See pricing

How voice typing stays on your device

What happens to your voice between the moment you speak and the moment the words appear in your note.

What on-device means

Turning speech into text needs a speech model. Many apps send your audio to a server that runs the model and sends the text back. In Mumblepad the model runs on your Mac, iPad or iPhone. Your audio has nowhere else to go.

Mumblepad is not released yet. This page describes how voice typing is being built.

Step by step

  1. Voice typing starts switched off. You turn it on in Settings, and the app asks for microphone access.
  2. You press the voice typing button in the editor. The speech model loads into memory.
  3. You speak. The model turns the sound into words, a short stretch at a time.
  4. The words appear at the cursor as ordinary text. Each stretch of audio is discarded as soon as it has been converted.
  5. You stop. The model leaves memory, so it uses none while you type.

What is kept and what is not

  • Kept: the text, as part of your note.
  • Not kept: the audio of voice typing. It is never written to storage.
  • Kept on your device: the audio of meetings you record and files you import, until you delete it.
  • Not sent: anything. There is no Mumblepad server to send it to.

How to check it yourself

  1. Turn on Airplane Mode, or switch off Wi-Fi and mobile data.
  2. Open a note and use voice typing.
  3. The words still appear. Nothing left the device, because nothing could.

Limits

  • Speech recognition makes mistakes, more so with background noise. Read important text before you rely on it.
  • The free plan includes 10 minutes of voice typing a day. Paid plans have no limit.
  • Dictating into other apps works on Mac only, and needs a paid plan.
Save 50%
Lifetime
Launch price $99Regular price $199
Pay once, keep it. First 101 buyers only.
See pricing

Keep reading