If you’ve ever wondered how people are really feeling during a conversation, you’re not alone. By and large, we rely on a huge number of cues — body language, speech, eye contact, and a million others — to determine the feelings of others. It’s an inexact science to say the least. Now, researchers at MIT have developed a wearable system to analyze the tone of a conversation.
The system uses Samsung Simband wearables, which are capable of measuring several physiological markers — heart rate, blood pressure, blood flow, and skin temperature — as well as movement thanks to an on-board accelerometer. This data is fed into a neural network which was trained to classify a conversation as “happy” or “sad”. Training consisted of capturing 31 conversations of several minutes duration each, where participants were asked to tell a happy or sad story of their own choosing. This was done in an effort to record more organic emotional states than simply eliciting emotion through the use of more typical “happy” or “sad” video materials often used in similar studies.
The technology is in a very early stage of development, however the team hopes that down the road, the system will be sufficiently advanced to act as an emotional coach in real-life social situations. There is a certain strangeness about the idea of asking a computer to tell you how a person is feeling, but if humans are nothing more than a bag of wet chemicals, there might be merit in the idea yet. It’s a pretty big if.
Machine learning is becoming more powerful on a daily basis, particularly as we have ever greater amounts of computing power to throw behind it. Check out our primer on machine learning to get up to speed.
Continue reading “Wearable Predicts Tone of Conversation from Speech, Vital Signs”
Using your smartphone to control your home automation system gives you a lot of flexibility. But for something as simple as turning the lights on and off, it can be a pain to go through the whole process of unlocking your phone, choosing the right browser page or app, and then finally hitting the button you need. It’d be much simpler if it could all be done at the touch of a single, physical button – but phones don’t have many of those anymore. [falldeaf] brings the solution – a four-button Bluetooth remote for your smartphone that’s wearable, to boot.
The project is built around the RFDuino, an Arduino platform used for quickly and easily building Bluetooth compatible projects. So far, so simple – four buttons wired into a microcontroller with wireless capability onboard. The real trick is the 3D-printed clothespin style case which allows you to clip the four-button remote onto your clothes. [falldeaf]’s first attempt was a palm-mounted setup that they found got in the way of regular tasks; we agree that the wearable version offers a serious upgrade in utility.
The smartphone side of things is handled with a custom app [falldeaf] coded using PhoneGap. This is where actions for the buttons can be customized, including using the buttons to navigate a menu system to enable the user to select more than just one function per button. It adds a high level of flexibility, so you can create all kinds of macros to control your whole home automation system from your button clip.
It’s really great to see a project that considers ergonomics and usability above and beyond just creating the baseline functionality. Follow this train of thought and you’ll find yourself enjoying your projects in the use phase well beyond the initial build. Another great example is this self-charging electrically heated jacket. Video after the break.
Continue reading “Bluetooth Automation Remote Hangs Around”
Halloween has come and gone, but this DIY voice changing Star Wars Stormtrooper helmet tutorial by [Shawn Hymel] is worth a look for a number of reasons. Not only is the whole thing completely self-contained, but the voice changing is done in software thanks to the Teensy’s powerful audio filtering abilities. In addition, the Teensy also takes care of adding the iconic Stormtrooper clicks, pops, and static bursts around the voice-altered speech. Check out the video below to hear it in action.
Besides a microphone and speakers, there’s a Teensy 3.2, a low-cost add-on board for the Teensy that includes a small audio amp, a power supply… and that’s about it. There isn’t a separate WAV board or hacked MP3 player in sight.
Continue reading “Stormtrooper Voice Changer Helmet uses Teensy to Mangle Audio”
Some of the more interesting consumer hardware devices of recent years have been smart glasses. Devices like Google Glass or Snapchat Spectacles, eyewear incorporating a display and computing power to deliver information or provide augmented reality on an unobtrusive wearable platform.
Raspberry Pi Zero Smart Glass aims to provide an entry into this world, with image recognition and OCR text recognition in a pair of glasses courtesy of a Raspberry Pi Zero. Unusually though it does not take the display option of other devices of having a mirror or prism in the user’s field of view, instead it replaces the user’s entire field of view with a display and re-connects them to the world through the Raspberry Pi camera.
