Also got a nice Dell 7390 for a similar price a year ago. Though you really can’t compare a laptop bought in 2019 with a laptop bought 6 years in the future. You’d need to compare it to a refurbished one available for a similar price in 2019 and then factor in how that turned out for you a few years later. I mean technology always progresses and you’ll always get more a few years later. But yes, I’ve always been a fan of refurbished enterprise-grade laptops instead of the super-cheap consumer ones which include as much cost-cuttings as possible and a legacy CPU which is upmarketed because it’s cheap. I think my old desktop Celeron N4500(?) was like 40€ when it was new, because it was leftovers in production. At that point you can always buy a used processor for the same price with double the processor cores.

Nice, thanks for the link! I wasn’t aware of that. Sadly as with all shiny new things it doesn’t fit all my requirements… I’d really like to speak to my house in my native language. But I figure English will do. I’m gonna try that.

Not sure if an ESP32-S3 is fast enough for more advanced DSP plus the rest of an voice assistant. At least I found some ESP32 libraries with noise reduction, echo cancellation… There is the ESP-ADF and a project called ESP32-SpeexDSP. But I didn’t try that yet. The Rockckip / Luckfox development board looks nice as well. A Cortex-A7 and a few hundred megabytes of memory might come in handy. And whatever the NPU does. But I don’t have a clue what kind of software and libraries we got for embedded Linux or custom processing units.

Anyway. I think the production-grade stuff mostly uses multiple microphones and a combination of beamforming and echo cancellation. I got 4 inmp441 microphones here. But I lack the software/libraries to tinker with that kind of signal processing.

Uh, noise cancellation is hard. First of all, the audio pipeline currently isn’t able to resample the microphones, so mic and output need to be connected to separate i2s buses, or it won’t work simultaneously in the first place.

And then I had some luck with the microwakeword component. It often triggers correctly even with noise in the background. And I have an automation that mutes all media players and the TV when the wake word is triggered. That’s my “noise cancelling”.

I think more elaborate noise cancelling is going to require some dedicated hardware (or maybe some proprietary ESP-ADF functions) and a microphone array. But that’s probably as expensive as an Voice PE?!

I’m not in a good place with the voice assistant anyway. Don’t own a graphics card. So it’s slow. And Whisper never gets all the words right for me. So it’s down to the speech-to-phrase addon. And that seems to be broken as of now. At least I get more connection errors than commands through. I think I’m going to do the Sendspin media player first. And then maybe add a microphone and voice assistant later.

I’m currently doing it the other way around. Assemble multiple satellites and spread them through the house. With upcoming Sendspin and Music Assistant this might do whole house audio soon. But I don’t own a Voice PE. I just bought some microcontrollers plus MAX98357A codec/amplifiers and connect them to random old speakers I have in my e-waste / upcycling bin. The one thing with an 3.5" audio jack might just go into the preexisting soundbar or stereo in the livingroom.

No worries. Your post was well-written. And I’m glad people could offer some advice. Not even the proficient Lemmy users get all of this right all the time. I just figured I’d drop you a comment in case the mods take action, to spare you the effort to also learn about the modlog and how to look up their note… But seems it wasn’t necessary 😄

Sorry, I don’t have an answer to your question, but two other communities that would fit: !homelab@lemmy.world and !homelab@selfhosted.forum

They’re both not really active, though. And someone asked about OpenSense hardware before and didn’t get any answers…

Just writing this so you have some other places to look up, in case your post gets deleted, I think you’re technically in the wrong community here. As per rule 3 in the sidebar, this community isn’t about hardware questions.

Sorry, I’m not not able to help with that. Maybe there’s a limit how many old messages your server or client syncs?

I suppose it’s old drama by now. And I didn’t check if there’s new one in the meantime. As of now, both projects are active. Both have a userbase. Judging by the lasst commits, it’s still the case that Tuwunel is a one-man-show and Continuwuity is a community project.

I think whether you do closed source software is a personal choice. Based on considerations of your application. Like money, of if you want to rely on a company and how well they do their job, if it’s still gonna be around in 7 years. If you can customize it enough to suit your needs. Or you base the decision on ideology.

I’ve been using Yunohost on the NAS. And it’s simple, works well and is pretty reliable, I didn’t get any major issues for many years now. (And in general, community maintained open-source software has served me well. So that’s what I do.)

Downsides as a proficient Linux user are: You can’t just mess with the config while the automatic scripts also mess with the config. You need to learn how they’re set up and work around that. Hope software has a config.d or overrides directory and put your customizations there. Or something will get messed up eventually. And you can’t just change arbitrary things. The mailserver or SSO or reverse proxy and a few other components are tightly integrated and you’re never gonna be able to switch from postfix to stalwart or something like that. Or retrofit a more modern authentication solution. It is a limiting factor.
And YunoHost doesn’t do containers, so I doubt it’s what you’re looking for anyway.

