Alexa, Bring Me A Beer!

June 14, 2021 by Jenny List 11 Comments

Voice controlled home assistants are the wonder of our age, once you’ve made peace with the privacy concerns of sharing the intimacies of your life with a data centre owned by a massive corporation, anyway. They provide a taste of how the future was supposed to be in those optimistic predictions of decades past: Alexa and Siri can crack jokes, control your lights, answer questions, tell you the news, and so much more.

But for all their electronic conversational perfection, your electronic pals can’t satisfy your most fundamental needs and bring you a beer. This is something [luisengineering] has fixed, an he’s provided the appropriate answer to the question “Alexa: bring mir ein bier!“. The video which we’ve also put below the break is in German with YouTube’s automatic closed captions if you want them, but we think you’ll be able to get the point of it if not all his jokes without needing to learn to speak a bit of Deutsch.

As he develops his beer-delivery system we begin to appreciate that what might seem to be a relatively straightforward task is anything but. He takes an off-the-shelf robot and gives it a beer-bottle grabber and ice hopper, but the path from fridge to sofa still needs a little work. The eventual solution involves a lot of trial and error, and a black line on the floor for the ‘bot to follow. Finally, his electronic friend can bring him a beer!

We like [Luis]’s entertaining presentational style, and the use of props as microphone stands. We’ll be keeping an eye out for what he does next, and you should too. Meanwhile it may not surprise you that this is not the first beer-delivery ‘bot we’ve brought you.

Continue reading “Alexa, Bring Me A Beer!” →

Speech Recognition On An Arduino Nano?

May 26, 2021 by Orlando Hoilett 24 Comments

Like most of us, [Peter] had a bit of extra time on his hands during quarantine and decided to take a look back at speech recognition technology in the 1970s. Quickly, he started thinking to himself, “Hmm…I wonder if I could do this with an Arduino Nano?” We’ve all probably had similar thoughts, but [Peter] really put his theory to the test.

The hardware itself is pretty straightforward. There is an Arduino Nano to run the speech recognition algorithm and a MAX9814 microphone amplifier to capture the voice commands. However, the beauty of [Peter’s] approach, lies in his software implementation. [Peter] has a bit of an interplay between a custom PC program he wrote and the Arduino Nano. The learning aspect of his algorithm is done on a PC, but the implementation is done in real-time on the Arduino Nano, a typical approach for really any machine learning algorithm deployed on a microcontroller. To capture sample audio commands, or utterances, [Peter] first had to optimize the Nano’s ADC so he could get sufficient sample rates for speech processing. Doing a bit of low-level programming, he achieved a sample rate of 9ksps, which is plenty fast for audio processing.

To analyze the utterances, he first divided each sample utterance into 50 ms segments. Think of dividing a single spoken word into its different syllables. Like analyzing the “se-” in “seven” separate from the “-ven.” 50 ms might be too long or too short to capture each syllable cleanly, but hopefully, that gives you a good mental picture of what [Peter’s] program is doing. He then calculated the energy of 5 different frequency bands, for every segment of every utterance. Normally that’s done using a Fourier transform, but the Nano doesn’t have enough processing power to compute the Fourier transform in real-time, so Peter tried a different approach. Instead, he implemented 5 sets of digital bandpass filters, allowing him to more easily compute the energy of the signal in each frequency band.

The energy of each frequency band for every segment is then sent to a PC where a custom-written program creates “templates” based on the sample utterances he generates. The crux of his algorithm is comparing how closely the energy of each frequency band for each utterance (and for each segment) is to the template. The PC program produces a .h file that can be compiled directly on the Nano. He uses the example of being able to recognize the numbers 0-9, but you could change those commands to “start” or “stop,” for example, if you would like to.

