In European medieval folklore, a practitioner of magic may call for assistance from a familiar spirit who takes an animal form disguise. [Alex Glow] is our modern-day Merlin who invoked the magical incantations of 3D printing, Arduino, and Raspberry Pi to summon her familiar Archimedes: The AI Robot Owl.
The key attraction in this build is Google’s AIY Vision kit. Specifically the vision processing unit that tremendously accelerates image classification tasks running on an attached Raspberry Pi Zero W. It no longer consumes several seconds to analyze each image, classification can now run several times per second, all performed locally. No connection to Google cloud required. (See our earlier coverage for more technical details.) The default demo application of a Google AIY Vision kit is a “joy detector” that looks for faces and attempts to determine if a face is happy or sad. We’ve previously seen this functionality mounted on a robot dog.
[Alex] aimed to go beyond the default app (and default box) to create Archimedes, who was to reward happy people with a sticker. As a moving robotic owl, Archimedes had far more crowd appeal than the vision kit’s default cardboard box. All the kit components have been integrated into Archimedes’ head. One eye is the expected Pi camera, the other eye is actually the kit’s piezo buzzer. The vision kit’s LED-illuminated button now tops the dapper owl’s hat.
Archimedes was created to join in Google’s promotion efforts. Their presence at this Maker Faire consisted of two tents: one introductory “Learn to Solder” tent where people can create a blinky LED badge, and the other tent is focused on their line of AIY kits like this vision kit. Filled with demos of what the kits can do aside from really cool robot owls.
Hopefully these promotional efforts helped many AIY kits find new homes in the hands of creative makers. It’s pretty exciting that such a powerful and inexpensive neural net processor is now widely available, and we look forward to many more AI-powered hacks to come.
With Google’s near-monopoly on the internet, it can be difficult to get around in cyberspace without encountering at least some aspect of this monolithic, data-gathering giant. It usually takes a concerted effort, but it is technically possible to do. While [Mat] is still using some Google products, he has at least figured out a way to get Google Home to work with location data without actually sharing that data with Google, which is a step in the right direction.
[Mat]’s goal was to use Google’s location sharing features through Google Home, but without the creepiness factor of Google knowing everything about his life, and also without the hassle of having to use Google Maps. He’s using a few things to pull this off, including a NodeRED server running on a Raspberry Pi Zero, a free account from If This Then That (IFTTT), Tasker with AutoRemote plugin, and the Google Maps API key. With all of that put together, and some configuration of IFTTT he can ask his Google assistant (or Google Home) for location data, all without sharing that data with Google.
This project is a great implementation of Google’s tools and a powerful use of IFTTT. And, as a bonus, it gets around some of the creepiness factor that Google tends to incorporate in their quest to know all the data.
Google Authenticator is a particularly popular smartphone application that can be used as a token for many two factor authentication (2FA) systems by generating a time-based one time password (referred to as TOTP). With Google Authenticator, the combination of your user name and password along with the single-use code generated by the application allows you to securely authenticate yourself in a way that would be difficult for an attacker to replicate.
That sounds great, but what if you don’t have a smartphone? That’s the situation that [Lady Ada] recently found herself in, and rather than going the easy route and buying a hardware 2FA token that’s compatible with Google Authenticator, she decided to build one herself based on the ESP8266. With the hardware and source documented on her site, the makings of an open source Google Authenticator hardware token are available for anyone who’s interested.
For the hardware, all you need is the ESP8266 and a display. Naturally [Lady Ada] uses her own particular spin on both devices which you can purchase if you want to create an identical device, but the concept will work the same on the generic hardware you’ve probably already got in the parts bin. Software wise, the code is written in CircuitPython, a derivative of MicroPython, which aims to make microcontroller development easier. If you haven’t tried MicroPython before, grab an ESP and give this a roll.
Conceptually, TOTP is relatively simple. You just need to know what time it is, and run an SHA1 hash. The time part is simple enough, as the ESP8266 can connect to the network and get the current time from NTP. The calculation of the TOTP is handled by the Python code once you’ve provided it with the “secret” pulled from the Google Authenticator application. It’s worth noting here that this means your 2FA secrets will be held in clear-text on the ESP8266’s flash, so try not to use this to secure any nuclear launch systems or anything, OK? Then again, if you ever lose it the beauty of 2-factor is you can invalidate the secret and generate a new one.
