About voice assistants

This post is very funny. Sadly, it was closed very fast but I still want to reply to it seriously.
Basically, there are opensource voice assistants that you can use. Just whoogle it, or searx more.
BUT! Not everybody needs these voice assistants in the first place. Out isos are already very big, and pasting something that nobody really needs is not the proper way of making an OS. "Being competitive" is not about copying others, but in fact being better in some ways. People usually use Linux because they are tired of dealing with proprietary OS's and their useless software. And voice assistant will remind people about their traumatic experience, not actually helping them to recover from years of using garbage.
New shortcut keys? Well, if you need some specific shortcuts you should have suggested them, but just saying that you need more doesn't do anything.
More app support? Dude, AUR is literally the biggest pool of repositories on Linux. And we are against Snap and Flatpak enabled by default because their questinable software practices introduce lots of security hazards. If you need some specific package that a lot of people use, you can suggest it in Chaotic-AUR, binaries that are made from AUR sources. But if you need an app that nobody really uses, you should just compile it and add another AUR package, it's usually very simple.


Me, personally, against Snap? Yes. But against Flatpak? No. Flatpak, in fact, is more secure than AUR because of its sandboxing feature. In a recent discussion in Plasma Mobile Matrix group, I heard that Flatpak is even more secure than Firejail.


first searx results


I haven't read it full, but in flatkill.org I just read the "Sandbox is a lie" :scream:

I really don't know why some say it is more secure than Firejail then. I also don't know why people even distribute in Snap and Flatpack, while both of them take space more than double the size and more than double the time to launch (maybe just because they're cross-distro?)


Because they are "user-friendly". People usually don't bother researching on topic that much, so they just fall under a false sense of security after somebody says words like "encryption", "sandboxing", "virtualizing", etc. But sometimes first searx results are your best friend, other times wiby.me really helps with quality guidelines. The point is - do your research beforehand and don't bother trusting someone when they don't know what they are talking about.


Going back to the subject of voice assistant, I think an open source privacy respecting voice assistant of some sort, if exists, could be appealing as an ultimate edition option. But not to me, since I've always been able and opted to type.

What do you all think, can an application that records you and sends it over the internet ever be trusted? Alternatively, can offline voice recognition be made well enough to solve that concern? The open source community should at least give it consideration, since it is an essential accessibility feature of other operating systems, in spite of it being an annoyance for some.

There is an open source voice assistant.

And I guess it is one of the most famous and well known open source voice assistant. But it do have lot of bugs to be fixed.

  1. It never really worked for me. Though I tried it around a year ago, when I was on vanilla Arch, it didn't open at all. Maybe it is due to my low spec hardware.
  2. We can't still add it to iso. You will find a lot of people saying Garuda "bloat", even with Barebones edition. We can't add another hefty package.

When I was using Android phone, I was using Google Assistant just for fun and just to show off. It was inaccurate and impractical for me, maybe because my speaking accent is not as the creators'. Also I don't want to speak in a foreign language unless when necessary.

A voice assistant in Garuda would be a junk for me and would be the first app I rmpkg. No matter how ever much it gets developed, it always remains inaccurate and impractical. I don't want to speak in a foreign language in my house and even if I do I don't want to alter my accent to be like someone else'.


Personally I could never trust anything such as Alexa that constantly eves drops on everything you say. I find these types of apps nothing but an annoyance on my phone. Manufacturers such as Samsung are making them more ad more intertwined with the OS and next to impossible to remove.

This really pisses me off, only adding insult to injury for their overpriced hardware. I want nothing to do with this tech on my home computers. I don't think most people want their home riddled with devices spying on everything they do.


Why I was called spy when I suggested the same thing

You were not called a spy.

competitive with Google assistant and Alexa

Some people don't like even these names. They try to stay away from them. You asked for something like them.

I just shifted from mac that's why I suggested

You asked for Garuda to be like Mac, they simply closed the thread.


I just feel that a voice assistant is irritating and worthless. Why, would you want a process listening all the time, even when you have suspended? Why should it have the possibility of writing calendars and setting alarms if it malfunctions? What if it just slows everything down?


we don't call you a spy, I think you just have not explained how exactly you want it to be implemented. That's why I will explain some possible options:

  1. Built-in, people are mad
  2. With Ultimate edition, newbies are mad because they haven't checked the box to disable it
  3. Small guide somewhere on the forum for somebody who's interested on how to install and configure some opensource voice assistants. Everybody is happy and no storage is wasted. Pretty much everything is already in AUR.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.