The display in question is an inexpensive set of “3D Virtual Stereo Digital Video glasses”, of the type that can be found fairly easily on your favourite auction site. They aren’t particularly high-resolution, but the Pi can easily drive them with its composite video output. The electronics and camera are mounted on a headband, in a custom 3D-printed enclosure. All files can be downloaded from the project page.
There is some Python software, but it’s fair to say that there is not a clear demo on the project page showing it working. However this is no reason to disregard this project, because even if its software has yet to achieve its full potential there is value elsewhere. The 3D-printed Raspberry Pi enclosure should be of use to many other similar wearable projects, and we’d almost say it’s worthy of a project all of its own.
MIT Media Lab and Microsoft have teamed up to take wearable devices one step further — they’ve glued the devices directly to the user’s skin. DuoSkin is a temporary tattoo created with gold leaf. Metallic “Flash” temporary fashion tattoos have become quite popular recently, so this builds on the trend. What the team has done is to use them to create user interfaces for wearable electronic devices.
Generally speaking, gold leaf is incredibly fragile. In this process to yield the cleanest looking leaf the gold is not actually cut. Instead, the temporary tattoo film and backer are cut on a standard desktop vinyl cutter. The gold leaf is then applied to the entire film surface. The cut film/leaf can then be “weeded” — removing the unwanted portions of film which were isolated from the rest by the cutting process — to complete the temporary tattoo. The team tested this method and found that traces 4.5 mm or more thick were resilient enough to last the entire day on your skin.
The gold leaf tattoos make excellent capacitive touch sensors. The team was able to create sliders, buttons, and even 2 dimensional diamond grids. These controls were used to move a cursor on a computer or phone screen. They were even able to create a wearable NFC tag. The gold leaf is the antenna, and the NFC chip itself is mounted on the temporary tattoo backer.
These devices all look great, but with the exception of the NFC chip, we’re not seeing the electronics driving them. Capacitive touch sensors used as a UI for a phone will have to have a Bluetooth radio and a battery somewhere. We’re that’s all hidden under the arm of the user. You can see what we’re taking about in the video after the break. That said, the tools and materials are ubiquitous and easy to work with. Take a quick read through the white paper (PDF) and you can be making your own version of this today.
Continue reading “Skin Bling: Wearable Electronics from Golden Temporary Tattoos”
A person who is deaf can’t hear sound, but that doesn’t mean they can’t feel vibrations. For his Hackaday Prize entry, [Alex Hunt] is developing the Shakelet, a vibrating wristband for that notifies hearing impaired people about telephones, doorbells, and other sound alerts.
To tackle the difficulty of discriminating between the different sounds from different sources, [Alex’s] wants to attach little sound sensors directly to the sound emitting devices. The sensors wirelessly communicate with the wristband. If the wristband receives a trigger signal from one of the sensors, it alerts the wearer by vibrating. It also shows which device triggered the alert by flashing an RGB LED in a certain color. A first breadboard prototype of his idea confirmed the feasibility of the concept.
After solving a few minor problems with the sensitivity of the sensors, [Alex] now has a working prototype. The wristband features a pager motor and is controlled by an ATMEGA168. Two NRF24L01+ 2.4 GHz wireless transceiver modules take care of the communication. The sound sensors run on the smaller ATTiny85 and use a piezo disc as microphone. Check out the video below, where Alex demonstrates his build:
Continue reading “Hackaday Prize Entry: Shakelet”
[Nick Sayer] can reflow, and he can prove it. He designed a simple blinking-LED circuit that uses SMD parts to, well, blink LEDs. That’s not the point, though. It’s designed to be a test platform for reflow soldering, and to use a minimum number of valuable parts. Plus, it says “I can reflow!” in exposed copper. What else do you want?
OK, as far as “proving it” goes, the badge isn’t 100% reliable — we hand-solder 0805 components all day long. But still, if you want to try your hand at reflowing a circuit board, and you don’t want to ruin a lot of expensive parts if you fail, something like this is a good idea.
The design is open, and it’s really the idea that’s the point here anyway. How about something that would be really onerous to hand-solder, but still cheap? We’re thinking a matrix of tiny LEDs and a shift register or something.
We just ran an article on a hand-soldering challenge board, this seems the perfect complement. Display both proudly on your desk and confound and amaze your coworkers!