[RELEASE] Echo Speaks v3 + Actions

First off, I love the App and that it's provided a lot of functionality tying in HE with the plethora of Echos around the house and has been a big selling point to the wife to invest time and resources into maturing our smart house.

I've tried scanning through all of the documentation and comments in this thread. But I am not able to pull off a custom execute function with Alexa. I have a number of devices that are not IFTTT or HE compatible, but are Alexa enabled. Most recently with the Tuya/SmartLife losing IFTTT support, I have been encouraged to find an alternative or lose some WAF.

The best workaround I see at the moment, is to create an Alexa Routine and use ES to send a command to Alexa to run a routine equivalent to "Turn Tuya On/Off."

Am I missing a function or action in the documentation?

that is how i am doing it. activating an alexa routine. dont know if its the only way, or correct way. but it does work.

So after several days now- Still NO errors. I guess Alexa gets lonely if you don't send her some direct commands....

@tonesto7, I am having the same thing show in my logs. One of my Dots is spamming my logs. I tried the suggestion above to send a command to the Dot in question, but it didn't seem to help, or at least not long term.

Echo Speaks seems to work fine, but I wonder whether this is related to the other error I regularly see in my logs "Too Many Requests Made to Amazon | Msg: Too Many Requests" on this same device.

I know you are busy at the moment, but is there anything we can do to help diagnose this issue?

Why not just do this through Alexa routines in the Alexa app?

Not 100% as it's only been 2 hours but I think i've fixed the lastvoiceactivity spamming issue, I reimported the Echo Speaks Device code from here and then on line 886 and the driver code I put // on the log.debug line to comment it out. So line 886 now looks like the below, for some reason this wasn't being controlled by the debug switch. No sure why it was only the Echo Show that it was affecting but seems fine now.

//log.debug("lastVoiceActivity: ${actData?.spokenText}")

@cjkeenan @neonturbo if you're still having this issue try it out?

Has anyone had any success getting the echo auto to do anything? i tried a test that said when my presence sensor gets home, play a message on echo auto. but it doesn't work. it does work on my other echo products. maybe its a limitation of echo auto?

1 Like

Cool, I will give that a shot!

It was getting so bad that I had to disable the one Echo Dot.

I think it might be a limitation. My Motorola phone has a native Alexa capability, and I can't do anything with that either. I also am pretty sure I tried my Echo Auto and I remember it being limited too. Even the Echo Flex has some limitations about speech.

does Text to Speech work on a gen 2 echo dot?
I tried using it in HSM and got nothing and via the device control page no joy

the echo speaks app shows text to speech as true...

Yes. I have one and it works fine, although I have never tried it specifically in HSM.

@tonesto7 seeing this in my logs since last update...

EchoApp (v3.6.2.0) | wakeupServerResp Exception: java.lang.Exception: No response data exists for async request

Does anyone know of a way to programmatically disable/enable the mic of an echo device? I want to enable/disable the mics based on presence (motion/GPS) as it has unsecured/passwordless access to my alarm system and I am trying to improve the WAF in any way I can.

Does not really matter to me where, in RM, or in an Alexa routine, or hell on my Raspberry Pi, just wondering if anyone has done this.

One workaround I was thinking was to randomize the wake word if the room didn't have presence, so yes it is still listening but no one can easily access it.

Or another idea I had revolved around the programmatic disabling of devices or apps, which would probably work if I disable my alarm integration but I don't know if that is even possible using RM, also does anyone know of a way to disable an app, such as the Amazon Alexa skill?

Why does Alexa have unsecured access to disarm your alarm?? Rather than working around that surely you need to change that?

It's out of convenience, I wanted to be able to arm and disarm with voice commands. Obviously I can disconnect that but I wanted to see if there was another way.

And when I say unsecured access I mean that I have a couple of routines that arm or disarm, not unfettered access, and I guess another way I could secure it is in that RM rule check for presence first.

Depending on your alarm system is there no way to be able to passcode that? I.e I have the Ring alarm system so I can arm and arm through alexa and use that to trigger other actions but to disarm you still have to provide a verbal passcode to alexa.

But to your original point, Echo speaks has a SetWakeWord exposed command, so theoretically maybe you could set it so that if your GPS location is not detected at home then a custom action is run on all devices to set the wake word to "computer" but then once your GPS returns it does the same to change it back to "Alexa"?

Not sure if this would work in real world scenario though?

No unfortunately, I wanted to get the Ring system but was vetoed into getting Simplisafe, but that is a discussion for another time. I think Alexa has integration with SS but not at a routine level for some reason last I checked, only at a skill app level, like "Alexa, ask simplisafe to....".

I was thinking the same thing but even more sophisticated involving random numbers and modulo to select one of the, I think 6?, wake words every like 30 minutes or so. All given that presence isn't there.

But programmatically muting would be my preference as I have recently had cross-room voice detection issues, probably due to the open floor plan design, but tapping into the motion sensors of that room might help alleviate that.

Also a thing to note, if I'm not mistaken, this is only an issue if one, they are in the house, in which case the alarm should be going off, or two, they are outside and screaming through the window in which case, I don't even know if that would be recognizable by Alexa.

Looked into it out of interest and whilst you could mute the alexas that are in other rooms it doesn;t stop them listening just stops you hearing their response, the only option to stop them listening is the mute switch on the device but that's hardware based and can't e enabled remotely.

1 Like

I was mucking about and redeployed the app. Unfortunately, it won't authenticate me now -- I've verified my username/password is correct. It appears to be an added layer of security that Amazon is now enforcing a one-time passcode on new authentications.

I tried adding my cookie information manually, but despite saying that saved, it didn't acknowledge the authentication.

Any chance this might be easy to fix?

EDIT: it is now letting me past the login/OTP, but I'm now getting "Looking for Something?
We're sorry. The Web address you entered is not a functioning page on our site" -- but manually entering the cookie, then logging in seems to work.

Hello guys, i'm running into a problem with this. I started working on this last night and today and I can't seem to get past the login to Amazon Alexa. I've confirmed my user/pw is correct by logging out of amazon website and the Alexa app and logging back in. I even changed my PW, but I can't seem to get past this screen below. I even tried creating a new amazon account to see if it works and it failed as well.

Where is this issue coming from and is there a fix?