Homey as an Alzheimer/Dementia assistant

I guess many of you know what the disease Alzheimer/Dementia is about. Possible some of you have personal experiences with people close by. There are some robot-animals available and also some other devices that can assist patients with Alzheimer/Dementia like in the link below:

From my own experience, I can report that those robot animals really work great. Many of the other devices are more of a rip-off with many entanglements.

I have experienced that these patients react very much to homey. Like homey commenting when the weather/temperature has changed, telling that there was post delivered, etc. The patient also likes to “talk to homey” and homey has the ability to capture the attention by using the LED-ring when listening/looking for the answer. Compared to other devices homey has a reasonable price… Who would have thought that :slight_smile:

Now for my questions, I could use some suggestions and ideas.

  1. Only “old” homeys will do (pre-2019) since a microphone is required. I would gladly use an Alexa or Google voice assistant, but they miss the attention capturing aspect like homeys LED-ring. Does anyone know of any connectable devices that could create a similar function?
  2. Besides conversation interaction, it is also important that agenda items can be announced via voice. Something like “it is 10:00 am, it is time to take a red pill”. I already tried some with google, but I did not find an option where a schedule item can be spoken at a specific time. Any suggestions here?
  3. This could be a huge “new market” for athom :slight_smile: It would be great if we could add extra functionality to homey as a helpful patient assistant. Any thoughts on this?
1 Like

Maybe a combination with the Google Nest Hub - Smart Speaker, so messages can also be displayed and spoken when needed.

Yes, that crossed my mind also. Unfortunately, the attention capturing visual is quite small on the screen, the patient will likely miss this.

Also, the agenda items are shown on the screen, but not read by some voice. These patients often do not comprehend written text on a screen, that is why calling out an “attention signal”, their name and following the message is quite necessary…

Like the idea, but not all “patients” will react the same, some are deaf and will react more on visual effects and some react better on audio signals, so it will be good to have both options and don’t forget to make a possibility for repeating the message when the patient said, “What are you saying?”, the display could also show the last spoken message.

True, even more reason to find a device that can signal interaction, like homeys LED-ring.

The “what are you saying” thingy might bring issues with the “old” homey. Previous to software version 2.x the card “something was said” worked reasonably well. Since 2.x and later all voice recognition has been downgraded dramatically.

In my experience only the built-in commands are mostly recognized and processed. This would be the main reason to get Alexa or Google for voice input…

What a great and lovely ideas.
I think you should drop the “older Homeys with onboard mic” idea.

When using an additional Google Home-/Ggl nest mini***, you can trigger flows which drive the LED ring with patterns and colors of your choosing, depending on what command G mini should respond.

With the Homey Ggl TTS app, you can let Homey speak, way more clearer and way louder (adjustable) than Homey’s onboard speech.

And while TTS can read any text out loud, you can create flows to let the weather, agenda items, and what not be spoken out loud.
Homey can read a Google Calendar, so anyone permitted could manage the medicine reminders for instance, and the flows with a matching trigger will send corresponding TTS messages.
Maybe it is even possible to “fill” the TTS text with the calendars’ subject text.

Also, an authorised person can send an email to a specific address, and with a matching sender and or subject,
you can trigger a flow.

You can even let a light blink as a reminder, or a sound.

There’s lots more possible with speech recognition, text-to-speech, and visual feedback with the LEDring.
I’d like to help with ideas and solutions.

***) Maybe Alexa can do this also, but I just don’t have one and I just happened to start with GHome

Thanks for your response @Peter_Kawa.

  1. The issue here is that we used to say “OK homey” then the LEDring would start reacting while waiting on the exact command to follow that. It is like homey is paying attention to what you are going to say.
    Using “OK google” will only light a small LED on the G-home device while awaiting the true command or question afterwards. Therefore a patient misses (does not see) the attention LED and does not think anyone is listening.

  2. Is google able to interact/react on “what did you say” while not first using “Ok-google”? Patient will respond, but not with some kind of keyword first…

Graag gedaan Ruud,

Vwb punt 1, Ik begrijp het.
Eigenlijk zou het hele Ggl apparaat een soort glow ring moeten hebben als het luistert.
Dit kan alleen Ggl oplossen, of door een modificatie met een extern LED apparaatje.

Het kan wél een “ik luister” en “begrepen” geluidje afspelen, beide los van elkaar aan of uit te schakelen.
Verder zou er, in mijn beleving, een “go” woord moeten zijn, of er moet langer worden gewacht tot spraak klaar is. Het overkomt mij wel dat als ik even nadenk tijdens het inspreken, bv voor hoeveel graden ik de thermostaat wil hebben, dat Ggl het tot dan toe ingesproken deel begint uit te voeren.
Dat is een kwestie van een time-out instelling die nu niet in de opties en instellingen zit.
Dit is enkel door Ggl aan te passen.

Ik neem aan dat zij dit een mooie toepassing vinden, en hier mss extra opties voor willen programmeren, en wellicht een speciale versie voor de doelgroep willen ontwerpen.
Je kunt er een mailtje aan wagen.

Verder heb je ook Alexa nog, maar ik heb daar geen apparaten van.

  1. Helaas nu nog niet. Dit zou ook een optie moeten zijn en kan enkel Ggl op verzoek programmeren.