Home Assistant is flexible enough to run on a wide range of hardware, but in my experience, nothing matches the Raspberry Pi 4B or 5. It isn’t just about convenience or power efficiency, though those matter too. The Pi unlocks certain features, integrations, and community support that you’ll struggle to find anywhere else. For anyone serious about automating their home, a dedicated Raspberry Pi is the best foundation you can give your setup.

As a journalist, I often have to sit through hours of recorded interview audio, trying to catch nuance, but mostly just transcribing it for important quotes. Sure, there are services like Otter, but outsourcing transcription can be expensive and risky, often both. File privacy is a legitimate concern, and accuracy isn't quite the best, anyway, for the amount of money these services charge. I've been looking for a tool that I can control, one that runs locally, and that can handle transcription. If it came with a few extra features, that would be even better. And that's how I came across FileWizard. This self-hosted, open-source app is astonishingly capable at wrangling files of all sorts, and just as easy to get up and running. Here's why FileWizard is a great choice not just for journalists, but for anyone in search of an easy transcription tool.

The Nintendo Switch 2 has already had an eventful launch year for fans that managed to pick up the console. New games hitting the market have kept Nintendo fans busy with fun adventures, including a few major games from classic Nintendo mascots like Donkey Kong and Mario. But as 2025 begins to wind down over the course of the fourth quarter of the year, the holiday season will bring more games to Nintendo's latest console for everyone. And some titles are going to make a bigger splash than others to end off the year on a high note.

SSD storage is some of the best you can buy right now, mainly because it's going to offer some of the best performance you can get no matter the form factor. Those looking for a simple and quick way to get their data from one device to another will want to look towards a portable SSD.

AI and Obsidian go together like peanut butter and jelly, but not all tools are created equal. I tested a local LLM with Obsidian, but I found it didn't perform the way I would have liked. When the chance arose to integrate my vault with Claude, an AI I use regularly for clerical tasks, I had to try it out. What I found completely revolutionized my workflow and took my Obsidian experience to the next level. I already loved the tool for its ability to aggregate my notes and esoteric interests in one place, but Claude allows me to combine that data in ways I wouldn't expect, and links Obsidian with any service connected to Claude.

One of my favorite parts about GTA 5 is the radio stations. You spend a lot of time in a vehicle, so it makes sense that Rockstar put a ton of time and effort into crafting some really memorable stations with some equally banging tunes.

Not everyone manages to get their hands on a brand-new console weeks before it's released, and even fewer decide to take the time to sit down and review the product for everyone else to see. However, when it does happen, it's pretty special, not only because we get to see someone get incredibly lucky, but also because it gives us an early and unbiased insight as to what to expect when the product finally arrives in our hands.

When I first started using Obsidian, it felt powerful, but also raw. I had the foundation of a fantastic networked note-taking app, yet something was missing from my personal workflow. I spent hours tweaking settings and tried to force the core features to do things they weren’t designed for. The real turning point didn’t come from a core update; it came from the incredible community.

Developers have gotten really creative with roguelikes. Ever since Slay the Spire made a huge splash by popularizing the card-based roguelike system, other developers have taken the opportunity to put their own spin on the genre.

