smiletolerantly
@smiletolerantly@awful.systems
- Comment on Don't fix the problem just change the parameters 4 days ago:
FWIW, I went to school in mid-2000. My sibling even later. They still taught it back then, and at least here, I am pretty sure they still do. (And why would they not, after all…)
- Comment on Don't fix the problem just change the parameters 4 days ago:
Literally noone I know in real life has any problem whatsoever reading analog clocks, no matter the “brain capacity”, neuro-typicality, state of drunkenness,… It is an extremely simple “skill”.
- Comment on Don't fix the problem just change the parameters 4 days ago:
Yeah, vibe of the time is a good description
- Comment on Don't fix the problem just change the parameters 4 days ago:
Because it’s not! Glad to help you clear that up.
- Comment on Don't fix the problem just change the parameters 4 days ago:
Disagree - it rarely matters to me if it’s 13:24:56 or 13:25:05, but I do find the instant and intuitive gauging of time deltas super useful (as in, how long it’s going to be on to the full hour / two quarter past / … ). Not saying you can’t get that info from a digital clock as well, of course you can; but the physically of analog clocks lends a good bit of intuition, I feel.
- Comment on Don't fix the problem just change the parameters 4 days ago:
I feel like I’m going insane reading these comments about how difficult it is to read analog clocks, how it needs too much understanding of maths, how it takes too long,…
Can someone please confirm: you just look, for a fraction of a second, at the clock face and know the time, right?
Learning to read the clock was like… A couple of lessons and some homework in the 2nd grade, and everyone got it.
- Comment on Intent recognition for HomeAssistant without an LLM? 1 week ago:
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.
- Comment on Intent recognition for HomeAssistant without an LLM? 1 week ago:
Holy shit YES!
That article is from yesterday, and the relevant section is: www.home-assistant.io/blog/…/voice-chapter-11/#im…
Awesome to see improvements there. Thanks a lot for linking!
- Comment on Apparently Palantir can access the content of social media accounts that were deleted a decade ago. 1 week ago:
well… at least he realizes that was bullshit…?
- Comment on Apparently Palantir can access the content of social media accounts that were deleted a decade ago. 1 week ago:
Depends - was the assault comment directed at assailants or victims?
- Comment on Intent recognition for HomeAssistant without an LLM? 1 week ago:
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!
- Comment on Try BentoPDF if you haven't / are unhappy with StirlingPDF 1 week ago:
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.
- Submitted 1 week ago to selfhosted@lemmy.world | 3 comments
- Comment on Intent recognition for HomeAssistant without an LLM? 1 week ago:
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.
- Comment on Intent recognition for HomeAssistant without an LLM? 1 week ago:
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.
- Comment on Intent recognition for HomeAssistant without an LLM? 1 week ago:
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!!
- Comment on Intent recognition for HomeAssistant without an LLM? 1 week ago:
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.
- Comment on Intent recognition for HomeAssistant without an LLM? 1 week ago:
Thanks for sharing your experience! I have actually mostly been testing with a good desk mic, and expect recognition to get worse with room mics… The hardware I bought are seeed ReSpeaker mic arrays, I am somewhat hopeful about them.
Adding a lot of alternative sentences does indeed help, at least to a certain degree. However, my issue is less with “it should recognize various different commands for the same action”, and more “if I mumble, misspeak, or add a swear word on my third attempt, it should still just pick the most likely intent”, and that’s what’s currently missing from the ecosystem, as far as I can tell.
Though I must conceit, copying your strategy might be a viable stop-gap solution to get rid of Alexa. I’ll have to pay around with it a bit more.
That all said, if you find a better intent matcher or another solution, please do report back as I am very interested in an easier solution that does not require me to think of all possible sentence ahead of time.
Roger.
- Comment on Intent recognition for HomeAssistant without an LLM? 1 week ago:
Never heard about willow before - is it this onw? Seems there is still recent activity in the repo - did the creator only recently pass away? Or did someone continue the project?
How’s your experience been with it?
And sure, will do!
- Submitted 1 week ago to selfhosted@lemmy.world | 19 comments
- Comment on I went to an anti-tech rally, where Gen Z dressed as gnomes and smashed iPhones. Here's what I learned. | Business Insider 1 week ago:
That is actually a really interesting approach to moderation, huh.
- Comment on New trailer for Star Trek: Voyager - Across the Unknown has a rather bald Borg crew member 1 week ago:
Amazing. She’s a great role model.
- Comment on New trailer for Star Trek: Voyager - Across the Unknown has a rather bald Borg crew member 1 week ago:
Ah! Finally! Something where I can look up at the sky and go:
What would Janeway do…?
without people looking at me like I’m a weirdo and that wasn’t what everyone does when faced with difficult questions.
- Comment on Loops Joins the Fediverse 2 weeks ago:
Yeah but why would I make myself dependent on Cloudflare.
- Comment on Loops Joins the Fediverse 2 weeks ago:
To be fair, you can simply selfhost MinIO.
- Comment on What's the best chat to self host? 2 weeks ago:
Lol, exact same situation here.
Quick question, did the migration to continuwuity break calls for you as well?
- Comment on I will be taking no followup questions. Thank you for your time 2 weeks ago:
What’s a tomato?
- Comment on Chat Control 2.0: what if Signal supports user-made Add-ons, one which encrypts all messages? Could that circumvent Chat Control 2.0? 3 weeks ago:
It’s not about prohibiting e2ee; it’s about enforcing client-side scanning.
Yes, that also breaks e2ee, but they can still go “nooo! E2ee is still perfectly fine and legal! You know, as ling as we get to read anything anyways”
And realistically, this will probably end up being implemented on an OS-level as well. So even using a self-hosted matrix server would not be immune.
Not to mention that both you and your conversation partner needs to take steps to evade this; one party is not sufficient.
- Comment on How realistic is it for me to learn Japanese so that I can experience Anime better? How long would it take to learn? Has anyone attempted this? 3 weeks ago:
Not OP, but: watched Haruhi a while back in English Dub, because there’s no German one. It was ok. The main POV character’s monotone voice was fitting in a fun way, but almost everyone else still has this fake energy to it (esp Haruhi and Asahina). Really hard to describe.
In general it’s baffling to me how fake English dubs sound, especially because there clearly are a lot of talented English voice actors doing the voices for cartoons etc.
I have the privilege of comparing the English Dubs to the German ones for a lot of shows, and it’s really interesting how, while the German VAs sound distinctly different from the Japanese originals, they sound natural and not overacted, while the English counterparts almost always sound like they were told “make it sound as fake as possible”.
- Comment on How realistic is it for me to learn Japanese so that I can experience Anime better? How long would it take to learn? Has anyone attempted this? 3 weeks ago:
What this lovely person said.
Also, and maybe I am alone here, but when I said learning to write, I really meant with a pen, on paper (or a tablet, I guess), not through an app where you need to smush your fingers in approximately the right place for the line to snap to the correct position; that does not really translate to being able to write.