About the same here, though I have to say… Reading the “3 hours per day” part out loud still seems… Insane somehow.

In a similar vein, I’m currently staying at my mom’s house, and the internet is too shit to use my Jellyfin. As a result, I haven’t been watching any shows, and my day seems to be infinitely longer, like a million more activities fit in the sake 24 hours.

Yes, and I do werether the recipient also knows how to use it.

So, for like, 1% of my mails.

More like: paying someone to maintain the hardware.

Anyways.

Just FYI, your mails with a provider like Proton are not E2E encrypted unless you exclusively wrote with other Proton customers (in which case I assume they are. No idea). Otherwise it’s just encrypted at rest.

I dint really see the benefit over doing it completely yourself, not even offering metadata to a provider, and also having encryption at rest, while maintaining full compatibility with mail clients 🤔

Not a VPS.

We host most stuff at home, and then additionally some services at Hetzner on an (auctioned) root server. Bloody nice to get really good hardware for cheap, plus unlimited data with either 1 or 10Gbit synchronous network speed, a dedicated IPv4,…

Stuff like my mail server lives there because it HAS to be available, and doing it at home, and doing it well, is next to impossible.

I’m planning a nix hydra + cache server, which will probably also live on the Hetzner server, simply because it’ll have pretty intense jobs to run a lot of the time and I’m not a fan of having the noise of spun-up fans at home.

Both solutions have their place, is what I’m saying / agreeing.

• every VM with state backs up its state to the NAS once a day
• client devices rsync most of their home folder to the NAS once an hour
• at 3:15 in the morning, a Borg backup job starts pushing the days changes to a Hetzner storage box

Through borg, I have the Option to go back to any point in time with the backups. I will probably never need this, hence why it happens in this step, not on the rsync job to the NAS.

Things like movies and tv shows are not backed up, they are replaceable. All in all, about 2tb of documents, pictures, and VM state is backed up to Hetzner, out of the 16tb on the NAS.

Pick and choose your battles.

Yeah. I just put the media location on my nas, and that is being mirrored to hetzner.

Actually… Just tried it. I am on 2025.10, so newer than what was mentioned there. It still does not understand any better than from what I remember. Bummer.

But hey, at least the acknowledge that there’s the need for something between dumb pattern matching and an LLM.

Holy shit YES!

That article is from yesterday, and the relevant section is: https://www.home-assistant.io/blog/2025/10/22/voice-chapter-11/#improved-sentence-matching

Awesome to see improvements there. Thanks a lot for linking!

Oh wow, awesome!

Thank you for your sacrifice :D

While I don’t like it, it’s not hidden either:

https://bentopdf.com/privacy.html

There should definitely be an option to disable this for self-hosting, but if it’s just a counter for how often each tool is used by all users combined… Eh…

(Stirling also has something similar)

Why not open a PR to make it configurable? The maintainer is super active and friendly.

Thanks for the recommendation! That looks interesting indeed.

This entire topic is probably a sinkhole of complexity. It’s great to have somewhere to look for inspiration!

Yeah those are good points. Also noticed the CDN thing, it’s a bit annoying for a privacy-first project… But should be an easy fix 😄

Stirling’s backend is Java. So, yeah, heavy and slow sounds about right.

Ah, thanks for mentioning. Yep, they have a docker image; as mentioned, a nixpkg will be available soonTM; and frankly, you can just build / download the release artifacts and put them on any static host.

Please read the title of the post again. I do not want to use an LLM. Selfhosted is bad enough, but feeding my data to OpenAI is worse.

Yep, that’s the idea! This post basically boils down to “does this exist for HASS already, or do I need to implement it?” and the answer, unfortunately, seems to be the latter.

Thanks, had not heard of this before! From skimming the link, it seems that the integration with HASS mostly focuses on providing wyoming endpoints (STT, TTS, wakeword), right? (Un)fortunately, that’s the part that’s already working really well 😄

However, the idea of just writing a stand-alone application with Ollama-compatible endpoints, but not actually putting an LLM behind it is genius, I had not thought about that. That could really simplify stuff if I decide to write a custom intent handler. So, yeah, thanks for the link!!

Thanks for your input! The problem with the LLM approach for me is mostly that I have so many entities, HASS exposing them all (or even the subset of those I really, really want) is already big enough to slow everything to a crawl, and to get bad results from all models I’ve tried. I’ll give the model you mentioned another shot though.

However, I really don’t want to use an LLM for this. It seems brittle and like overkill at the same time. As you said, intent classification is a wee bit older than LLMs.

Unfortunately, the sentence template matching approach alone isn’t sufficient, because quite frequently, the STT is imperfect. With HomeAssistant, currently the intent “turn off all lights” is, for example, not understood if STT produces “turn off all light”. And sure, you can extend the template for that. But what about

• turn of all lights
• turn off wall lights
• turnip off all lights
• off all lights
• off all fights
• …

A human would go “huh? oh, sure, I’ll turn off all lights”. An LLM might as well. But a fuzzy matching / closest Levensthein distance approach should be more than sufficient for this, too.

Basically, I generally like the sentence template approach used by HASS, but it just needs that little bit of additional robustness against imperfections.