I'm still waiting on my LED lights, which is making me nervous. Perhaps I shouldn't have ordered from eBay
So, in the mean time, I've been working on some automation stuff with lights (when I get them). I've always wanted voice control, but something that is continuous, and does not require pushing buttons prior to the voice command. I found out about how Tasker can send HTTP Post/Get commands to OpenHAB. Tasker also has a plugin called Autovoice, which can run in "continuous" mode, so it will always monitor your voice. So here's an example of how to do this integration.
Here's the holiday lights screen.
Video of a quick demo.
I have an item defined this way:
Switch itm_holiday_on "Lights Enable"
Under Tasker, define a new Task to turn off the switch.
New Task > Action > Net > HTTP Get
itm_holiday_on = OFF
Do a similar task for turning it on, but just replace the attribute with "=ON". You can test the task by using the play button at the bottom. Executing these tasks should turn the switch off or on.
Next, we use voice commands to activates profiles that run these tasks.
In the Profiles tab on Tasker, create a new profile.
Event > Plugin > Autovoice > Recognized
Edit the configuration for this event. Set the "Command Filter" to whatever you want the voice command to be. Then go down to pick a "Trigger Word". In my case, the Trigger Word is "Computer", and the Command is "on" or "off". Exit out of the profile creation and you'll be prompted to pick a task for this event. Pick either the on task or the off task created previously.
Next, open up the Tasker plugin called Autovoice. Set it to run continuously. This will make it run voice recognition all the time, and act as the "computer" in Star Trek.