Voice to text, on a laptop or a phone
Record in your browser or upload a recording you already have, and read the words back in minutes, in 99 languages.
Voice to text means turning spoken words into written ones, whether you're dictating now or working from a recording made last week. voxnoto does both in the browser: press record, or upload a file, and the text comes back split by speaker.
How to turn your voice into text
Open voxnoto in your browser
On a laptop or a phone. There is nothing to install.
Record or upload
Press record and talk, or upload a recording you already have.
Read the words back
The text is ready in minutes. Copy it, export it or translate it.
Two ways to get your voice into text
Record now. Open voxnoto on a laptop or phone and press record. Useful for dictating a note, catching a conversation in the room, or capturing an idea before it goes.
Upload a recording. Any audio or video file up to 8 hours: MP3, M4A, WAV, MP4 and the rest. Useful for interviews, meetings, lectures and anything already sitting on your phone.
Why people convert voice to text
Speaking is faster than typing. Most people speak around 150 words a minute and type 40. For a first draft, the difference is an hour a day.
Text is searchable, audio isn't. A folder of recordings is a folder you never open. A folder of transcripts answers a question in seconds.
Hands and eyes are often busy. Driving, walking, cooking, working with your hands, the thought arrives anyway, and voice is the only way to catch it.
Reading is faster than listening. A twenty-minute recording is two minutes of reading. For anyone who has to go through several, that's the whole point.
What you get back
Punctuated text. Sentences and paragraphs, not one unbroken block.
Speakers, when there are several. Each turn attributed, names editable once and applied throughout.
A summary and a translation. The short version for long recordings, and a translation to read it in another language.
Speaking so the text comes out clean
- Keep a steady pace. Rushing costs more accuracy than a strong accent does.
- Hold the phone or headset 20-30 cm away.
- Pause between thoughts instead of saying "um", pauses become paragraph breaks.
- Say unusual names once, clearly, and fix them in the transcript afterwards.
- In a noisy place, get closer to the microphone rather than louder.
Accuracy, honestly
Clear speech in a quiet room, one person at a time, transcribes very well. Overlapping voices, heavy background noise and very compressed files are where any system loses accuracy, ours included.
That's why the transcript stays editable and the recording stays in your account. For anything you'll quote or publish, read the text against the audio before it goes out.
Questions and answers
How do I convert voice to text?
Record in your browser or upload a recording. The text is ready in a few minutes.
Do I need to install anything?
No. voxnoto works in your browser on a laptop or phone.
Does it work in my language?
99 languages are supported, detected automatically.
How big can my files be?
On Premium you can upload files up to 8 hours long and run several at once. The free plan gives you 10 minutes of transcription a day: longer files are transcribed up to that daily limit.
Can I edit the text afterwards?
Yes, including speaker names.
Is there a free plan?
Yes. The free plan gives you 10 minutes of transcription every day, no card needed.
Did not find your answer?Browse all questionsEmail support
