Logic Machine Forum
[LM APPS] Voice Configuration app - Printable Version

+- Logic Machine Forum (https://forum.logicmachine.net)
+-- Forum: LogicMachine eco-system (https://forum.logicmachine.net/forumdisplay.php?fid=1)
+--- Forum: Application Store (https://forum.logicmachine.net/forumdisplay.php?fid=11)
+--- Thread: [LM APPS] Voice Configuration app (/showthread.php?tid=1092)



[LM APPS] Voice Configuration app - Monolithos - 18.11.2017

Hi,

I have been playing around with the app, following Daniel's tutorial.

I'm a native Dutch speaker so I use Dutch in the configuration.

Got it all working right away but I don't really like talking full sentences with a machine when only a few keyword are needed.

I found a very nice and probably unintended (please don't get rid of it) use of the configuration to prevent you have to use (and maybe forget) the longer commands

so instead of saying 'activate scene diner in the dining room' I just want to say 'activate scene dining' since it's quite obvious where you'll be dining.
Same example for your downlighters above your couch (how many couches does a normal person have?) : instead of saying 'turn on downlighters above couch in living room' I wanted to say 'lights couch on' or even 'couch on' and 'dim couch to 10'

I found it was possible :
Just create a place (Places tab/desc field) that has the same name as the names field in the objects tab.

When giving a command, a location and an object is expected. If both have the same name, that name apparently only needs to be said once.

So, for instance for scenes
Place tab : create a place with desc-fiels named 'scene' (the id can be whatever you want)
Objects tab : create line with command 'activate', type 'set', value (the scenenumber), place (the id you entered during the creation of the place 'scene'), names (the name you want google or siri to respond to) and of course the knx object for your scene in the 'objects' field

A simple command like 'activate scene tv' will then be possible. No need to specify the room

just wanted to share this for all those other people who feel stupid talking to a dumb smartphone  Wink


RE: [LM APPS] Voice Configuration app - Daniel - 21.11.2017

(18.11.2017, 14:38)Monolithos Wrote: Hi,

I have been playing around with the app, following Daniel's tutorial.

I'm a native Dutch speaker so I use Dutch in the configuration.

Got it all working right away but I don't really like talking full sentences with a machine when only a few keyword are needed.

I found a very nice and probably unintended (please don't get rid of it) use of the configuration to prevent you have to use (and maybe forget) the longer commands

so instead of saying 'activate scene diner in the dining room' I just want to say 'activate scene dining' since it's quite obvious where you'll be dining.
Same example for your downlighters above your couch (how many couches does a normal person have?) : instead of saying 'turn on downlighters above couch in living room' I wanted to say 'lights couch on' or even 'couch on' and 'dim couch to 10'

I found it was possible :
Just create a place (Places tab/desc field) that has the same name as the names field in the objects tab.

When giving a command, a location and an object is expected. If both have the same name, that name apparently only needs to be said once.

So, for instance for scenes
Place tab : create a place with desc-fiels named 'scene' (the id can be whatever you want)
Objects tab : create line with command 'activate', type 'set', value (the scenenumber), place (the id you entered during the creation of the place 'scene'), names (the name you want google or siri to respond to) and of course the knx object for your scene in the 'objects' field

A simple command like 'activate scene tv' will then be possible. No need to specify the room

just wanted to share this for all those other people who feel stupid talking to a dumb smartphone  Wink

Hi
Thanks for your feedback.
No worries we will not change this. Our goal was to make it as much flexible as possible.
The system just expect you to say the 3 words. Command, Name and Room but it do not care if you will call your room TV or couch  Smile . The resat of the sentence is deleted anyhow as we just doing string parsing.
You can use any combination of sentence as you want as long you use the 3 key words then command will be executed.
More feedback are welcome.
BR