​
Overview

Custom Modes in Superwhisper provide flexibility and advanced options not available in other built-in modes. These include the ability to record from system audio and speaker separation when using Nova Cloud models.

There are several scenarios where you might want to use Custom Mode features without AI processing:

  • You need to record and transcribe audio from applications or websites
  • You want clean speaker-labeled transcripts without AI post-processing

​
Step-by-Step Process

1

Create a Custom Mode

First, create a Custom Mode with the specific features you need:

  • Enable system audio capture if needed
  • Activate speaker separation if required
  • Give your mode a descriptive name
2

Locate the Mode Configuration File

Navigate to the Superwhisper configuration folder:

  1. Open your Documents folder (or the custom location if you’ve modified the setting)
  2. Find the “Superwhisper” folder
  3. Go to the “modes” subfolder
  4. Locate the JSON file for the Custom Mode you just created
3

Edit the Configuration File

  1. Open the JSON file in any text editor
  2. Find the field labeled languageModelEnabled
  3. Change the value from true to false
  4. Save the file
4

Restart Superwhisper

Close and restart the Superwhisper application for the changes to take effect.

5

Verify the Changes

After restarting, open your Custom Mode. You’ll notice that the AI processing sections of the interface are no longer visible, confirming that AI has been deactivated for this mode.

When AI is deactivated, you don’t need to add any custom system prompts or AI instructions. The mode will function similar to a voice mode but with the advanced capabilities you’ve enabled.

Voice Modes

Learn more about voice modes and how to select the right model for your needs.

Speaker Separation for Meetings

Discover how to set up speaker separation for meeting transcription and processing.