The keyboard is the slowest part of your stack.

Voca (usevoca.net) is cloud voice dictation for Windows, macOS and Linux, built for people who drive AI agents. You speak at about 150 words a minute and type at about 60. One hotkey puts your voice into your editor, your terminal, your browser and Slack. See how many hours a year you spend typing.

The Voca dashboard: minutes used this month, team credits, time saved, the record and upload panel, and recent transcriptions.

Try for free

No account, no card, nothing to install. Speak a couple of sentances and see the results for yourself.

up to 60 secondschecking

One hotkey

Use everywhere.

Use everywhere, from Clauce Code and Codex, to Slack and Teams. Voca sits at the operating system so you can use it anywhere.

Cursor

CtrlShift;

you sayRefactor this to use the repository pattern and add a test for the empty case.

Claude Code

CtrlShift;

you sayRun the suite, then fix whatever fails without touching the migrations.

Your terminal

CtrlShift;

you sayRebase onto main, keep my version of the lockfile, then push.

Slack, ChatGPT, anything else

CtrlShift;

you sayThe same hotkey. Voca types into the focused field, so it works in software that has never heard of it.

Including Linux.

Every operating system included.

Use Voca everywhere, on Linux, Windows, MacOS or Android.

  • Linux

    X11 and Wayland Support. How it works on Linux

  • macOS

    Apple silicon and Intel.

  • Windows

    Windows 10 and 11.

  • Android

    Transcribe while you are on the move. Get it on Google Play

  • Any browser

    The same as the desktop app, just on the browser.

Speed

Fast transcription

Get a cleaned up transcript a couple seconds after you finish speaking.

0
seconds of audio transcribed
0
seconds saved against typing
0
seconds of audio kept
5hrs
free every month

Trust

Your audio does not stick around.

  • Deleted on return

    Audio lives in one temporary file while it is transcribed, then it is gone. Not backed up, not queued, not analysed.

  • Never used for training

    No model is trained on your voice or your text, by us or by anyone we send audio to.

  • Transcripts are opt-in

    History is off until you turn it on, and you can delete it whenever you like.

  • Billing metadata only

    We keep a duration, word count, a timestamp and a model name. Not the words. Read the detail.

How it works

Three simple steps.

  1. Press the hotkey

    Recording starts on your machine.

  2. Say what you want

    Audio goes up, text comes back, the audio is deleted.

  3. The text lands where your cursor is

    In the editor, the terminal, the browser field, wherever you were already typing.

Pricing

Pay for what you say.

Five hours free every month. No card to start.

Free
$0
  • 5 hours of transcription a month
  • Transcript history, opt-in
  • Transcription Cleanup
  • Dictionary (define your own words)
  • Browser and desktop app
  • No card required
Start free
Pay as you go
$0.10/hour
  • Top up from $2
  • Credits never expire
  • Works across your devices
Buy credits

Stop typing your prompts.

Not sure? Use the demo and decide for yourself, will you use us?

Try the demo