[Peter] admits that you can’t implement the type of speech recognition on an Arduino Nano that we’ve come to expect from those covert listening devices, but he mentions small, hands-free devices like a head-mounted multimeter could benefit from a single word or single phrase voice command. And maybe it could put your mind at ease knowing everything you say isn’t immediately getting beamed into the cloud and given to our AI overlords. Or maybe we’re all starting to get used to this. Whatever your position is on the current state of AI, hopefully, you’ve gained some inspiration for your next project.

“Alexa, Stop Listening To Me Or I’ll Cut Your Ears Off”

February 2, 2021 by Dan Maloney 88 Comments

Since we’ve started inviting them into our homes, many of us have began casting a wary eye at our smart speakers. What exactly are they doing with the constant stream of audio we generate, some of it coming from the most intimate and private of moments? Sure, the big companies behind these devices claim they’re being good, but do any of us actually buy that?

It seems like the most prudent path is to not have one of these devices, but they are pretty useful tools. So this hardware mute switch for an Amazon Echo represents a middle ground between digital Luddism and ignoring the possible privacy risks of smart speakers. Yes, these devices all have software options for disabling their microphone arrays, but as [Andrew Peters] relates it, his concern is mainly to thwart exotic attacks on smart speakers, some of which, like laser-induced photoacoustic attacks, we’ve previously discussed. And for that job, only a hardware-level disconnect of the microphones will do.

To achieve this, [Andrew] embedded a Seeeduino Xiao inside his Echo Dot Gen 2. The tiny microcontroller grounds the common I²S data line shared by the seven (!) microphones in the smart speaker, effective disabling them. Enabling and disabling the mics is done via the existing Dot keys, with feedback provided by tones sent through the Dot speaker. It’s a really slick mod, and the amount of documentation [Andrew] did while researching this is impressive. The video below and the accompanying GitHub repo should prove invaluable to other smart speaker hackers.

Continue reading ““Alexa, Stop Listening To Me Or I’ll Cut Your Ears Off”” →

On-Air Sign Helps Keep Your Broadcasts G-Rated

January 18, 2021 by Kristina Panos 7 Comments

Like many of us, [Michael] needed a way to let the family know whether pants are required to enter the room — in other words, whenever a videoconference is in progress. Sure he could hang a do not disturb sign, but those are easy to forget. There’s no need to worry about forgetting to change status because this beautiful wall-mounted sign can be controlled with Alexa.

Inside the gorgeous box made from walnut, curly maple, and oak is an ESP32, some RGB LEDs, and three MOSFETs. [Michael] is using the fauxmoESP library to interface the ESP32 with Alexa, which emulates a Phillips Hue bulb for the sake of using a protocol she already knows. [Michael] can change the color and brightness percentage with voice commands.

The sign is set up as four different devices — one default, and one for each color. Since talking to Alexa isn’t always appropriate, [Michael] can also change the color of the LEDs using sliders on a website that’s served up by the ESP. Check out the full build video after the break.

Need something quick and dirty that works just as well? Our own [Bob Baddeley] made a status indicator that’s simple and effective.

Continue reading “On-Air Sign Helps Keep Your Broadcasts G-Rated” →

Automating Your Car With A Spare Fob And An ESP8266

January 6, 2021 by Tom Nardi 47 Comments

Despite the name, home automation doesn’t have to be limited to only the devices within your home. Bringing your car into the mix can open up some very interesting possibilities, such as automatically getting it warmed up in the morning if the outside air temperature drops below a certain point. The only problem is, not everyone is willing to start hacking their ride’s wiring to do it.

Which is exactly why [Matt Frost] went the non-invasive route. By wiring up an ESP8266 to a cheap aftermarket key fob for his Chevrolet Suburban, he’s now able to wirelessly control the door locks and start the engine without having to make any modifications to the vehicle. He was lucky that the Chevy allowed him to program his own fob, but even if you have to spend the money on getting a new remote from the dealer, it’s sure to be cheaper than the repair bill should you cook something under the dash with an errant splice or a misplaced line of code.