Google has announced their soon to be available Vision Kit, their next easy to assemble Artificial Intelligence Yourself (AIY) product. You’ll have to provide your own Raspberry Pi Zero W but that’s okay since what makes this special is Google’s VisionBonnet board that they do provide, basically a low power neural network accelerator board running TensorFlow.
The VisionBonnet is built around the Intel® Movidius™ Myriad 2 (aka MA2450) vision processing unit (VPU) chip. See the video below for an overview of this chip, but what it allows is the rapid processing of compute-intensive neural networks. We don’t think you’d use it for training the neural nets, just for doing the inference, or in human terms, for making use of the trained neural nets. It may be worth getting the kit for this board alone to use in your own hacks. An alternative is to get Modivius’s Neural Compute Stick, which has the same chip on a USB stick for around $80, not quite double the Vision Kit’s $45 price tag.
The Vision Kit isn’t out yet so we can’t be certain of the details, but based on the hardware it looks like you’ll point the camera at something, press a button and it will speak. We’ve seen this before with this talking object recognizer on a Pi 3 (full disclosure, it was made by yours truly) but without the hardware acceleration, a single object recognition took around 10 seconds. In the vision kit we expect the recognition will be in real-time. So the Vision Kit may be much more dynamic than that. And in case it wasn’t clear, a key feature is that nothing is done on the cloud here, all processing is local.
The kit comes with three different applications: an object recognition one that can recognize up to 1000 different classes of objects, another that recognizes faces and their expressions, and a third that detects people, cats, and dogs. While you can get up to a lot of mischief with just that, you can run your own neural networks too. If you need a refresher on TensorFlow then check out our introduction. And be sure to check out the Myriad 2 VPU video below the break.
Even if you don’t want to add an AUX audio output port to your Google Home Mini, you’ll still want to see a pair of videos from [SnekTek]. After all, you’ll eventually want to open it up, and putting it over some boiling water might not have been your first idea. You can see both videos, below.
However, he did want to add an AUX port. The biggest challenge was finding a place to put the connector. Even after identifying a likely spot, a bolt interfered with the case closing and so he removed it. The one bolt didn’t seem to bother the final result.
One of the tasks I dread is configuring a web server to send email correctly via Gmail. The simplest way of sending emails is SMTP, and there are a number of scripts out there that provide a simple method to send mail that way with a minimum of configuration. There’s even PHP mail(), although it’s less than reliable.
Out of the box, Gmail requires OAUTH2 for authentication and to share user data, which has the major advantage of not requiring that you store your username and password in the application that requires access to your account. While they have an ‘allow less secure apps’ option that allows SMTP access for legacy products like Microsoft Outlook, it just doesn’t seem like the right way forward. Google documents how to interact with their API with OAUTH2, so why not just use that instead of putting my username and password in plaintext in a bunch of prototypes and test scripts?
Those are the thoughts that run through my head every time this comes up for a project, and each time I’ve somehow forgotten the steps to do it, also forgotten to write it down, and end up wasting quite a bit of time due to my own foolishness. As penance, I’ve decided to document the process and share it with all of you, and then also make it work on an ESP8266 board running the Arduino development environment.
Old Radio Shack intercom; brand new Google Voice interface for a Raspberry Pi. One of these things is not like the other, but they ended up together in this retro-look Google Voice interface, and the results are pretty slick.
The recipient of the Google hive-mind transplant was one of three wireless FM intercoms [MisterM] scored for a measly £4. Looking much as they did when they were the must-have office tool or home accessory for your modern mid-80s lifestyle, the intercom case was the perfect host for the Pi and the Google AIY hat. Only the case was used — not even the original speaker made it into the finished product. The case got a good scrubbing, a fresh coat of paint to perk up the gone-green plastic, and an accent strip of Google’s logo colors over the now-deprecated station selector switch. [MisterM] provided a white LED behind the speaker grille for subtle feedback. A tap of the original talk bar gets Google’s attention for answers to quick questions, and integration into the family’s existing home automation platform turns the lights on and off. See it in action after the break.
[MisterM] was lucky enough to score an AIY hat for free, and as far as we know they’re still hard to come by. If you’re itching to try out the board, fear not — turns out you can roll your own.