Please forgive me, but I'm in a little deep. I just got my hubitat. got it configured to work with my Ring Alarm and cameras. and got a Safety routine that uses pushover to send me an audible Alert to my iphone.
I'd like to do something similar with Alexa. as a target of the safety monitor routine. The Garage sensor is a simple Ring contact sensor.
basic routine is, garage door opens, triggers a timer for 10min, if it's still open after x minutes I'd like Alexa to announce the garage door is still open..
thoughts? and thanks for your patience with the new guy.
This is straightforward with one small caveat. Echo Speaks is a community supported app that allows TTS and a whole lot of other functionality with Alexa. If you have it installed already the rest is simple.
If you don't have it installed already there's a recently introduced gotcha. It requires some sort of external service to authenticate to Amazon. That can either be run on a cloud service like Heroku, or as a docker container, or stand-alone. Heroku used to be free for hobbyist use so we all went that route, but now it's not free. So you either have to shell out some bucks for Heroku or you have to set up something external to HE running the authentication server. I made the switch from Heroku to rPi recently and it works great.
Once you've done that, the rest is really easy. Here's a simple example that yells if you leave the from door open while the heat is on and it's chilly out.
How many different things do you want Alexa to say? Do you need to construct strings for her to say at runtime? Do you have interest in getting deep into Alexa integration? If so, the Echo Speaks solution may be for you. Be mindful that it is not trivial to set up, especially now with Heroku situation.
On the other hand, if the number of different things you want Alexa to say can be counted on fingers and toes, I prefer to use a virtual contact with switch for each one. When you want her to say that turn on the switch in HE. On the Alexa side, have a routine that triggers on the contact open that causes her to announce (not say) your chosen phrase, and then turn off the switch.
As you have deduced, things are a little fluid with EchoSpeaks at the moment, given that there are ways to host a local server with another device as mentioned by @HAL9000 , as well as being able to still use the Heroku service (now for a fee of course). Adding to the mix is that the EchoSpeaks developer @tonesto7 , has been promising a new version (since the time of the Heroku announcement a number of months ago) that can be hosted solely on the Hubitat Hub. Please do not contact him about this as he has understandably been very busy in developing this as, well as his “real” work and life and has admonished us in the community for our impatience, lol).
Because I trust @tonesto7 to deliver on his promises (he is an incredible contributor to the HE community), although I have an RPi to try the docker container, I am waiting for @tonesto7 ’s all HE version. I would ultimately like to keep things in Hubitat, keeping additional hardware to a minimum, so in the meantime, I have reverted back to my most used and important announcements working directly with the Amazon Echo skill (like it was prior to my using EchoSpeaks) using virtual contacts as @HAL9000 suggested. The beauty of EchoSpeaks is that you can avoid quickly run up against the 200 routine limit in Alexa (one routine is needed for every announcement you wish to make).
I did find a partial solution in pushTV an app for android that will let you use pushover to push to an android device including a Kindle fire (my target) but alas it only works when the kindle screen is unlocked.
I've got it pushing to my iphones at the moment and that will do for now... thanks for the info.
