MMM-Voice-Control

dr4ke616

So, ive never installed before on a SunFounder Sound Voice Sensor, but if you’re running it on a Pi using the Magic Mirror framework, it should (hopefully) be a case of installing it like any other module. There are details in the README on how to do so.

However maybe there are some other steps required in order to get it to work on the Sensor that I am not aware of.

Baltibu

@dr4ke616 nice one :-)
i Get it finally work ! it answer very quickly … but it s not so easy to get it work with other Modules …
in the readme file, there are some infos . but they didnt help so much .

carteblanche

@dr4ke616 great work on creating this module!

I have installed the module on my Raspberry Pi 3 and am having some issues getting it to work. I checked the console in debug and saw the following error message for MMM-Voice-Control. Any idea what is causing the “network” error? thanks in advance!

SpeechRecognitionError {error: “network”, message: “”, type: “error”, target: SpeechRecognition, currentTarget: SpeechRecognition…}
bubbles:false
cancelBubble:false
cancelable:false
composed:false
currentTarget:
SpeechRecognition
defaultPrevented:false
error:“network”
eventPhase:0
message:“”
path:Array[0]
returnValue:true
srcElement:SpeechRecognition
target:SpeechRecognition
timeStamp:9665.535
type:“error”

dr4ke616

@carteblanche

Kinda hard to say. Could be any number of things.

One thing to note is that the module uses annyang package, which relies on Google Speech API. So perhaps the error you are seeing is related to rate limiting on Google’s Speech API. I believe it is something like 50 requests per day (when using for free).

dr4ke616

Hey @Baltibu

Yeah, i haven’t done much work on actually integrating it with other modules yet. At the moment it just sends notification like messages to other modules, so other modules can then react to it.

There is probably a better more generic way to achieve this. So I am open to ideas.

Baltibu

hey thanks for your answer…
i tried to do the same but the problem i didnt get a comminication between two defferent module …
I get your MMM-Dublin-bus als Referenz . but it still difficult.
by the other should I owerwrite the NotificationReceived module ? or the socketNotificationReceived ?

dr4ke616

@Baltibu

So, it should (if all working ok) be the case that you add a command here. The first paramater references a language entry in the translations directory.

For example in commands.js we register the key dublinbus_start to a function that sends a notification. Then in your translations files you can add what to actually say (in what ever language), for example here.

Then in your own module you override notificationReceived and check for the message sent from your command, as in DUBLINBUS_START.

I understand that this is probably overly complex way of doing this. I do believe this can be done in a better way.

Baltibu

@dr4ke616 thanks a lot .
I added some Command , and i cann see that all my module get the Nofication from the Sender( MMM-Voice-Control) …
but any of this module react of my Notifation …
I will try tonight one more time … hope that I get it work.
I will give some feedback…

dr4ke616

@Baltibu

The current design of this module is focused more on allowing other developers to create modules with voice recognition capabilities rather than it just “work out of the box” with all other modules running on your mirror.

carteblanche

@dr4ke616 I haven’t used annyang before and not familiar with how it works vs other options, but I have used Google Speech API (+ snowboy hotword detection) using a different smart-mirror project…

if this module is using Google Speech API, I would think we would need to input our own API keys but I didn’t see that in the config.