Skip to content
wisper

How it works

From thought to text in three steps.

Speak naturally, let Wisper refine it, and get on with your day. Here's what happens at each step, and what happens to your data along the way.

The three steps

  1. Step 1: Start speaking

    Activate Wisper and talk the way you'd explain something to a colleague.

    • No training session and no voice profile to set up. Wisper works with natural speech from the first minute.
    • Pause to think, change your mind mid-sentence, or circle back. You don't need to sound polished.
    • Wisper listens only while dictation is active, and you can always see when it is.
  2. Step 2: Let AI refine it

    Wisper turns what you said into what you meant to write.

    • Speech recognition converts your audio into words, using your custom vocabulary for names and terms.
    • A language model then removes fillers and false starts, resolves self-corrections, and punctuates by meaning.
    • Your tone, facts and phrasing are preserved. Wisper makes minimal edits rather than rewriting you.
  3. Step 3: Keep moving

    Clean text is ready moments after you stop speaking.

    • Use it straight away in the email, document or message you're working on.
    • Give important writing a quick read, as you would anything you type. Most of the time there's nothing to change.
    • By default the audio is already gone. Only the text you asked for remains.

Under the hood

What happens in the seconds after you speak.

  1. 01

    Capture

    Audio is recorded only while dictation is active and sent over an encrypted connection.

    Encrypted in transit

  2. 02

    Recognise

    Speech recognition turns sound into words, guided by the vocabulary you've taught it.

    Audio deleted after processing

  3. 03

    Understand

    A language model reads the whole passage to work out your intent, including corrections and emphasis.

    Never used for training without opt-in

  4. 04

    Deliver

    Refined text comes back to you, formatted and ready to use wherever you're writing.

    Yours to keep, export or delete

See the difference

The same thought, spoken and written.

Real-world dictation is messy. Pick an example to see what Wisper does with it.

Replying to a client between meetings

What you say

“hi Sam um thanks for sending this over so I had a look and I think the the proposal is mostly fine but can we, actually let's move the kickoff to the fifteenth instead of the tenth because our design lead is out that week”

What you get

Hi Sam, Thanks for sending this over. I've had a look and the proposal is mostly fine, but can we move the kickoff to the 15th instead of the 10th? Our design lead is out that week.

Your first week

Build the habit in a few days.

  1. Day 1

    Teach it your words

    Add the names, products and acronyms you use most to your vocabulary.

  2. Day 2

    Start with replies

    Dictate email and chat replies that need more than a line. They're the quickest win.

  3. Day 3

    Capture meetings

    Send a spoken summary straight after your next call, while it's fresh.

  4. Day 5

    Draft something long

    Talk through a first draft of a proposal, report or doc, then refine it by hand.

Go deeper

Learn more about writing with your voice.

Start writing at the speed of thought.

Join the waitlist and be among the first to experience Wisper.

No spam. One email when your invite is ready.