

This is why I started using speech to phrase instead of speech to text. Sure it’s technically less flexible, but I only need a couple of voice commands for occasional tasks that are impractical to automate with sensors and it’s blazing fast


This is why I started using speech to phrase instead of speech to text. Sure it’s technically less flexible, but I only need a couple of voice commands for occasional tasks that are impractical to automate with sensors and it’s blazing fast


Last time I checked it only runs well on M1 devices, with M2 being somewhat usable. M3 though M5 are a complete no-go unfortunately :(


Is this really causation though? Could it not just be that kids that spend less time looking at screens are less likely to be short-sighted AND more likely to spend time outside?
How the artist’s name is supposed to be parsed:
Analog Nowhere
How my brain parsed it:
Anal OG now here
I have almost 50 containers running on a £75 used mini pc from Ebay, chucked in the bottom of a cupboard. Anyone that tells you you need a giant server rack to self host is a gatekeeping moron


Does anyone know of an RSS feed for the newsletter? I’ve been trying to use RSS for more stuff recently
The best way I found to combat spam was to change my email address. Was it a pain? Yes. Was it worth it? Absolutely yes.
Suddenly I can just give my email address out to services I trust, and I set up a temporary alias for anything I don’t (although I tend to avoid those these days anyway).


Yes, that seems to be the aim here. Also note that this is being provided by a community member and is in no way affiliated with the HA team


How the hell are you supposed to pronounce this?
“Great”?
“Greet”??
“Grit”???
“Grite”???
The long and short of it is this:


Get rid of that thing immediately, it’s recording everything you say and sending it to Palantir


How does this compare to Jellyseer?
Ikea recently released some matter over thread smart buttons that are ~$3 each


NixOS. And be really annoying about declarative software environments


I disagree that Nix is a solution in search of a problem, in fact it solves arguably the two biggest problems in software deployment: dependency hell and reproducibility (i.e. the “It works on my machine” problem)
Every package gets access to the exact version of all the dependencies it needs (without needless replication like Flatpaks would have) and sharing a flake to another machine means you can replicate that exact setup and guarantee it will be exactly the same
Containers try to solve the same problems, and succeed to a somewhat decent extent, although with some overhead of course.
I’m not trying to criticize you or your setup at all, if Debian alone works for you, that’s fine. The beauty of open source and self hosting is that we can use whatever tools we want, however we want. I do though think it’s good practice to be aware of what alternatives are out there should our needs change, or should our tools change to no longer align with our needs.


Small setups can very easily turn into large setups without you noticing.
The only bare-metal setup I’d trust to be scaleable is Nix flakes (which I’m actually very interested in migrating to at some point)
It’s hard to recommend any Chromium-based browser in 2026 as Chromium no longer supports Manifest v2 extensions, breaking support for essential content blockers like uBlock Origin.
Even Brave, which has a built-in adblocker not based on extensions, is loaded with AI and Crypto bs, and as far as I’m aware, the adblocker portion of its code is closed source. This isn’t even mentioning the horrific political opinions expressed by the CEO of that company.
I keep a copy of raw ungoogled Chromium around for webUSB firmware flashing, but that’s it.
If I were you, I would try and figure out what is causing firefox-based browsers to run slowly - i.e. toggling on/off hardware acceleration in browser settings and see if that makes a difference.
This is a single unofficial client for the mentioned software. LibreTranslate itself is fantastic.


Doing this is generally a bad idea, because audio exported from YouTube is pretty poor quality, and music videos often have bits of talking or silence that make sense in context of the video but aren’t part of the actual song (designed to prevent exactly this). There was a cli tool I used last year that could download music from Spotify directly.
Edit: The tool I was talking about is Zotify
Make sure to set the --download-quality flag to very_high if you have premium to ensure it downloads in max quality
If you have long playlists (more than a few hundred songs), you should also use the --skip-previously-downloaded and --song-archive flags as per the docs to make sure you can start again from where you left off, as Spotify will start to rate-limit your connection and downloads will fail (if this happens, just kill the tool, wait a few minutes and start again)
It works with wildcards, but only if they’re pre-defined in some yaml. It’s great for i.e. defining areas a robot vacuum should clean or days of the week for a weather forcast, but might be a bit of a pain for shopping list items unless you only ever use your voice for a select few.
As for what I’m using, it’s just called speech-to-phrase and I believe it’s now the default when setting up a new Voice PE (although that might just be on lower powered hardware). If it wasn’t set up for you, it can be installed as an add-on and then configured as a Wyoming protocol service.