ok, got it… voice detection is sensitive…

@strawberry-3-141 said in MMM-voice wont register commands:

@sdetweil the base module hides all modules. the mode has voice and those ones https://github.com/fewieden/MMM-voice/wiki/Supported-Modules

not sure what rules apply for modifications of my module

sorry, that description of mode doesn’t help… what does MODE mean? that you can supply words via voice or text strings, or ??? but its not used?

ok, on my installation, ‘hide modules’ doesn’t do anything…

@strawberry-3-141 said in MMM-voice wont register commands:

@shazglass as the author of MMM-voice I can tell you, that my module doesn’t support a command called hide clock at all. You are probably using hello-lucy, at least that’s a mod I’m aware of which enhances the commands. You can see the command list here https://github.com/fewieden/MMM-voice#usage

then does ‘hide modules’ only hide the ‘supported’ modules list? vs ‘all/any’?

also, could you add some text to the description of ‘mode’ … its unclear to me the value of this , and the possible choices…

i have that, but the text ‘recognized’ is not what i said.

if I speak slowly, i can get ‘magic mirror’
then wake up is ‘wake go’ or ‘wake all’,

hide calendar is ‘hide who’
open help is ‘open go’

the smart-mirror project also uses ‘go to sleep’ and ‘wakeup’

‘hide newsfeed’ is ‘hide to show the’, ‘hide is the’

none of the commands, when recognized, do anything…

hide modules

go to sleep (not on PI so the tvservice doesn’t exist), but no error is logged

i am running MMM-voice on an ODroid Xu4, which is very fast.

I also use another voice enabled smart mirror on this same device.

with a Logitech webcam as the mic.

MMM-voice gets the hotword correct every time, but almost none of the following words.

the same hardware setup works correctly with arecord as the wav source, but uses google speech detection for everything after the hotword.

it appears the system is using the default language model (lm) and dictation (dic) files.

anyone have a suggestion on debugging this?

YOU cannot push to the core repo… you can create a fork (under YOUR github userid), and then clone/push to THAT repo…

you can then propose changes to the master repo, by submitting a pull request, for your branch back to the master (and some branch, usually dev) … and the master owner can accept it , or not… and then merge it with the master branch whenever…

in my module, i need to know the dimensions of the MM main window.

in another mirror project I added an on.resize handler to the main window code, and had it broadcast a window size change to my plugin/module…

how can I do that in MM?

@idoodler thanks… i changed where the data is used and moved the routines to the node_helper… all good…

i have done some work over the weekend to get my two modules and their helpers working to do the basics…

database access is working, rest api server is working. (scheduler helper)
browser window creation and destruction is working, and image loading (handler helper)

one thing of note, a json object transports from module to helper and back. BUT any data added to the object in the helper (like a browserwindow object), does NOT transport (becomes undefined on receipt.)

@hriereb just checked… nothing free there…

cheapest looked like 400 eu /year

@hriereb i see… i created a new userid and see the same thing… IBM now owns this stuff and has decided not to provide free api access anymore… what a shame…

you can’t install in Desktop… its protected…

use your home folder… then make a shortcut to start the mirror

i got an api key the other day without any trouble. just have to login and go to key settings…

https://www.wunderground.com/weather/api/d/docs

thanks… not a nodejs developer generally, so didn’t know about the npmjs site…

also, i run the MM on an ODROID not PI. a lot faster.

i understand how ws works… i built a custom ws server and interface in front of an api emulation platform to support software testing a couple years ago.

thanks… the image is not shown in the dom, it is shown in a separate ‘browserwindow’, floating over the dom. i just loadUrl the image into the window, after creating and positioning it.
the mirror code doesn’t know the images are there

pictorial calendar reminders.

I have tested that the helper needs to do all this window mgmt

The scheduler module does scheduling (compare db data to calendar entries), start/stop viewer(s), and provide indirect services to load images depending on their source (file, dropbox, google drive, one drive).

The viewer module just walks thru a list of urls (getNext()) and opens a window hidden, loads image, hides old window, shows new window, deletes old window. (avoid dead window during load and, minimize screen flash)

yeh, but all this notification event handling really messes up the code i have which is all angular promise based.

I want to open my own browserwindow (show image), but u can’t do that in a module, only in a helper,
but then the window handle object needs to be stored inside the object which is managing the list of images for this ‘viewer’. that object is stored in the module… so, how do I transport the window handle object to the module… is the object valid in the other context (helper vs module).

see the topic ‘what can i do in a module’. (which u already have)

@idoodler

do you have any pointers to examples of websockets from the module? or your “other JS code” means some OTHER library of JS code…

looks like I would have to build another js class to wrap the ws services…(run the server side in the helper) cause I can’t require() in the module itself… load those classes via the scripts…

cool… that would allow my components to communicate in a synchronous mode…

not class variable (I got that now), but shared would solve a ton of problems.