You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm currently using the best effort inference server, but most of my use will likely be relatively simple and will be entirely running commands on home assistant. At least that's all I can imagine at the moment anyway...
Was considering hosting my own server, but reading about the ability to run directly on device up to 400 commands. I think that's well within the number I would need, even if I made a heck of a lot of aliases for the devices and various scripted automations we have.
I presume a couple of things, but have no idea if I'm right about them - so figured I'd ask and see if anyone knew this stuff yet:
Does the inference server allows for much more natural language than on device processing?
Which is to say, if I go on device I will have to be fairly precise and say Hi ESP, turn on ABC light
And ABC light would have to be the name, or alias, in HA for a light, or well really I guess any addressable entity/device that can be turned of/off.
The point being - There's not a translation of natural language to "Home Assistant Speak" such as...
With the inference server I could say something like Hi ESP, turn the light on in the kitchen and presuming that I've done things right, the server would actually translate that to turn on ABC Light aka put it in "Home Assistant Speak"
Is that correct?
And I guess 1.A is: If I ever wanted the ability to send a variable - as in:
Hi ESP, set a timer for X minutes
or
Hi Esp, send a message to the bedroom speaker that says $a_String_of_Words
would that require a the inference server? Or is that not even possible?
Anything that is listed as an exposed entity to assistant is what the available commands would be.
Is that correct?
And if so - do I have to re-flash each time I add a new command or will they be picked up from HA automagically if I'm running on device?
I've got one more to flash and put into service and a long weekend coming up, so I'll probably give a whirl regardless, but I'm certainly interested in anyone's feedback on this stuff if you're digging into it right now.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
I'm currently using the best effort inference server, but most of my use will likely be relatively simple and will be entirely running commands on home assistant. At least that's all I can imagine at the moment anyway...
Was considering hosting my own server, but reading about the ability to run directly on device up to 400 commands. I think that's well within the number I would need, even if I made a heck of a lot of aliases for the devices and various scripted automations we have.
I presume a couple of things, but have no idea if I'm right about them - so figured I'd ask and see if anyone knew this stuff yet:
Which is to say, if I go on device I will have to be fairly precise and say Hi ESP, turn on ABC light
And ABC light would have to be the name, or alias, in HA for a light, or well really I guess any addressable entity/device that can be turned of/off.
The point being - There's not a translation of natural language to "Home Assistant Speak" such as...
With the inference server I could say something like Hi ESP, turn the light on in the kitchen and presuming that I've done things right, the server would actually translate that to turn on ABC Light aka put it in "Home Assistant Speak"
Is that correct?
And I guess 1.A is: If I ever wanted the ability to send a variable - as in:
Hi ESP, set a timer for X minutes
or
Hi Esp, send a message to the bedroom speaker that says $a_String_of_Words
would that require a the inference server? Or is that not even possible?
Is that correct?
And if so - do I have to re-flash each time I add a new command or will they be picked up from HA automagically if I'm running on device?
I've got one more to flash and put into service and a long weekend coming up, so I'll probably give a whirl regardless, but I'm certainly interested in anyone's feedback on this stuff if you're digging into it right now.
Beta Was this translation helpful? Give feedback.
All reactions