Announcement

Collapse
No announcement yet.

OpenHAB apparently has working Alexa TTS announcements.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    OpenHAB apparently has working Alexa TTS announcements.

    Reference these two links:

    https://community.openhab.org/t/open...-binding/44889

    https://www.openhab.org/addons/bindi...ocontrol/#note

    I'm not an OpenHAB user and never have been, at first glance it looked a little too DIY for my tastes.

    I've been googling trying to find an answer on Text-To-Speech (TTS) announcements via Amazon Echos, a feature the Google Home has had for a while. I find it odd Amazon doesn't have a clear and obvious feature for this. The articles that kept surfacing were related to this OpenHAB solution. Who knows, maybe Google slightly suppress search results related to Alexa in favour of their own platform.

    Skim reading the second link, it seems the author is using the API available from alexa.amazon.com

    The question keeps coming up here on the forums, usually in child posts of other Alexa related threads, about "How can I make an event fire off a text string, to be spoken out loud by an echo/some echos/all echos?".

    So I'd thought I'd post this bit of information in hopes of grabbing the attention of some HSPI developers.

    HS Speak is certainly fit to purpose for making announcements, but it requires speakers and clients distributed around the home. If you own an Echo of any variety you already have a WiFi connected speaker, so why double up on hardware.

    #2
    I don't believe there is an official API. It looks like he simulates a user operating the web based Alexa site. If Amazon changes this, it could break his binding.

    Originally posted by Fellhahn View Post
    HS Speak is certainly fit to purpose for making announcements, but it requires speakers and clients distributed around the home. If you own an Echo of any variety you already have a WiFi connected speaker, so why double up on hardware.
    Only that the Amazon TTS route would be reliant on the cloud (Internet, Amazon & MyHomeSeer) and that is not 100% reliable.
    Jon

    Comment


      #3
      Originally posted by jon00 View Post
      Only that the Amazon TTS route would be reliant on the cloud (Internet, Amazon & MyHomeSeer) and that is not 100% reliable.
      I could live with this for now. Alexa turning on and off my lights is cloud based too, and not 100% reliable... But it beats trying to run microphones to everyroom and somehow get homeseer speaker client to offer all of the same features.


      Joe (zimmer62)

      BLSecurtiy, AC-RF2, RCS Serial Thermostats, RFXCOM SMarthome SwitchLinc, mcsXap, Global Cache GC100, SqueezeBox, TWA_ONKYOINTEGRA, BLLogMonitor, BLPlugins, BLRadar, BLSpeech, BLZLog.aspx, HSTouch (Windows, iPhone, iPod), USB Mimo touchscreens, VMWare Server, Vortexbox, Windows Home Server, MyMovies, Windows Media Center, X10, ZWave, and much much much more.

      Comment


        #4
        Although using Alexa speakers would be great Amazon needs to make it so you can send directly and I doubt we'll see this anytime soon.

        This is why I use Spuds AirplaySpeak to talk to RaspBerryPi Zero Ws running and Airplay app so I can send sound around the house.

        I also wrote PollyC so I could use Amazon Polly TTS to provide excellent voices and to cache them locally so I only need to use the cloud if I use a new phrase.

        It works great and I'm now getting ready to deploy once I finish up a good enclosure for the speakers. Only caveat, do not run anything else on the Pi Zero as it will chop up the sound. Once I have all of this to a point I like I'll publish it here but in anyone has questions post in on my PollyC thread.
        HomeSeer Version: HS3 Standard Edition 3.0.0.548
        Linux version: Linux auto 4.15.0-72-generic #81-Ubuntu SMP Tue Nov 26 12:20:02 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
        Number of Devices: 484 | Number of Events: 776

        Enabled Plug-Ins: 3.0.0.13: AirplaySpeak | 2.0.61.0: BLBackup
        3.0.0.70: EasyTrigger | 1.3.7006.42100: LiftMaster MyQ
        4.2.3.0: mcsMQTT | 3.0.0.53: PHLocation2 | 0.0.0.47: Pushover 3P
        3.0.0.16: RaspberryIO | 3.0.1.262: Z-Wave

        Z-Net version: 1.0.23 for Inclusion Nodes
        SmartStick+: 6.04 (ZDK 6.81.3) on Server

        Comment


          #5
          Originally posted by Timon View Post
          I use Spuds AirplaySpeak to talk to RaspBerryPi Zero Ws running and Airplay app so I can send sound around the house.
          Does this play sounds in sync, and very quickly? If I'm skipping the cloud I need to work as fast as HSSpeaker wired to multizone stereo system.

