Science fiction has regularly portrayed smart computer assistants in a fanciful way. HAL from 2001: A Space Odyssey and J.A.R.V.I.S. from the contemporary Iron Man films are both great examples. They’re erudite, wise, and capable of doing just about any reasonable task that is asked of them, short of opening the pod bay doors.
Cut back to reality, and you’ll only be disappointed at how useless most voice assistants are. It’s been twelve long years since Siri burst onto the scene, with Alexa and Google Assistant following years later. Despite years on the market, their capabilities remain limited and uninspiring. It’s time for voice assistants to level up.
A powerful robot awaiting for a verbal command to crush its foes might sound like something from a science fiction film, but now it’s a permanent fixture of the [Making Stuff] garage. (Video, embedded below.) Thankfully this robot’s sworn enemy are aluminum cans, and the person controlling it with their voice isn’t a maniacal scientist, just a guy who’s serious about recycling. Well, we hope so anyway.
The star of the show is a heavy duty wall-mounted can crusher that [Making Stuff] built from some scrap steel and a pneumatic cylinder hooked up to the garage’s compressed air system. A solenoid operated valve allows an Arduino with attached ESP-01 to extend the cylinder whenever the appropriate command comes over the network. In this case, the goal was to tie the crusher into Google Assistant so a can would get smallified whenever one of Google’s listening devices heard the trigger phrase.
Note the ejector air line.
Obviously, those who’d rather keep Big Data out of their recycling bin don’t have to go down the same path. But that being said, having to give a specific voice command to activate the machine does provide a certain level of operational safety. At least compared to trusting some eBay sensor to tell the difference between an aluminum can and a fleshy appendage.
After crushing a few cans with his new toy, [Making Stuff] noticed a fairly troubling flaw in the design. Each time a can was crushed he had to reach into the maw of the machine to push its little flattened carcass out of the way. In other words, he was one bad line of code away from having one good hand.
The solution ended up being a new hose that runs from the exhaust port of the valve to the crushing chamber: once the cylinder retracts, the air exiting the valve pushes the crushed can out the rear of the machine and into a waiting pail underneath. Very slick.
Getting a child’s attention can be difficult at the best of times. Add deafness into the picture, and it’s harder again. [Jake]’s daughter recently had to go without her cochlear implants, raising this issue. Naturally, he whipped up some hardware to solve the problem.
[Jake]’s solution was to devise a vibrating wristband that could be used to get his daughter’s attention. An Adafruit Trinket M0 is used to vibrate a pager motor, using a DRV2605 motor driver. This is paired with a Tile Bluetooth device, allowing the unit to interface with Google Assistant. This allows [Jake] to get his daughter’s attention with a simple voice command to a smartphone, tablet or smart speaker.
While [Jake]’s daughter will regain her cochlear implants soon, they do have limitations as far as hearing distant sounds and working in high-noise environments. It’s likely that this little gadget will prove useful well into the future, and could serve others well, too. Wearable notification devices are growing more popular; this OLED ring is a particularly good example. Video after the break.
Had too much self-quarantine? [Sharathnaik] had, so he decided to build a robot companion named Ewon. Using a Raspberry Pi, Ewon isn’t a robot that moves around, but rather an expressive Google assistant. Using some servo-driven ears and a display, Ewon reacts to you based on keywords you use in your queries. For example, it might perk up and smile at the mention of ice cream. Or look unhappy if you mention sadness.
The project is simple because of the Google Assistant API. However, we liked the 3D printed body and some of the additional features the robot adds.
Careful, the walls have ears. Or more specifically, the smart speaker on the table has ears, as does the phone in your pocket, the fitness band on your wrist, possibly the TV, the fridge, the toaster, and maybe even the toilet. Oh, and your car is listening to you too. Probably.
How does one fight this profusion of listening devices? Perhaps this wearable smart device audio jammer will do the trick. The idea is that the MEMS microphones that surround us are all vulnerable to jamming by ultrasonic waves, due to the fact that they have a non-linear response to ultrasonic signals. The upshot of that is when a MEMS hears ultrasound, it creates a broadband signal in the audible part of the spectrum. That creates a staticky noise that effectively drowns out any other sounds the microphone might be picking up.
By why a wearable? Granted, [Yuxin Chin] and colleagues from the University of Chicago have perhaps stretched the definition of that term a tad with their prototype, but it turns out that moving the jammer around does a better job of blocking sounds than a static jammer does. The bracelet jammer is studded with ultrasonic transducers that emit overlapping fields and result in zones of constructive and destructive interference; the wearer’s movements vary the location of the dead spots that result, improving jamming efficacy. Their paper (PDF link) goes into deeper detail, and a GitHub repository has everything you need to roll your own.
We saw something a bit like this before, but that build used white noise for masking, and was affixed to the smart speaker. We’re intrigued by a wearable, especially since they’ve shown it to be effective under clothing. And the effect of ultrasound on MEMS microphones is really interesting.
The modern hacker wields a number of tools that operate on the principle of heating things up to extremely high temperatures, so a smoke alarm is really a must-have piece of equipment. But in an era where it seems everything is getting smarter, some might wonder if even our safety gear could benefit from joining the Internet of Things. Interested in taking a crack at improving the classic smoke alarm, [Vivek Gupta] grabbed a NodeMCU and started writing some code.
Now before you jump down to the comments and start smashing that keyboard, let’s make our position on this abundantly clear. Do not try to build your own smoke alarm. Seriously. It takes a special kind of fool to trust their home and potentially their life to a $5 development board and some Arduino source code they copied and pasted from the Internet. That said, as a purely academic exercise it’s certainly worth examining how modern Internet-enabled microcontrollers can be used to add useful features to even the most mundane of household devices.
In this case, [Vivek] is experimenting with the idea of a smoke alarm that can be silenced through your home automation system in the event of a false alarm. He’s using Google Assistant and IFTTT, but the code could be adapted to whatever method you’re using internally to get all your gadgets on the same virtual page. On the hardware side of things, the test system is simply a NodeMCU connected to a buzzer and a MQ2 gas sensor.
So how does it work? If the detector goes off while [Vivek] is cooking, he can tell Google Assistant that he’s cooking and it’s a false alarm. That silences the buzzer, but not before the system responds with a message questioning his skills in the kitchen. It’s a simple quality of life improvement and it’s certainly not hard to imagine how the idea could be expanded upon to notify you of a possible situation even when you’re out of the home.
Adding a camera to Google’s AIY Voice Kit makes for a versatile object identification system.
The device is built around Google’s AIY Voice Kit, which consists of a Raspberry Pi with some additional hardware and software to enable it to process voice queries. [Liz] combined this with a Raspberry Pi camera and the Google Cloud Vision API. This allows WhatIsThat to respond to users asking questions by taking a photo, and then identifying what it sees in the frame.
It may seem like a frivolous project to those with working vision, but there is serious potential for this technology in the accessibility space. The device can not only describe things like animals or other objects, it can also read text aloud and even identify logos. The ability of the software to go beyond is impressive – a video demonstration shows the AI correctly identifying a Boston Terrier, and attributing a quote to Albert Einstein.
Artificial intelligence has made a huge difference to the viability of voice recognition – because it’s one thing to understand the words, and another to understand what they mean when strung together. Video after the break.
