What time is it?

[ikucukkaya]

The "what time is it" command is shown in the available commands however it has no effect. Evan I open a new issue because you told me if i didn't misunderstand you :d

[decentralgabe]

I've been playing with getting this working today and am nearly there. I've been using the SpeechSynthesisUtterance API. Here's the code I've got so far (in a new file named time.js). Haven't gotten sound working quite yet, but this method does work if I run it from a non-mirror js console.

(function() {
  'use strict';

  function getTime() {
    var currentTime = new Date();
    var hours = currentTime.getHours();
    var minutes = currentTime.getMinutes();
    var t = "The time is ";
    if (minutes < 10) {
      minutes = "0" + minutes;
    }
    if (hours > 11) {
      if (hours != 12) {
        hours = hours - 12;
      }
      t += hours + ":" + minutes + " " + "PM";
    }
    else { // (hours <= 11)
      t += hours + ":" + minutes + " " + "AM";
    }

    return t;
  }

  function TimeService() {
      var service = {};

      service.speakTime = function(msg) {
        var msg = new SpeechSynthesisUtterance(getTime());
        speechSynthesis.speak(msg);
      };

      return service;
  }

  angular.module('SmartMirror')
    .factory('TimeService', TimeService);
}());

[decentralgabe]

I had been exploring Artyom, but it looks like, unfortunately, Google has discontinued the use of the Speech API in embedded/dev chromium environments.

https://discuss.atom.io/t/webkitspeechrecognition-google-webkit-speech-api-throws-network-error-in-electron-app/27958/3

Not sure what to look into next...

[evancohen]

@glcohen I think what you are talking about is unrelated to this issue... As for the Speech API, the workaround is in the documentation and is no longer an issue for us.

[decentralgabe]

@evancohen

Could you point me to the documentation with the workaround?

As far as I can tell the problem is that speechSynthesis won't load any voices (speechSynthesis.getVoices() returns a null set), because synthesis isn't supported in Chromium. The rest of the Speech API is not the issue, I have voice commands working just fine with Google API keys.

Edit: Should I open up a new issue for this? I assumed the desired functionality to "what time is it" is to have the time spoken to the user.

[evancohen]

Having the mirror talk back to you is not really a missing feature, I have expressly kept it out of the mirror up until now. See my response in #78 for details.

You should open up a new issue for this and we can discuss the merits of the mirror responding to you via voice there (I'll try to help you get speech synthesis working regardless).

[justbill2020]

removed suggestion label as this is being worked on and is labeled "next"

[justbill2020]

#386 duplicate