标签：#smarthome

Voice and chat assistants for the home share a deceptively hard job: turning messy natural language into precise, structured commands. “Make it cozy in here” has to become a concrete intent plus the right slots — which device, which room, which value. Domux is an open model from iFlytek that focuses on exactly this problem: command understanding for smart-home assistants, framed as intent parsing and slot filling. What it is Task: smart-home command understanding — intent parsing + slot filling Base model: fine-tuned on google/gemma-4-E2B-it Modality: multimodal (image + text input) Target: edge / on-device deployment rather than large cloud models License: Gemma Why the compact base matters Building on the small Gemma-4-E2B base keeps Domux in a size class meant to run close to the device. For home assistants, that direction is attractive: keeping command understanding on-device can reduce round-trips and keep more interaction local, instead of routing every utterance to a large hosted model. Try it The model card is on Hugging Face (access is gated — you may need to log in and request access): 👉 https://huggingface.co/iFlytekOpenSource/Domux We're sharing open work like this because on-device, task-focused models are a practical piece of the foundation-model and serving story — not everything needs to be a giant cloud model.

Cellars sweat. On a warm humid day the air you let in is warmer than the cold concrete, and the moment it touches a cold surface it gives up its water. That's condensation, and over enough summers it's how a basement grows mould in the corners you never look at. I wanted a number that warned me before that happened. The catch I already knew going in: a raw relative-humidity reading isn't that number. This is Part 08 of the series. The hub was already in the house running other things, and the install was genuinely the easy 20 minutes. The part worth writing about is what came after the sensor showed up: turning its two raw readings into a dew-point spread, and deciding when that spread means "act" versus "ignore the spike." Why the H100, and why 868 MHz is the whole point The hub is a TP-Link Tapo H100 , a little smart hub that acts as a radio bridge for TP-Link's battery sensors — the T100 motion, the T110 contact, and the one I care about here, the T310 temperature/humidity sensor. Here's the load-bearing detail, and the reason I reached for this hub instead of a WiFi sensor: the Tapo sensors don't talk WiFi. They talk to the H100 over 868 MHz sub-GHz radio . That matters in a cellar more than anywhere else. Sub-GHz is long-range and punches through concrete and floors in a way 2.4 GHz WiFi simply doesn't. There's no WiFi worth having down in my cellar, and no interest in running a repeater into a damp room just to read a sensor. The T310 sits down there on a battery, the H100 upstairs where the network is, and the radio link between does the work. One H100 supports up to 64 sensors — for a house, more headroom than I'll ever use. Getting the sensors into Home Assistant The native TP-Link integration doesn't expose the H100's child sensors — it's built for the plugs and bulbs. The one that works is the community Tapo Controller integration (petretiandrea's TP-Link Tapo ), installed through HACS , the same custom-integration store I've leaned on throughout this ser

I didn't put Shelly Plus 1PM and 2PM behind my wall switches for the app or the cloud. I put them there because each one quietly does two jobs at once: it switches a circuit locally, and it measures exactly what that circuit draws. The moment I had them wired in, every light group and every roller shutter in the flat became both controllable AND measurable in Home Assistant — with no separate energy meter and, deliberately, no cloud in the loop. This is the part most Shelly write-ups skip: the metering isn't a bonus, it's the reason to use them. This is how I set them up, kept them fully local, wired the physical switches so they still work like switches, and pulled per-channel energy into the HA Energy Dashboard. Why the Plus series for in-wall use The Plus 1PM and 2PM are Gen2 devices. That matters: they speak a local RPC API over HTTP and WebSocket, they're supported by the native Home Assistant Shelly integration out of the box, and nothing about them needs the internet once they're on your Wi-Fi. The 1PM is a single relay with power metering — I use it behind light circuits. The 2PM has two relays with metering and a dedicated cover/roller-shutter mode — that's what sits behind my shutters. The catch you plan for up front: both need a neutral wire in the switch box. Older installations often don't have one at the switch, so check before you order. If the neutral is there, the rest is easy. The first thing I change: kill the cloud, pin the IP Straight out of the box I do two things in the Shelly web UI (just browse to the device's IP). First, I disable Bluetooth and Cloud under Settings — I don't want the device reaching out to anything, and Home Assistant talks to it directly on the LAN. Second, I give it a fixed address via a DHCP reservation on my router, so the native integration never loses it after a reboot. That's the whole hardening story. No firmware flashing, no third-party firmware — stock Shelly firmware with the cloud switched off is already a fully

A note before the post: the mistake in the first section is genuinely mine. It cost me an evening of forking conditions in Home Assistant before I accepted the fix didn't belong in Home Assistant at all. I've left it in rather than writing around it, because it's the part I'd have wanted to read first. The first time motion-controlled lighting actually worked in my place, it didn't feel clever. It felt obvious — I walked into a dark hallway and the light was already on by the time I'd registered it was dark. That's the bar. Not smart , just attentive. Getting there with seven KNX motion sensors took me less code than I expected and one insight I wish I'd had on day one. This is Part 05 of the series. The earlier parts cover the boring-but-load-bearing groundwork: running Home Assistant in Docker and wiring up HACS . Here I'm assuming HA is up, talking KNX, and you just want the lights to behave. One sensor, two jobs, two addresses Here's the mistake I made, and it's the whole reason this post exists. KNX exposes each motion sensor to Home Assistant as a binary_sensor with device_class: motion , fed by a KNX group-address state object you configure in knx.yaml with a state_address per sensor. Simple enough. So I wired all seven sensors with one group address each and pointed both the lighting automation and the presence logic at the same signal. That works right up until you want the two to behave differently. A light should react to the smallest twitch, instantly, generously. Presence and security want the opposite: a debounce, a grace window, some scepticism before they commit. When both ride the same group address, every change you make to one quietly deforms the other. I spent an evening forking conditions in Home Assistant trying to make one signal mean two things. The fix isn't in Home Assistant at all. It's in ETS: give each physical PIR a second group address . One drives comfort lighting, the other feeds presence and the alarm path. I use a flat convention —

I’ve spent thousands of dollars over the years on smart home gear. Like many tech enthusiasts, I started with the usual suspects: smart bulbs, plugs, sensors, voice assistants, and eventually more “advanced” hubs. Every time, the marketing promised intelligence. Every time, I got glorified timers and motion detectors wearing a fancy label. After multiple attempts, I’ve reached the same conclusion many others quietly reach: most “smart home” products are not smart. They are automated, and there’s a massive difference. What “Smart” Actually Means A genuinely smart home system should do three things well: Understand context. Not just that a door opened or motion was detected, but why and what it means right now. Integrate devices meaningfully. Devices shouldn’t just talk to each other; they should share rich, semantic information so the system can reason across them. Be predictive and proactive. It should anticipate needs based on patterns, current state, and human behavior, instead of waiting for a trigger. Current systems almost never do any of these at a level that feels intelligent. The Core Problems (From Someone Who Actually Tried) Take a simple example: the dishwasher. A basic automation might detect the door was opened and then closed, then start the cycle. But it has zero idea whether: Dishes were actually loaded Someone was just checking if the cycle finished More dishes are coming in 30 seconds The person is about to run a quick rinse first The same gap appears everywhere: Lighting at night. The system doesn’t know if you just got up to use the bathroom, you’re wide awake working, or there was an emergency. It just sees “motion after 11 p.m.” and either blasts you with light or leaves you in the dark. Multi-person households. One person’s preference for dim evening lighting conflicts with another person’s need for bright light. Guests have no idea how anything works and accidentally trigger routines. “I’m just doing a quick house tour” vs. actual activity. T