Turn a voice note into tasks, without sending your voice anywhere
Talk through everything on your plate and get back a tidy list of tasks, each with an owner, a due date and a priority. The transcription happens on your device, so the recording of your voice never leaves the browser.
The UnaTask team, Building UnaTask. Jul 8, 2026. 4 min read.
You are walking to the next meeting and your brain picks that exact moment to file five things at once. Ping the designer about the logo, the invoice is due Friday, book the venue, ask Sofia about the budget, and do not forget the thing from this morning you have already half forgotten. No free hand, no good moment to type, and you know how this ends.
By the time you sit down, two of the five are gone. The other three survive as one enormous note that reads like a run on sentence, which you will decode later, badly, and split into real tasks that never quite match what you meant.
So do the natural thing instead. Tap the mic and talk, in whatever order it comes, and let the ramble become a list.
One breath in, separate tasks out
You say it all in one go, and it comes back as a tidy set of separate tasks, not one blob. Each one gets the details you mentioned: who it is for, when it is due, how urgent it is, which project it belongs to, and even the one that has to wait on another. You glance over the rows, fix anything that landed wrong, add one it missed, and create them all in a single click.
Your voice stays on your device
The speech to text happens right in your browser. The recording of your voice is never uploaded anywhere, and the model that does the listening loads once and is kept by your browser after that, so every capture from then on starts the moment you tap the mic. Only the plain text of what you said goes on to become tasks. The convenience of talking, with the privacy of a notebook.
Capture at the speed of talking. Keep it as private as a notebook.
The same walk, nothing dropped
Run the walk again. You talk for twenty seconds, drop the phone back in your pocket, and by the time you sit down there is a clean list waiting: five tasks, each assigned, dated and sorted, ready to act on. Nothing decoded, nothing lost between the thought and the list. Open voice capture from your dashboard, or from inside a project, and just say what needs doing.