          If it plays out of sync, or it takes a while to connect up and speak, I'll do just fine with the Alexa speakers depending on the cloud to work.

          I'm trying not to buy a HTD Lync 12 system... I already have Amazon Echo's in every room and the people who live in my house won't give those up without a big fight.

          Joe (zimmer62)

          BLSecurtiy, AC-RF2, RCS Serial Thermostats, RFXCOM SMarthome SwitchLinc, mcsXap, Global Cache GC100, SqueezeBox, TWA_ONKYOINTEGRA, BLLogMonitor, BLPlugins, BLRadar, BLSpeech, BLZLog.aspx, HSTouch (Windows, iPhone, iPod), USB Mimo touchscreens, VMWare Server, Vortexbox, Windows Home Server, MyMovies, Windows Media Center, X10, ZWave, and much much much more.

          Comment


            #6
            I tried the openhab bending. For 3 months or more. Didn't use it that much... But what i know is that i had to reconnect the binding to amazon site every weeks or less.. Maybe because i didn't use it enough.

            Comment


              #7
              PollyC caches all TTS requests and won’t to to the Internet if it has the TTS string in its cache. So if it’s cached it’s extremely fast and if not it depends on your internet connection.

              Id like to see HS adopt Polly for TTS as long as it has the caching feature. If I was setting this up for multiuser use I’d cache it at both the local HS3 and at MyHS level. HS is free to use my PollyC as long as they keep the author credits in the python source.

              HomeSeer Version: HS3 Standard Edition 3.0.0.548
              Linux version: Linux auto 4.15.0-72-generic #81-Ubuntu SMP Tue Nov 26 12:20:02 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
              Number of Devices: 484 | Number of Events: 776

              Enabled Plug-Ins: 3.0.0.13: AirplaySpeak | 2.0.61.0: BLBackup
              3.0.0.70: EasyTrigger | 1.3.7006.42100: LiftMaster MyQ
              4.2.3.0: mcsMQTT | 3.0.0.53: PHLocation2 | 0.0.0.47: Pushover 3P
              3.0.0.16: RaspberryIO | 3.0.1.262: Z-Wave

              Z-Net version: 1.0.23 for Inclusion Nodes
              SmartStick+: 6.04 (ZDK 6.81.3) on Server

              Comment


                #8
                FYI:

                Since the binding version 2 of the beta 2.4 binding ( echo control binding), i didn't have any connections drop since two weeks. So it works rock solid.

                I personally use it in event to control the sound output of my alexa depending of the hour of the day ( so 75% in the day, 30% at 11 pm, etc; i change the sound like 5 times in the day). I created a virtual device wich acts as a master volume control for all the alexa.

                ----

                A new cool feature: you can now have room awareness for everything you want. You transfer this information to Homeseer via node-red and mqtt. Then, depending of wich echo hears the message and what was the message ( so this depend of the mqtt topic payload you receive) you do not have to say the room.

                So you create an Alexa routine like : turn on the tv. And then depending on what room you are the good tv will turn On


                So yes the binding read the last spoken command for each echo. One drawback for now is the speed you tell alexa command. Ex: IF you say alexa turn on the tv, then the last hear command will be ''alexa turn on the tv. But if you say ''alexa'' then pause 1 sec, then say ', turn on the tv the plugin will capt ''turn on the tv''.
                I found this to be a problem with mcsmqtt. The plugin do no trigger for two same topic with different payload on the same event ( maybe need to do 2 events separately..). It will trigger on the first one you create.

                I have talked about this with the binding author, and the word alexa will be filtered out in the next beta release ( beta 4)

                Comment


                  #9
                  More easy to setup... a Node-red module : https://flows.nodered.org/node/node-...b-alexa-remote and yes there is a ''tts'' (limited by amazon code of course) node.

                  I will stay with openhab for now.... but it might help others!!

                  Comment


                    #10
                    Matt, if you associate a device with your Alexa topic you can filter “Alexa” from the payload so the two messages will trigger the same. I suspect you will use the List device type so different payloads will trigger different events.

                    Comment


                      #11
                      Thanks Michael , I will give this a try

                      Comment

                      Working...
                      X