Make 3D Scenes With A Holodeck-Like Voice Interface

April 29, 2024 by Donald Papp No comments

The voice interface for the holodeck in Star Trek had users create objects by saying things like “create a table” and “now make it a metal table” and so forth, all with immediate feedback. This kind of interface may have been pure fantasy at the time of airing, but with the advent of AI and LLMs (large language models) this kind of natural language interface is coming together almost by itself.

A fun demonstration of that is [Dominic Pajak]’s demo project called VoxelAstra. This is a WebXR demo that works both in the Meta Quest 3 VR headset (just go to the demo page in the headset’s web browser) as well as on desktop.

The catch is that since the program uses OpenAI APIs on the back end, one must provide a working OpenAI API key. Otherwise, the demo won’t be able to do anything. Providing one’s API key to someone’s web page isn’t terribly good security practice, but there’s also the option of running the demo locally.

Either way, once the demo is up and running the user simply tells the system what to create. Just keep it simple. It’s a fun and educational demo more than anything and will try to do its work with primitive shapes like spheres, cubes, and cylinders. “Build a snowman” is suggested as a good starting point.

Intrigued by what you see and getting ideas of your own? WebXR can be a great way to give those ideas some life and looking at how someone else did something similar is a fine way to begin. Check out another of [Dominic]’s WebXR projects: a simulated BBC Micro, in VR.

VAR Is Ruining Football, And Tech Is Ruining Sport

April 29, 2024 by Lewin Day 16 Comments

Another week in football, another VAR controversy to fill the column inches and rile up the fans. If you missed it, Coventry scored a last-minute winner in extra time in a crucial match—an FA Cup semi-final. Only, oh wait—computer says no. VAR ruled Haji Wright was offside, and the goal was disallowed. Coventry fans screamed that the system got it wrong, but no matter. Man United went on to win and dreams were forever dashed.

Systems like the Video Assistant Referee were brought in to make sport fairer, with the aim that they would improve the product and leave fans and competitors better off. And yet, years later, with all this technology, we find ourselves up in arms more than ever.

It’s my sincere belief that technology is killing sport, and the old ways were better. Here’s why. Continue reading “VAR Is Ruining Football, And Tech Is Ruining Sport” →

Farewell MFJ

April 29, 2024 by Al Williams 13 Comments

We were sad to hear that after 52 years in operation, iconic ham radio supplier MFJ will close next month. On the one hand, it is hard not to hear such news and think that it is another sign that ham radio isn’t in a healthy space. After all, in an ideal world, [Martin Jue] — the well-known founder of MFJ — would have found an anxious buyer. Not only is the MFJ line of ham radio gear well regarded, but [Martin] had bought other ham radio-related companies over the years, such as Ameritron, Hygain, Cushcraft, Mirage, and Vectronics. Now, they will all be gone, too.

However, on a deeper reflection, maybe we shouldn’t see it as another nail in ham radio’s coffin. It is this way in every industry. There was a time when it was hard to imagine ham radio without, say, Heathkit. Yet they left, and the hobby continued. We could name a slew of other iconic companies that had their day: Eico, Hammarlund, Hallicrafters, and more. They live on at hamfests, their product lines are frozen in time, and we’re sure we’ll see a used market for MFJ gear well into the next century.

Continue reading “Farewell MFJ” →

DIY Passive Radar System Verifies ADS-B Transmissions

April 29, 2024 by Bryan Cockfield No comments

Like most waves in the electromagnetic spectrum, radio waves tend to bounce off of various objects. This can be frustrating to anyone trying to use something like a GMRS or LoRa radio in a dense city, for example, but these reflections can also be exploited for productive use as well, most famously by radar. Radar has plenty of applications such as weather forecasting and various military uses. With some software-defined radio tools, it’s also possible to use radar for tracking aircraft in real-time at home like this DIY radar system.

Unlike active radar systems which use a specific radio source to look for reflections, this system is a passive radar system that uses radio waves already present in the environment to track objects. A reference antenna is used to listen to the target frequency, and in this installation, a nine-element Yagi antenna is configured to listen for reflections. The radio waves that each antenna hears are sent through a computer program that compares the two to identify the reflections of the reference radio signal heard by the Yagi.

Even though a system like this doesn’t include any high-powered active elements, it still takes a considerable chunk of computing resources and some skill to identify the data presented by the software. [Nathan] aka [30hours] gives a fairly thorough overview of the system which can even recognize helicopters from other types of aircraft, and also uses the ADS-B monitoring system as a sanity check. Radar can be used to monitor other vehicles as well, like this 24 GHz radar module found in some modern passenger vehicles.

Continue reading “DIY Passive Radar System Verifies ADS-B Transmissions” →

AI Can Now Compress Text

April 28, 2024 by Jenny List 29 Comments

There are many claims in the air about the capabilities of AI systems, as the technology continues to ascend the dizzy heights of the hype cycle. Some of them are true, others stretch definitions a little, while yet more cross the line into the definitely bogus. [J] has one that is backed up by real code though, a compression scheme for text using an AI, and while there may be limitations in its approach, it demonstrates an interesting feature of large language models.

The compression works by assuming that for a sufficiently large model, it’s likely that many source texts will exist somewhere in the training. Using llama.cpp it’s possible to extract the tokenization information of a piece of text contained in its training data and store that as the compressed output. The decompressor can then use that tokenization data as a series of keys to reassemble the original from its training. We’re not AI experts but we are guessing that a source text which has little in common with any training text would fare badly, and we expect that the same model would have to be used on both compression and decompression. It remains a worthy technique though, and no doubt because it has AI pixie dust, somewhere there’s a hype-blinded venture capitalist who would pay millions for it. What a world we live in!

Oddly this isn’t the first time we’ve looked at AI text compression.

Hack In Style With This Fallout Cyberdeck

April 28, 2024 by Julian Scheffers 10 Comments

There’s always an appeal to a cool-looking computer case or cyberdeck – and with authentic-looking Vault-Tec style, [Eric B] and [kc9psw]’s fallout-themed cyberdeck is no exception.

The case looks like it came straight out of one of the Fallout games and acts the part: while (obviously) not capable of withstanding a direct nuclear bomb impact, it can protect the sensitive electronics inside from the electromagnetic pulse and shockwave that follows – if you keep it closed.

And it’s not just the case that’s cool: This cyberdeck is packed full of goodies like long-range radios, SDRs, ADSB receivers, a Teensy 4.1, and dual Raspberry Pis. But that’s just the hardware! It also comes with gigabytes upon gigabytes of Wikipedia, Wikihow, TED talks, and other information/entertainment, for the less eventful days in the wastelands.

If you, too, would like to have one, fret not! The parts list and design files are public, even though some assembly is required.

Hackaday Links: April 28, 2024

April 28, 2024 by Dan Maloney 19 Comments

Well, it’s official — AI is ruining everything. That’s not exactly news, but learning that LLMs are apparently being used to write scientific papers is a bit alarming, and Andrew Gray, a librarian at University College London, has the receipts. He looked at a cross-section of scholarly papers from 2023 in search of certain words known to show up more often in LLM-generated text, like “commendable”, “intricate”, or “meticulous”. Most of the words seem to have a generally positive tone and feel a little fancier than everyday speech; one rarely uses “lucidly” or “noteworthy” unless you’re trying to sound smart, after all. He found increases in the frequency of appearance of these and other keywords in 2023 compared to 2022, when ChatGPT wasn’t widely available.

Continue reading “Hackaday Links: April 28, 2024” →