WTH don't intents contain the entity_id of the speaker/device that recorded the voice input?

The entity_id of the device that captured the voice input should be part of the intent (And the intent should be emitted as an event).

This gives context to intents. For example “Turn on the lights” while spoken in the living room could turn on the lights only in the living room, while “turn on the lights” spoken in the kitchen could turn on only the lights in the kitchen.

If you add your assist entities to an area they will default to turning on lights in that area if you don’t specify a location or entity.

1 Like

Thanks, but if you want to use the info in your automation you can’t, afaik.