Automating Voice Interaction

Hello everyone,

I’m an enthusiastic user of Home Assistant and have successfully implemented the Assist function. Currently, I’m using a Wyoming Satellite with a microphone array as the input source (in combination with the Stream + OpenWakeword addon on Home Assistant) and an ESPHome Media Player for text-to-speech (TTS) output. I use google STT, openai echo TTS and the build in OpenAI Conversation integration, which all works pretty charming.

I would like to expand my automations by adding a feature that allows me to trigger the STT audio imput of the Wyoming Satellite from an automation or script—just as if the wake word had been recognized.

A specific example: If the outside temperature is lower than the inside temperature, I want the Media Player to say, “It’s cooler outside than inside. Should I turn on the ventilation?” After this announcement, I would like to trigger STT (Speech-to-Text) as if the wake word had been activated. The conversation should then be passed to the Assist function (OpenAI-Connected) to perform the desired action.

Has anyone had experience with implementing such a feature? I would appreciate any help or suggestions!

Thank you in advance!