The hardware for this project is about as simple as it gets. The fob is powered by the 3.3 V pin on the Wemos D1 Mini, and the traces for the buttons have been hooked up to the GPIO pins. By putting both boards into a custom 3D printed enclosure, [Matt] came up with a tidy little box that he could mount in his garage and run off of a standard USB power supply.

On the software side of things [Matt] has the device emulating a smart light so it can easily be controlled by his Alexa, with a few helpful routines sprinkled in that allow him to avoid the awkward phraseology that would be required otherwise. There’s also a minimal web server running on the microcontroller that lets him trigger various actions just by hitting the appropriate URLs, which made connecting it to Home Assistant a snap. One downside of this approach is that there’s no acknowledgement from the vehicle that the command was actually received, but you can always send a command multiple times to be sure.

This isn’t the first time we’ve seen an ESP8266 used to “push” buttons on a remote. If you’ve got a spare fob for your device, or can get one, it’s an excellent way to automate it on the cheap.

Robots Can Finally Answer, Are You Talking To Me?

November 8, 2020 by Matthew Carlson 25 Comments

Voice Assistants, love them, or hate them, are becoming more and more commonplace. One problem for voice assistants is the situation of multiple devices listening in the same place. When a command is given, which device should answer? Researchers at CMU’s Future Interfaces Group [Karan Ahuja], [Andy Kong], [Mayank Goel], and [Chris Harrison] have an answer; smart assistants should try to infer if the user is facing the device they want to talk to. They call it direction-of-voice or DoV.

Currently, smart assistants use a simple race to see who heard it first. The reasoning is that the device you are closest to will likely hear it first. However, in situations with echos or when you’re equidistant from multiple devices, the outcome can seem arbitrary to a user.

The implementation of DoV uses an Extra-Trees Classifier from the python sklearn toolkit. Several other machine learning algorithms were considered, but ultimately efficiency won out and Extra-Trees was selected. Another interesting facet of the research was determining what facing really means. The team had humans ‘listeners’ stand in for smart assistants. A ‘talker’ would speak the key phrase while the ‘listener’ determined if the talker was facing them or not. Based on their definition of facing, the system can determine if someone is facing the device with 90% accuracy that rises to 93% with per-room calibration.

Their algorithm as well as the data they collected has been open-sourced on GitHub. Perhaps when you’re building your own voice assistant, you can incorporate DoV to improve wake-word accuracy.

Continue reading “Robots Can Finally Answer, Are You Talking To Me?” →

A Smart Speaker That Reminds You It’s Listening

October 15, 2020 by Matthew Carlson 20 Comments

[markw2k9] has an Alexa device that sits in his kitchen and decided it was time to spruce it up with some rather uncanny eyes. With some inspiration from the Adafruit Uncanny Eyes project, which displays similar animated eyes, [markw2k9] designed a 3d printed shell that goes on top of a 2nd generation Amazon Echo. A teensy 3.2 powers two OLED displays and monitors the light ring to know when to turn the lights on and show that your smart speaker is listening. The eyes look around in a shifty sort of manner. Light from the echo’s LED ring is diffused through a piece of plexiglass that was lightly sanded on the outside ring and the eye lenses are 30mm cabochons (a glass lens often used for jewelry).

One hiccup is that the ring on the Echo will glow in a steady pattern when there’s a notification. As this would cause the OLEDs to be on almost continuously and concerned for the lifetime of the OLED panels, the decision was made to detect this condition in the state machine and go into a timeout state. With that issue solved, the whole thing came together nicely. Where this project really shines is the design and execution. The case is sleek PLA and the whole thing looks professional.

We’ve seen a few other projects inspired by the animated eyes project such as this Halloween themed robot that is honestly quite terrifying. The software and STL files for the smart speaker’s eyes are on Github and Thingiverse.

Continue reading “A Smart Speaker That Reminds You It’s Listening” →