›
This is the forum archive of Homey. For more information about Homey, visit the Official Homey website.

The Homey Community has been moved to https://community.athom.com.

This forum is now read-only for archive purposes.
Closed

Capture specific part of speech to a variable in a Flow?

mtudormtudor Member
in Archive
Does anyone know if it is possible to capture a specific part of speech input / recognition string to a variable.  For example, if I want to write a flow a music app that goes something like:

When: You said anything
And: <text> contains "play"
And : <text> contains "playlist {playlistName} please"
Then: Start playlist {playlistName}

The user could say "OK Homey, play me playlist party tunes please" and the flow could read the playlist that was requested and take appropriate action!

Comments

  • jordenjorden Member
    That requires building an app
  • [Deleted User][Deleted User]
    Would be a nice app. 
  • mtudormtudor Member
    Thanks for the reply @jorden.  I thought that might be the case - thanks for confirming - but it might be quite a nice feature to include in flows.  What do you think @Emile?

    I wonder if it would actually be possible to write an app that could do this as its function...
  • EvertorNEvertorN Member
    I built some basic parsing like that in my Kodi app. It's pretty hard to get right since the speech recognition of Homey is far from perfect. You'd better build it in a conversation way ("I wanna listen to music ".." what play list do you want to listen? ") 


  • mtudormtudor Member
    Ah yes, good advice on the conversation approach.  Would be nicer to process it all in one go but I can see that would have the potential to be hit and miss with speech recognition at it's current state.
  • EmileEmile Administrator, Athom
    These kind of commands we generally include in the core of Homey itself!
  • mtudormtudor Member
    @Emile, are you saying that you would include the "capture" functionality in Homey itself or that you would include functionality to act on specific things said in speech (variable things) as part of the core functionality / app functionality?

    If the latter, I still think it would be powerful to be able to capture sections of speech in the flow editor for when Homey doesn't have something already built in.  It would mean that you could write a smaller number of flows rather than having to create one for every possible condition.

    I think for what I need I can probably get around it for now but in future, it might be worth a look!  Or an app!  Or core changes ;)
This discussion has been closed.