Using Voice Dictation

Kantrad lets you dictate your reports by speaking instead of typing. The speech recognition runs entirely on your computer, so no audio is ever sent over the internet. Your recordings stay completely private.

Setting Up Dictation for the First Time

You have two ways to get started:

During the welcome wizard:

  1. When you first open Kantrad, the setup wizard includes a Set Up Dictation step.
  2. Click it and Kantrad will download the recommended speech model (about 142 MB).
  3. Once the download finishes, dictation is ready to use.

From Settings (if you skipped the wizard):

  1. Open Settings, then go to Voice Dictation.
  2. Click Quick Setup to download and load the recommended model.
  3. That is all — you are ready to go.

Choosing a Speech Model

Kantrad offers several speech models. The recommended one works well for most people, but you can choose a different one if you prefer:

ModelDownload SizeSpeedAccuracyLanguages
Tiny (English)78 MBFastestGoodEnglish only
Base (English)142 MBFastGoodEnglish only
Small (English)488 MBModerateHighEnglish only
Medium (English)1.5 GBSlowHighestEnglish only
Tiny (Multilingual)78 MBFastestGood99 languages
Base (Multilingual)142 MBFastGood99 languages
Small (Multilingual)488 MBModerateHigh99 languages

Tip: The Base English model is the best balance of speed and accuracy for most radiologists. If you need to dictate in another language, choose one of the multilingual models.

Dictating Your Report

  1. Place your cursor in the report where you want the text to appear.
  2. Press F2 on your keyboard or click the microphone icon in the report toolbar.
  3. The microphone button turns red to show it is recording. Start speaking.
  4. When you are done, press F2 again or click the microphone icon.
  5. Your spoken words are converted to text and inserted at the cursor position.

Dictating in the AI Chat

You can also use voice dictation when chatting with the AI assistant:

  1. Click the microphone button next to the chat input box.
  2. Speak your question or instruction.
  3. Click the microphone button again to stop.
  4. Your words are converted to text and sent to the AI.

Speaking Punctuation and Formatting

While dictating, you can say these words to insert punctuation:

Say this…To get this…
“period” or “full stop”.
“comma”,
“colon”:
“semicolon”;
“question mark”?
“exclamation mark”!
“new line” or “next line”A line break
”new paragraph”A blank line (new paragraph)
“hyphen”-
“dash”
“open paren” / “close paren”( / )

Kantrad automatically capitalises the first word after a period, question mark, or exclamation mark.

After Setup, It Just Works

Once you have set up dictation, the speech model loads automatically every time you open Kantrad. You do not need to configure it again.

If you manually unload the model in Settings, it will not load automatically until you load one again.

Microphone Permission on macOS

The first time you use dictation, your Mac will ask for permission to use the microphone. Click Allow.

If you accidentally clicked Don’t Allow, you can fix it:

  1. Open System Settings on your Mac.
  2. Go to Privacy & Security, then Microphone.
  3. Find Kantrad in the list and turn it on.

No Licence Key Needed

Voice dictation is available to everyone — you do not need a licence key. It works independently of the AI features.