I’m a bit split on the entire promise of turnkey selfhosting solutions. Some of them work really well. And they’re badly needed to enable regular people to emancipate themselves from big tech. Whether you as an expert want to use them is an entirely different question. I think that just depends on application. If you have a good setup, that might be better suited to your needs. And if done right might be very low maintenance as well. So switching to a turnkey solution would be extra work and it might not pay off. Or it does pay off, I think that really depends on the specifics.

Isn’t that a Nintendo Switch game? You’d need to install and run an emulator for that, like you did with Yuzu on Windows. I don’t think Yuzu is around anymore, but there are some sucessors, Eden and Citron? I’d install one of those. At least Eden has SteamOS mentioned on it’s homepage. You need to install it, though. The SteamOS or Linux version from their homepage, not copy the entire emulator over from Windows. After that you can transfer the game files and load them into the emulator. Any variant to copy files between computers should work. A windows network share, USB stick, microSD card, a cloud drive or filedrop/sync tool…

Isn’t that processor almost 15 years old? Nor sure what kind of price is alright for that. 250 seems a bit much but I don’t really know. I mean it’d probably work fine and 24GB is plenty of RAM for a few selfhosted services. I don’t think you need a graphics card for most services. Though Jellyfin can make use of the video encoder in it. Or use the encoder that comes with the iGPU in the processor. (Edit: Not sure in this case as it’s a very old generation.) And maybe the machine learning features in Immich can make use of the graphics card. Otherwise a GPU just wastes power in a server.

No idea about power consumption. I don’t know where you live, some countries have really cheap electricity. Some don’t and you maybe don’t want to run a random (old) gaming pc, because some waste a lot of power and some don’t and there is no good way to tell except measure it.

deleted by creator

Sure. Sadly I don’t have the proper tools around to do that. And in my case I wasn’t too sad. These devices had 100mbps ethernet and a slow wifi standard. Now they’re on e-waste and I got an upgrade to Gigabit ethernet and 5GHz wifi 😆

Not sure if I get your point. Abstraction is a concept used by IT people to deal with complexity. You’ll use Docker containers in order not to have 200 very specific problems and learn about the intricate details of all of them. Or use a turnkey solution because a working day has a finite amount of hours and you can just not care and have somebody else set the XY value of Postgres to 128 because that’s somehow needed for software M on python x.xx… Of course you’re then not going to learn about these things. It is not “bad”, though, in itself to abstract these issues away from you. Same for the other things I mentioned, networking, virtualization. Abstraction there allows to swap out complex things, do things once and in a clean way because it’s easy to miss things without abstraction and you always need to pay attention to a bazillion of specifics. Also helps with backups, deal with issues because things should break within confined layers, punch above one’s weight, security, do something once and roll it out several times…

I think what you want to avoid is poorly designed or written software. Or poorly done setups. Or not learn about important things. Abstraction is generally something you want, especially with complex things.

Maybe try something like YunoHost. That’s a web server Linux distribution. And it’s supposed to take care of the set up and come with somewhat safe/secure defaults. You’d need some kind of server, though. Or run it in a VM to isolate it from your home services. They have PeerTube, Lemmy, PieFed installable with a few clicks. (There are other projects as well, Yunohost isn’t the only option to help with the set up.)

But yes, some kind of isolation is probably nice with web services. Also from the home network, and from storage with personal data on it.

Yes, OpenWRT lasts way longer. Main thing that ends support is hardware requirements. My old devices with only a few megabytes of memory got dropped eventually. Not because of the chipset, a modern OpenWRT would just not fit any longer. I rarely see other reasons for them to discontinue updates.

You mean AMD or Intel? I can’t find any variant with an ARM processor. According to the internet, both the Intel and AMD version should work with Linux. My wife actually owns the Intel X13 Gen1. With Linux Mint Debian edition on it. Seems to work fine, she didn’t ever complain. Just be aware these are 5 year old devices. She paid 404€ for a refurbished one. We went with the 16GB RAM option, since that’s soldered and not upgradable. Also had an i7 processor at that price point.

If that traffic is going through an encrypted Wireguard tunnel, I don’t see a reason to encrypt it a second time. Judging by your description, it’s already encrypted on transport between the router and VPS. HTTPS would add nothing there. It will however add encryption within your DMZ, if you expect something nefarious going on within your DMZ.

Well known in the industry how you don’t assess programmers by lines of code. You kind of want them to be efficient and clean. Spend their day thinking and design clever solutions… Not pump out lots of unmaintainable low quality stuff. And have a million lines of that by tomorrow. But yeah, guess every aspect of this aligns well. You should be using Linux by now. Or at least do the switch in the near future.

Well, I tinkered around a bit with Speech Note which has a good amount of features and is easy to install as a Flatpak. I think it has an option to do this, but requires a bit off fiddling, an extra tool and permissions for the Flatpak. I didn’t find any software with a particularly good integration into the Desktop, though.

Also read about Blahst but didn’t try it yet. Maybe it’s an option.

Good question. maybe I was wrong, I’ve used Debian for so long, I’ll either just update it or use the advanced installer… It’s certainly on the installation media now. Not sure if it copies it over to the system. Would make sense, though if it did…