I have an issue I’m hoping someone can help with regarding voice processing. When HA’s built-in intents were in their infancy, I setup an automation with a sentence catch-all wildcard that directed all commands to a python program for processing (effectively avoiding the built-in intents altogether).
Given the maturing of HA voice and continued conversation etc I would like to use the built-in intents again. The problem I have os that even if I disable/delete the sentence catch-all automation, HA still matches all incoming voice requests to that wildcard (confirmed by using the Assist test feature in developer tools) and will not trigger any built-in intent.
Does anyone know how I can get rid of this wildcard trigger matching? Any clues would be greatly appreciated.