Announcement

Collapse
No announcement yet.

Kinect Plug-in - Feature Requests

Collapse
This is a sticky topic.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    #61
    I have been working on trying to get my kinect VR setup to trigger russound zones based what kinect heard the voice command that way it can turn on and off the zones based on where the voice command came from. would it be possible to add an event trigger that if any voice command is recognized run an event

    Thanks Ken
    HSPRO,HS3PRO, BLsecurity, , MCSxap, HSTouch Android, UltraGCIR, CM15a, USBUIRT, BLUSBUIRT, WIFIRGB, BLAB8SS, BLcontrol, BLGData, BLLAN,BLOccupied, BLRadar, BLVolume, iTunesDAAP, UltraGCIR3, Airplayspeak, BLalarm, BLbackup, BLLED, BLrandom, BLReminders, BLRF, BL Speech, Hyperion, IFTTT, KINECT, XBMC, MCSprinkers PRO, PHLocation, ULtrapioneer, Ultralog, ultraweatherbug, Z,troller, GC-100, GC WIFICC, GC-WIFI SERIAL, Nitrogen logic depth camera controllers,

    Comment


      #62
      Originally posted by surovich View Post
      1. Can you add a global set of variables that can be picked up from any Kinect, similar to what you do now for each instance, but in a script its getting old to check for each variable in each instance since its a shared script.
      in the config page (for instances which have a name) there is a setting called "Use Global Vars Specifics to Instance", untick this and it will always use the variables KinectVar0, KinectVar1, ... whatever the instance is

      Originally posted by surovich View Post
      2. Can you get the remote plugin to know the speaker client name (if running on the Kinect system), or let us add it as a passed parameters when we launch the plugin. This way I can direct the output back to single speaker.

      3. Right now when I create an event with a Kinect phrase, it replies to all speakers - so in that drop down, I think it would be great to see a selection for "All Kinect's" for recognition, and a speaker client drop down for replies "Speaker 1" "Speaker 2" or "All Speaker Clients"
      I could let you assign a speaker to an instance in the config page, and let you get that information through a scripting call, but I'm not sure that would help you because to call a scripting function from one plugin instance you still have to know the name of this instance...

      Comment


        #63
        Originally posted by kenrad View Post
        I have been working on trying to get my kinect VR setup to trigger russound zones based what kinect heard the voice command that way it can turn on and off the zones based on where the voice command came from. would it be possible to add an event trigger that if any voice command is recognized run an event

        Thanks Ken
        not sure I understand, do you mean you want a trigger like "Any phrase was recognized" which would fire when any other "A phrase was recognized" trigger also fires?

        Comment


          #64
          Yeah when and phrase is triggered. That way if a voice command is recognized from that instance the event will run. This would allow me to turn on russound zones for tts response in the selected area
          HSPRO,HS3PRO, BLsecurity, , MCSxap, HSTouch Android, UltraGCIR, CM15a, USBUIRT, BLUSBUIRT, WIFIRGB, BLAB8SS, BLcontrol, BLGData, BLLAN,BLOccupied, BLRadar, BLVolume, iTunesDAAP, UltraGCIR3, Airplayspeak, BLalarm, BLbackup, BLLED, BLrandom, BLReminders, BLRF, BL Speech, Hyperion, IFTTT, KINECT, XBMC, MCSprinkers PRO, PHLocation, ULtrapioneer, Ultralog, ultraweatherbug, Z,troller, GC-100, GC WIFICC, GC-WIFI SERIAL, Nitrogen logic depth camera controllers,

          Comment


            #65
            SPud,

            to make this a little more clear, a trigger that would fire when any phrase was recognized so for instance if I have and event that runs when the phrase turn on the light is heard it also trigger the event that any phrase recognized so no matter what phrase was recognized the event would run this would allow the speaker zones to be triggered for tts response every time the kinect vr recognized a phrase and you could have tts confirmation with out changing zones in other parts of the house.
            HSPRO,HS3PRO, BLsecurity, , MCSxap, HSTouch Android, UltraGCIR, CM15a, USBUIRT, BLUSBUIRT, WIFIRGB, BLAB8SS, BLcontrol, BLGData, BLLAN,BLOccupied, BLRadar, BLVolume, iTunesDAAP, UltraGCIR3, Airplayspeak, BLalarm, BLbackup, BLLED, BLrandom, BLReminders, BLRF, BL Speech, Hyperion, IFTTT, KINECT, XBMC, MCSprinkers PRO, PHLocation, ULtrapioneer, Ultralog, ultraweatherbug, Z,troller, GC-100, GC WIFICC, GC-WIFI SERIAL, Nitrogen logic depth camera controllers,

            Comment


              #66
              WAV file output

              I saw somewhere that there was a feature request to save a WAV file of the spoken capture, for uploading via API to a voice recognition service.

              OR - and perhaps an intermediate step - get the raw text from the Kinect SDK recognition engine.

              My application is to do things with the raw text that I didn't previously define in a recognition string. Maybe do it as a sub-grammar. For example, let's say:

              I define an event, "Computer, let's schedule something." Action: load grammar "Scheduling".

