I'm facing a frustrating issue with the LLM module (llm-audio) in my KWS (Keyword Spotting) project. My goal is to create a system where:

The module detects the activation word (e.g., "Hi") but does not produce any sound during this process. Instead, it should display a notification or message on the screen.
The module responds only as text on the screen without any audio output.
I’ve tried several approaches as documented in the LLM API, including:

Configuring the module with setup and setting enoutput: false to disable audio output.
Pausing the module with pause.
Stopping tasks or resetting the module with sys_reset(True).
Unfortunately, none of these methods have worked. The speaker continues to emit sound, and I haven’t been able to suppress it.

If anyone has experience or insights into how to fully mute the LLM module, I would greatly appreciate your help. Thank you in advance!