The late English writer Douglas Adams is best known as the author of the 1979 book The Hitchhiker’s Guide to the Galaxy. But there is much more to Adams than what is written in his Wikipedia entry. Whether or not you need to know that his birth sign is Pisces or that libraries worldwide store his books under the same string of numbers — 13230702 — you can if you head to an overlooked corner of the Wikimedia Foundation called Wikidata. There, images, text, keywords, and other information related to Adams are stored both in a webpage and, for the robots among us, in formats designed for machines like JSON. Now, Wikidata is getting a new AI-friendly database that makes it easier for large language models to ingest the information. The database comes from the Wikipedia Embedding Project out of the German chapter of the Wikimedia Foundation, Wikimedia Deutschland, which oversees Wikidata. The Berlin-based team spent the past year using a large language model to turn the 19 million entries within Wikidata from clunkily structured data into vectors that capture the context and meaning around the Wikidata entry. In this vectorized format, information is best imagined like a graph with dots and interconnected lines — Adams would be connected to “human” as well as the titles of his books, Lydia Pintscher, Wikidata portfolio lead, told The Verge.While the front-end user experience will remain the same — no, Wikipedia is not becoming a chatbot, the project leaders say — the back end will become easier for AI developers to access when building, for example, their own chatbots using the data. The goal of the project is to level the playing field for AI developers outside the monied core of Big Tech, Pintscher said. Companies like OpenAI and Anthropic have the resources to vectorize Wikidata, just like Pintscher and her team did. It’s the smaller outfits that most benefit from the new access to curated data stored in the vaults of Wikidata. “Really, for me, it’s about giving them that edge up and to at least give them a chance, right?” Pintscher said. She points to Govdirectory as an example project that harnessed Wikidata’s vast data curated by volunteers for good. The platform allows users to find the social media handles and emails for public officials across the world. Most AI chatbots prioritize popular words and topics across the internet. In addition to giving Little Tech a leg up, the team hopes that easier access to Wikidata will result in AI systems that better reflect niche topics not widely represented across the internet, Pintscher said. This could be a better way to get information into ChatGPT, for instance, than “generating a ton of content and then waiting for the next time for ChatGPT to retrain, and maybe, or maybe not, taking into account what you contributed,” Pintscher said. In practice, the vectors will allow AI systems to better access the context around information in addition to the information itself, Philippe Saadé, Wikidata AI project manager, told The Verge. The team used a model from AI company Jina AI to turn Wikidata’s structured data, captured through September 18th, 2024, into vectors. IBM company DataStax currently provides the infrastructure to store the vector database to the project for free. The team is waiting for feedback from developers who use the database before updating it with information added over the last year. While the current database does not include entirely new information added in the last year, Saadé says small edits or tweaks to existing Wikidata will not diminish the database’s usefulness. “At the end of the day, the vector that we’re computing is like a general idea of an item, so if some small edit has been made on Wikidata, it’s not going to be super relevant,” he said.

Roland TR-1000 Rhythm Creator Roland has largely decided sit out the analog synth renaissance of the last decade. But it’s (belatedly) making amends for that error with the launch of the TR-1000 Rhythm Generator. Unlike previous instruments that tried to cash in on Roland’s long legacy through digital technologies like its Zen-Core engine, Analogue Circuit Behavior (ACB) modeling, and Roland Cloud subscription service, the TR-1000 embraces the true analog sound of iconic Roland machines like the TR-808 and TR-909, Unlike other TR instruments from Roland over the last several years that simply invoke the branding, the TR-1000 actually recreates 16 analog sound circuits from the 808 and the 909 while adding plenty of modern sound-shaping and sequencing conveniences. It pairs those authentic analog voices with digital ACB sound, FM percussion, PCM samples, and more. The TR-1000 even has extensive sampling tools for time stretching loops and chopping up drum breaks. In short, if there is a way to make a percussive sound, Roland has crammed it into this very serious looking (and very expensive) gray box.Roland decided to lean even further into the analog world here, by including a stereo analog filter and drive section for adding grit and creating buildups. There are also a ton of knobs, buttons, and faders on the front that suggest made the TR-1000 will require less diving through arcane menus than your typical modern Roland instrument. While it does have some performance-friendly features, its substantial bulk seems best suited to a desk in a recording studio.All of these luxuries will set you back quite a bit — $2,699.99 to be exact, which means the TR-1000 isn’t for casual hobbyists. Still, that’s about half of what a functioning TR-808 or TR-909 will cost you on the vintage market so, maybe the TR-1000 is actually a bargain?

This might sound absurd coming from someone who works in tech, but I was never really a "mouse person." For years, I flat-out refused to use one, mostly because the only mice I'd ever touched were the flimsy, soulless ones you’d find sitting in computer labs.