, writers and businesses
- Accessibility and voice typing
- Troubleshooting: microphone and permission issues
This section is a placeholder for a future content hub; each guide will get its own page once written.
How your speech is processed
In this version, transcription is performed by your browser's built-in speech recognition engine (for example, Chrome uses Google's speech service). Audio may be sent to your browser vendor's servers to produce text; Speech to Text itself does not run its own transcription servers in this version and does not store your audio.
Transcribed text stays in your browser tab unless you choose to copy or download it. Your theme and language preference are saved locally on your device using localStorage, and can be cleared through your browser settings.
Exact data handling depends on your browser vendor's own privacy policy. Read our full Privacy Policy for details, including what will change if cloud or AI-based transcription is added in a future version.
Frequently asked questions
Is this free?
Yes. The current version has no paid tier and does not require an account.
Does it work offline?
The page and editor work offline once loaded, but live speech recognition needs a network connection because it relies on your browser's speech service.
Why isn't my language available?
The interface lists many languages, but actual recognition depends on what your browser and operating system support. If a language isn't recognized, the tool will tell you rather than fail silently.
Is my speech stored?
Speech to Text does not store your audio. Processing is handled by your browser's speech engine — see "How your speech is processed" above for details.
How accurate is it?
Accuracy depends on your microphone, accent, background noise, and browser. We don't publish an accuracy percentage because it varies by these factors.