So Iām thinking of purchasing a google nest mini for TTS.
With all the recent changes google has done, is this feature still working?
I have one. With the Chromecast Integration (Beta) device driver, it works okay. I have sometimes had it not respond, but going to the device page and initializing and refreshing it restores the function. However, those have been quite rare since I put the speaker on an UPS. Also, I reserved an IP address for it in my router so it does not change.
I use it to announce mode change, announce when a contact sensor has been open over 1/2 hour and arming/disarming mainly. I have other announcements that are set, but they would only happen if something like the CO or Smoke alarm went off or an alarm were triggered.
p.s. I have the microphone turned off because I want it for a speaker. I don't want it listening to me. I value my privacy. I got it for no cost because I have been with Google Fi for my phone service and it was an offer that was hard to refuse.
They work well for me. I have it announce when the garage door has been opened / closed, if a water leak sensor has been triggered (and what sensor it came from ā¦ I have 12). Also, because there are several tasks you can't automate with Amazon Alexa, I use Chromecast TTS to "tell" Alexa what to do!
-
Alexa Guard can enable with Geofence, but cannot disable. I have a rule set up that when I enter an area near home (triggered by a virtual presence switch) that a Google Home says "Alexa, disable Guard mode" Works like a charm!
-
I have a sleep sequence set up where I want my bedroom echo show to play rain sounds to "lull" me to sleep, but you can't automate a sleep timer or shut off the screen. In this scenario, after my "good night" Alexa routine runs, I have a Google Mini tell "Alexa, set a 15 minute sleep timer" and "Alexa, screen off" .. this works great as well.
Cool.
I have an Alexa, but the way I have TTS working isn't correct really.
I have routines setup to say something once a virtual contact sensor opens.
I rather do this work under the hub.
I'm thinking of picking one up and selling my Alexa Echo Dot.
@StephenH or @Tim-in-Ca -- How are you automating playback of sounds on your Google speakers? I have my speakers set up and can play a sound by URL in the device page, but the "Simple Automation" app does not allow me to target the speaker as it would a light switch. Is this a project for Rule Machine or is there another app best used to watch sensors and drive sounds?
I use mine only in the Notification App. I occasionally have to rediscover them to get them to work again.
Mine is set up for TTS announcements. I use Notifications to let me know of mode change. I have a Rule Machine rule to notify armed/disarmed. Here is a screenshot of my rule:
Thanks @Tim-in-Ca & @StephenH, that helps. I must be missing a piece of the puzzle still--from the Notifications app I don't see a way to specify the URL or a sound file. Another app must be required.
Have you installed the Google Mini as a device using the Chromecast Audio driver? You also need to go to the Settings (left-hand column) and select Hub Details. Then select your default TTS voice. After that, the TTS works as that part is built-in. It does not require a sound file for TTS notifications. Don't forget to configure and initialize on the device page also. You can test there by typing something in the Speak text entry box and clicking on the Speak button. I also have the Stop after TTS is complete enabled.
Hopefully, that will be of assistance.
Edit:
Okay, I saw that you are posting in another topic about the same subject. If you are trying to play an mp3 file, then it is an entirely different process. For that, you will need to have a server to be able to stream the mp3 file. I have a web server on my main system (running Linux). I can have an mp3 file there and then I would just provide the URL to that file such as entering the link (h t t p ://www.myserver/mysoundfile.mp3) in the appropriate location such as the Play Track in the device settings or in a rule. I have tried this and it does work, but I mainly wanted the Mini for TTS notifications, so I don't use it this way. I also have the microphone switched off because I don't want to try to use it for voice commands. I'm perfectly happy not having Google listen in on my everyday life, not that I have anything worth listening to. It would be pretty boring listening anyway.
I have the Mini integrated, files are available on http, and I can play them from the Mini device page. The hangup was--as it always is--which Hubitat app(s) were needed to put it all together. I now see that I must use Rule Machine. Too bad, since the Notification app has everything I want... except the ability to play a sound instead of TTS. With so many complaints about RM performance I was hoping to avoid it. C'est la vie.
Rule machine is not THAT bad. It has a learning curve, but it is logical once you start to figure it out. After all, what I posted to you above is a rule machine rule. I don't have issues with the performance. Every once in a great while, the Mini stops responding, but that is likely in response to a power issue. Reinitializing the Mini gets it going again. That has only been a couple of times since I got it last year as a free product for being a Google Fi customer. The more recent updates have made it more reliable as it is no longer listed as a beta driver.
It's not the complexity of RM that worries me. There just seem to be more unhappy than happy users. But, I guess that is the nature of a forum. All I ever do is ask questions and complain! Maybe in a year I will be able to post help and pay it back. 
For me, I never really used RM much until the RM4 was released. I think I had one RM3 rule, but I found that for a couple of things, RM was the way that worked best. I started when I was having trouble with the Thermostat Scheduler. I wrote rules to control the thermostat and then to periodically check to make sure the setting was appropriate and to change it if it somehow had not switched at the scheduled time. This has been rock-solid. My notification rule has been very reliable as well. I think that many of the complaints about RM are about people using faulty logic to set the rule up so that they are trying to do contradictory things. Sometimes, laying it out on paper first to make sure one knows what he/she is wanting to do will help in constructing the rule(s) to accomplish the desired goal instead of just banging away at the keyboard and hoping for the best.
1 Like
Stephen.
Wouldn't it be enough to put the MP3 file on the Hubitat as a file and specify that URL?
That I don't know. If you try it, please let me know the results. The rules I have now work for resetting the volume when it changes to 45% at night. The speech part has been very reliable lately so I use the rule, "if it ain't broke, don't fix it."