              I speak, "Schedule a new event called Guests for 8pm on Saturday." Obviously, I can't pre-define what this will be. But maybe the sub-grammar phrase is "Schedule a new event," and executes a script. The script can pull the raw string from a Global Var maybe, or from a plugin call. Then either I can try to parse it in my script, or maybe I send this to a Google API for execution.

              Just some ideas. Without your code and with no knowledge of the Kinect SDK, I have no idea how hard this ask is to implement.

              Still one of the most useful(/cool/sexy) plugins ever for me!

              hjk
              ---

              Comment


                #67
                Originally posted by hjk View Post
                I saw somewhere that there was a feature request to save a WAV file of the spoken capture, for uploading via API to a voice recognition service.

                OR - and perhaps an intermediate step - get the raw text from the Kinect SDK recognition engine.

                My application is to do things with the raw text that I didn't previously define in a recognition string. Maybe do it as a sub-grammar. For example, let's say:

                I define an event, "Computer, let's schedule something." Action: load grammar "Scheduling".

                I speak, "Schedule a new event called Guests for 8pm on Saturday." Obviously, I can't pre-define what this will be. But maybe the sub-grammar phrase is "Schedule a new event," and executes a script. The script can pull the raw string from a Global Var maybe, or from a plugin call. Then either I can try to parse it in my script, or maybe I send this to a Google API for execution.

                Just some ideas. Without your code and with no knowledge of the Kinect SDK, I have no idea how hard this ask is to implement.

                Still one of the most useful(/cool/sexy) plugins ever for me!

                hjk
                ---
                I can let you add wildcard (*) at the end of a voice command, and then save to a wav file this part of the command.

                so for example you would set up a voice command trigger like "Schedule a new event, *" then when you speak "Schedule a new event called Guests for 8pm on Saturday." you would get a wav file containing "called Guests for 8pm on Saturday"

                Then you could do whatever you want with this wav file, like calling an online voice recognition service.

                Comment


                  #68
                  SPUD! That would be amazing! Taking it a stretch further (not for me specifically, but maybe for others who might request) -- could you include an option/API call/alternate WAV with the whole text, in addition to this wildcard capability?

                  Comment


                    #69
                    On the events page, would it be possible to state the instance of the Kinect plugin involved in the trigger?

                    I have two remote plugin instances. When I go into the event details, the trigger just shows up as:
                    Kinect: A phrase was recognized
                    Phrase = Blah Blah

                    Would it be possible to show it as:
                    Instance: Living Room
                    Kinect: A phrase was recognized
                    Phrase = Blah Blah

                    I can see the instance involved only by going into edit mode for the trigger.

                    Thanks.

                    Comment


                      #70
                      Originally posted by JimBob View Post
                      On the events page, would it be possible to state the instance of the Kinect plugin involved in the trigger?

                      I have two remote plugin instances. When I go into the event details, the trigger just shows up as:
                      Kinect: A phrase was recognized
                      Phrase = Blah Blah

                      Would it be possible to show it as:
                      Instance: Living Room
                      Kinect: A phrase was recognized
                      Phrase = Blah Blah

                      I can see the instance involved only by going into edit mode for the trigger.

                      Thanks.
                      yes, will do

                      Comment


                        #71
                        Originally posted by spud View Post
                        yes, will do
                        Super, cheers spud

                        Comment


                          #72
                          Originally posted by JimBob View Post
                          On the events page, would it be possible to state the instance of the Kinect plugin involved in the trigger?

                          I have two remote plugin instances. When I go into the event details, the trigger just shows up as:
                          Kinect: A phrase was recognized
                          Phrase = Blah Blah

                          Would it be possible to show it as:
                          Instance: Living Room
                          Kinect: A phrase was recognized
                          Phrase = Blah Blah

                          I can see the instance involved only by going into edit mode for the trigger.

                          Thanks.
                          done in version 3.0.0.28

                          Comment


                            #73
                            Superb. Working a treat.

                            Many thanks,
                            John

                            Comment


                              #74
                              in version 3.0.0.28, I have added support for wildcard in "A phrase was recognized" triggers

                              so you can now specify one or several * anywhere in the phrase to be recognized, and the recognition engine will accept any words or group of words instead of the *. Moreover the plugin will record as wav files the wilcard words you spoke.
                              the files are saved as Wave/Kinect<instance_name>Wildcard<wildcard_index>.wav
                              for example if the instance name is "LivingRoom" and you only have one wildcard in the phrase, the file is saved as Wave/KinectLivingRoomWildcard0.wav

                              screenshot below shows a simple example
                              if you say "Kinect, my name is spud"
                              HS will answer back "Hello spud. Nice to meet you! "
                              the "spud" part being your own recorded voice

                              Click image for larger version

Name:	wildcard.jpg
Views:	2
Size:	59.0 KB
ID:	1176686

                              in the same version I have also added the possibility to record any recognized phrase as a wav file, to do that you need to tick the "Save Recognized Phrase as Wav File" checkbox in the config page
                              Last edited by spud; March 11, 2015, 03:43 PM.

                              Comment


                                #75
                                Just to confirm the camera mirror issue has been resolved ok.

                                I like the idea of the console log timestamps too.

                                I'm now thinking of applications for using the wildcard ;-)

                                Great plugin spud.

                                Comment

                                Working...
                                X