Speach to text controller: Home Assistant Cloud (I messed around with whisper but I found that HA cloud responded to my Dinky-Di Bonza Australian accent just marginally better so while I'm testing things I'm going to leave it there for now then once I'm set up I'll move back to whisper and start dialing things in there more) That said the following is what I have in place. So that's probably the first thing people should consider when it comes to the response times here because the more ram and CPU you have the more you can dial up the settings.
It's an 11th gen i5 NUC with 16gb of ram. So probably to begin with I should say that the hardware I'm running on is grossly OP for its mission.
