Ive just done this. The trigger can be (location+time of the day) in your case. Google home probably has a URL where the text recognised by the voice command can be passed and it would act.
In my case, the switch has a URL with arguments to turn on and off the switch. On press of a shortcut that executes a flow I passed that URL+arguments using the HTTP request block and it worked like a charm.
I would like to see a plugin/block like IFTTT has for Google Assistant. In IFTTT, the widget is configured for 3 different text (recognised by voice) and the action supports a HTTP call through webhooks. Webhooks needs a cloud accessible URL on a public IP. I want a private LAN URL and I cannot get a static public IP for my WAN. My WAN is also a RFC1918 address!! The voice recognition block here takes the recognised text into output variables that we need to comapre and then act. I'm not sure how to pass the variable from the speech recognition block to the next comparison block.
Can someone help?