Speech to Text
Dictate in your browser and watch your words appear as text
Speech to text turns live microphone audio into editable text while you speak. On this page you can record in your browser for up to 2 minutes, with nothing to install, then review and copy the result. For an existing recording, upload the file instead: the first 30 minutes are transcribed free.
No account? The first 30 minutes of any file are free. Create a free account for premium on your first file.
- Up to 99%
- Accuracy on clear audio
- 99
- Languages
- Anytime
- Cancel, no contract
- Free
- 30 min, no signup
Why use Speech to Text?
Words As You Speak
Your microphone audio is transcribed live, so the text builds up on screen while you talk.
Speaker Labels
When more than one person talks, the live transcript separates the voices, so a quick interview or brainstorm stays readable.
Dictation and Voice Typing
Draft notes, emails and outlines by voice. Speak naturally and punctuation is added for you.
Edit, Copy, Export
Save the recording to open it in the editor, fix any word, and copy the text. Downloads follow your plan: TXT on a free account, every format on paid plans.
How it works
Click Record
Allow microphone access and start speaking. There is nothing to install.
Speak Naturally
Talk at your normal pace. The text appears while you speak, for up to 2 minutes per recording.
Review and Save
Stop, read the transcript, then save it to edit, copy or export it.
Speech to text online: a practical guide
Live speech to text and voice typing
Record your voice and see the words appear as you speak.
Speech to text is useful whenever typing would slow you down: a note between meetings, the first draft of an email, an idea on a walk. The live engine processes your microphone audio in real time, so you can keep talking without switching between voice and keyboard.
- Recording from your browser microphone
- Text that appears while you speak
- Automatic punctuation
- Speaker separation when more than one person talks
- Editing before you copy or export
To remember: live recordings on this page stop at 2 minutes. For anything longer, upload a file.
How do I do speech to text?
- Open this page and click Start Recording. Allow microphone access when your browser asks.
- Speak at your normal pace. Punctuation is added for you.
- Stop the recording and read the transcript.
- Save it to open the editor, correct any word, then copy it or download it on your plan.
Live recording or upload: which one?
| You have | Use | Free limit |
|---|---|---|
| A thought to dictate | Live recording on this page | 2 minutes per recording |
| A lecture, meeting or interview file | Upload the file | First 30 minutes without an account |
| A long recording you transcribe often | A free account | Up to 30 files a day, up to 30 minutes each |
Speech to text for students
For lecture recordings, a transcript is easier to search and revisit than audio. Record the lecture, upload the file, and use the transcript as a study draft. Check key terms and names against the recording before you rely on them.
Accuracy, and where it drops
The premium engine reaches up to 99% accuracy on clear speech. Background noise, several people talking at once, fast speech and unusual names lower that, so check names and numbers before you rely on a transcript.
My take: start here for quick voice typing or a short dictated note. For a lecture, interview or meeting longer than 2 minutes, upload the file and plan to review the transcript before you use it.
Speech to Text vs typing it out by hand
Why upload-and-go beats manual transcription every time.
| Capability | CATT | Manual transcription |
|---|---|---|
| Time to first draft | Minutes per file | 8 to 10 hours per audio hour |
| Accuracy | Up to 99% on clear audio, premium engine | Depends on the typist |
| Speaker labels | On the premium engine | Tagged by hand |
| Languages | 99 supported | Whatever you speak |
| Exports | TXT, DOCX, PDF, SRT, VTT | Whatever you build by hand |
| Cost | Free for the first 30 minutes | Your time, every time |
Weighing transcription services instead? Read the full CATT vs Otter.ai comparison
Simple pricing
Free for files up to 30 minutes, no card needed. Paid plans start at $14.99.
Free
- Up to 30 files a day, each up to 30 minutes (your first file on the premium engine, up to an hour)
- Speaker labels on your first file
- Copy any transcript or download it as TXT
Cancel anytime. Try the free plan before you upgrade. See all plans
Frequently asked questions
No. Speech to Text works in your browser. Allow microphone access, click record and start speaking.
Live recordings on this page are capped at 2 minutes, enough to dictate a note. For longer material, record with the Online Voice Recorder or upload a file: the first 30 minutes of an uploaded file are transcribed free.
Live speech runs on our premium engine, which reaches up to 99% accuracy on clear speech. A quiet room and one person speaking at a time give the best result. Check names and technical terms before you share the text.
Yes. Upload MP3, WAV, M4A and other common audio or video files. Without an account the first 30 minutes are free; a free account transcribes up to 30 files a day, up to 30 minutes each, and its first file runs on the premium engine for up to an hour.
Yes. The live transcript separates speakers when more than one person talks. For uploaded files, speaker labels come with the premium engine: your first file on a free account, the Weekly Pass or a paid plan.
Yes. Arabic, Hindi and Polish are in the spoken-language list, along with English, Spanish, French, German, Japanese and more than 20 others, and auto-detect covers 99 languages. Pick your language before recording for the most reliable result.
Yes, in a mobile browser that allows microphone access. Open this page, tap record and grant the permission when your browser asks. There is no app to install.
Upload it on the homepage transcriber, which is built for existing files. This page is built for live speech from your microphone. Both use the same editor once the transcript is ready.