As most of the Lemmy users are security/privacy minded, after a question about alternative voices I’m wondering. Why would you allow a virtual assistance to spy on you constantly? (Especially when you don’t need to due to other limitations)
And as a side note, is there a way to kill Siri completely on IOS (not just go trough all app settings and disable siri there)? (company phone, so alas not jail broken, Android is rooted and GA has been removed, as is the Google app itself)
You must log in or register to comment.
Actually, it can be useful in the car to send a message to someone like „I’ll be there a couple of minutes late“ or „shall I bring home something?“ without distracting yourself.
Would you like to send “How much for a nickel bag?” to Grandma?..Sending
With E2E encryption? Sure.
Edit: Grandma’s got the good stuff. I’ll risk it.
It’s easier to say “hey Google, turn off all lights” than go to Home Assistant and turn each off by hand.
Rhasspy works great for that
I’m sure. But my raspberry pi 3 can barely run Hass without overheating. (had to remove adguard from the same device because that made it overheat a few times a week)
Maybe when I upgrade it to a proper home server, I’ll look into replacing my Google home speakers.
I’m running domoticz on a 3 with an rf-link. Although it just manages some lights, it’s usually 99%+ idle.
I have a remote for that, kaku remote which sends an off for device one and domoticz turns the lights off for me. When I forget, there is a fixed time in a schedule for off and 2 times for 1, 1 fixed, 1 30m before sun down.
I agree, I find using a web interface/app to cumbersome, the remote is a lot nicer to use. (And I have the original remotes available as well, for when I messed to much with domoticz. ;) )
Even easier to pair a sensor to Home Assistant so the lights can turn on automatically when you enter the room. Then you don’t need to interact with a voice assistant or an app to turn on a light.
Why would you allow a virtual assistance to spy on you constantly?
Because it’s not? A low-power process on my phone is listening for the wake word. When it hears other stuff, it ignores it. When it hears the wake word, it processes my request, tied to a separate anonymous identifier used only for Siri itself. I’m not really losing any privacy at all.
And as a side note, is there a way to kill Siri completely on IOS (not just go trough all app settings and disable siri there)?
It’s just the first two toggles (Listen for “Hey Siri” and Press Side Button for Siri) in the Siri & Search menu that you’d need to turn off. There’s not much to it.
I’m all about informed consent - I’m wiling to compromise privacy as long as I know how much I’m compromising in exchange for convenience - using virtual assistants to shoot off texts, make calendar events, etc is invaluable to me for many reasons and it improves QOL more than it harms it - I do what I can to limit how much access there is to my queries, but thats just how I approach it
People use stuff like Alexa or Google home mostly as a voice activated timer, weather reporter, or speaker. If they have light bulbs maybe they use it that too. It can also be handy for voice activated controls for a movie if it can connect to something like a fire stick.
Most people don’t actually know what might be involved in terms of privacy concerns. They see the convenance of tech or something that’s neat. A vast majority of people simply aren’t doing things like uninstalling Google apps or know what a GPL refers to.
There are open source and local versions of home automation and voice assistants.
I have Alexa because it was on sale and useful to me. I don’t really think out loud and can unplug the thing if I want to talk about unionizing. When I replace things or buy new ones I’ll be keeping in mind compatibility with open source or local solutions. Alexa does integrate with a lot of stuff though. I do think it will be discontinued at some point. Amazon loses money on the product and it hasn’t been increasing sales like they hoped. No one trusts the damn thing to buy crap off Amazon for them.
I have an Amazon echo that is a glorified weather reporter and cooking assistant. It’s pretty useful for setting timers and for calculating conversions for measurements. I also use it as a white noise machine for bed time.
I do get the random interruptions where it some how responds to some random conversation I am having and I promptly cuss it out. It’s probably foolish for my privacy, but I don’t actively plan crimes in front of it, but who knows how it might bite me in the ass in the future.
The one advantage that Alexa legitimately had over Siri was the ability to set multiple timers. Very handy when doing multiple dishes at a time.
I’d only set one up if it were fully open-source and self-hosted. It looks like there are a couple of options out there that meet those criteria (like Mycroft), but I haven’t looked into it enough yet.
Not sure why people downvoted this. I’d love to self-host a virtual assistant, both for privacy and for the fun of getting it to work.
I can relate to this stance. Same here with home automation. I have minor automation which is only accessible from inside my network (and back-uped with the original remotes, so I don’t rely on it)
When you can limit it to the home, it could be nice to work with, but I don’t trust apple, amazon or google with permanent access to the mic. I heard Siri stating she doesn’t understand the question via smart watches while in online meetings or even worse, in areas where phones aren’t allowed.
I’m still using Clippy.
Why would you allow a virtual assistance to spy on you constantly?
- part of the job
- “nothing of value was lost” environment
- a bet
- there would be no other way (family, local conditions or other factors enforcing such a scenrio)
These are some scenarios that I could think of…
Summed up: convenience, and curiosity.
I can lay in bed and ask if to turn off the lights, and all lights in the house shut off (including the ones the kids forgot to turn off). That convenience is phenomenal.
And my kids who are too young to own their own devices can ask it questions. Whatever is on their random little minds, which very frequently leads them to a topic that they then come to us to discuss in depth. I love that it helps them foster their curiosity about the world around them and often enough leads to healthy conversations with Mom or Dad.
that’s really sweet - how often do you find that the assistants do a good enough job answering their questions?
Usually by the second or third question they got as much out of Google Assistant as they are going to get (because they just keep asking variations of the same thing, but don’t quite grasp that it is really the same question, or they have now gotten to a hyper specific question). By question 5 or 6 they typically are done with Google and come to us to discuss further.
I use one so that I can control my lights and music at home. they’re also helpful to automate things like playing music when I turn my car on or put my headphones in.
I don’t I turn them all off. Too many privacy issues. Give me something that runs 100% on device/premises or fuck off.
I use it almost exclusively to set timers when I’m cooking so I don’t have to touch anything. Once in a while I’ll use it to initiate navigation or look something up but search results have been getting increasingly shitty/useless these last few years.
Not in their current form.
If a completely self-hosted/controlled one arrives in a decade, maybe.
I use Siri very infrequently, and mostly for simple things.
“What time is it?” Is probably the most frequent one. Mostly if I’m walking and don’t have my watch on. It’s nicer to be able to get a quick answer through my headphones than have to take my phone out.
Timers and alarms also. But that’s about it
