The groundbreaking findings, published in the journal Nano Energy, suggest "we're on the path toward wearable devices powered by human motion," said Nelson Sepulveda, associate professor of electrical and computer engineering and lead investigator of the project.

"What I foresee, relatively soon, is the capability of not having to charge your cell phone for an entire week, for example, because that energy will be produced by your movement," said Sepulveda, whose research is funded by the National Science Foundation.

The innovative process starts with a silicone wafer, which is then fabricated with several layers, or thin sheets, of environmentally friendly substances including silver, polyimide and polypropylene ferro electret. Ions are added so that each layer in the device contains charged particles. Electrical energy is created when the device is compressed by human motion, or mechanical energy. The completed device is called a bio-compatible ferro electret nano-generator, or FENG. The device is as thin as a sheet of paper and can be adapted to many applications and sizes. The device used to power the LED lights was palm-sized, for example, while the device used to power the touch screen was as small as a finger.

Advantages such as being lightweight, flexible, bio-compatible, scalable, low-cost and robust could make FENG "a promising and alternative method in the field of mechanical-energy harvesting" for many autonomous electronics such as wireless headsets, cell phones and other touch-screen devices, the study says. Remarkably, the device also becomes more powerful when folded.

"Each time you fold it you are increasing exponentially the amount of voltage you are creating," Sepulveda said. "You can start with a large device, but when you fold it once, and again, and again, it's now much smaller and has more energy. Now it may be small enough to put in a specially made heel of your shoe so it creates power each time your heel strikes the ground."

Sepulveda and his team are developing technology that would transmit the power generated from the heel strike to, say, a wireless headset.

The Bluetooth Special Interest Group (SIG) officially adopted Bluetooth 5 as the latest version of the Bluetooth core specification this week.

Key updates to Bluetooth 5 include longer range, faster speed, and larger broadcast message capacity, as well as improved interoperability and coexistence with other wireless technologies. Bluetooth 5 continues to advance the Internet of Things (IoT) experience by enabling simple and effortless interactions across the vast range of connected devices.

"Bluetooth is revolutionizing how people experience the IoT. Bluetooth 5 continues to drive this revolution by delivering reliable IoT connections and mobilizing the adoption of beacons, which in turn will decrease connection barriers and enable a seamless IoT experience." Mark Powell, Executive Director of the Bluetooth SIG

Key feature updates include four times range, two times speed, and eight times broadcast message capacity. Longer range powers whole home and building coverage, for more robust and reliable connections. Higher speed enables more responsive, high-performance devices. Increased broadcast message size increases the data sent for improved and more context relevant solutions.

Bluetooth 5 also includes updates that help reduce potential interference with other wireless technologies to ensure Bluetooth devices can coexist within the increasingly complex global IoT environment. Bluetooth 5 delivers all of this while maintaining its low-energy functionality and flexibility for developers to meet the needs of their device or application.

Consumers can expect to see products built with Bluetooth 5 within two to six months of today’s release.

Source: Bluetooth.com

Author of this post :

University of Washington researchers have taken a first step in showing how humans can interact with virtual realities via direct brain stimulation. In a paper published online Nov. 16 in Frontiers in Robotics and AI, they describe the first demonstration of humans playing a simple, two-dimensional computer game using only input from direct brain stimulation -- without relying on any usual sensory cues from sight, hearing or touch.

The subjects had to navigate 21 different mazes, with two choices to move forward or down based on whether they sensed a visual stimulation artifact called a phosphine, which are perceived as blobs or bars of light. To signal which direction to move, the researchers generated a phosphine through transcranial magnetic stimulation, a well-known technique that uses a magnetic coil placed near the skull to directly and noninvasively stimulate a specific area of the brain.

"The way virtual reality is done these days is through displays, headsets and goggles, but ultimately your brain is what creates your reality," said senior author Rajesh Rao, UW professor of Computer Science & Engineering and director of the Center for Sensorimotor Neural Engineering.

"The fundamental question we wanted to answer was: Can the brain make use of artificial information that it's never seen before that is delivered directly to the brain to navigate a virtual world or do useful tasks without other sensory input? And the answer is yes."

The five test subjects made the right moves in the mazes 92 percent of the time when they received the input via direct brain stimulation, compared to 15 percent of the time when they lacked that guidance.

The simple game demonstrates one way that novel information from artificial sensors or computer-generated virtual worlds can be successfully encoded and delivered noninvasively to the human brain to solve useful tasks. It employs a technology commonly used in neuroscience to study how the brain works -- transcranial magnetic stimulation -- to instead convey actionable information to the brain. The test subjects also got better at the navigation task over time, suggesting that they were able to learn to better detect the artificial stimuli.

"We're essentially trying to give humans a sixth sense," said lead author Darby Losey, a 2016 UW graduate in computer science and neurobiology who now works as a staff researcher for the Institute for Learning & Brain Sciences (I-LABS). "So much effort in this field of neural engineering has focused on decoding information from the brain. We're interested in how you can encode information into the brain."

The initial experiment used binary information -- whether a phosphine was present or not -- to let the game players know whether there was an obstacle in front of them in the maze. In the real world, even that type of simple input could help blind or visually impaired individuals navigate.

Theoretically, any of a variety of sensors on a person's body -- from cameras to infrared, ultrasound, or laser rangefinders -- could convey information about what is surrounding or approaching the person in the real world to a direct brain stimulator that gives that person useful input to guide their actions.

"The technology is not there yet -- the tool we use to stimulate the brain is a bulky piece of equipment that you wouldn't carry around with you," said co-author Andrea Stocco, a UW assistant professor of psychology and I-LABS research scientist. "But eventually we might be able to replace the hardware with something that's amenable to real world applications."

Author of this post :

Speech recognition systems, such as those that convert speech to text on cellphones, are generally the result of machine learning. A computer pores through thousands or even millions of audio files and their transcriptions, and learns which acoustic features correspond to which typed words. But transcribing recordings is costly, time-consuming work, which has limited speech recognition to a small subset of languages spoken in wealthy nations.

At the Neural Information Processing Systems conference this week, researchers from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) are presenting a new approach to training speech-recognition systems that doesn't depend on transcription. Instead, their system analyzes correspondences between images and spoken descriptions of those images, as captured in a large collection of audio recordings. The system then learns which acoustic features of the recordings correlate with which image characteristics.

"The goal of this work is to try to get the machine to learn language more like the way humans do," says Jim Glass, a senior research scientist at CSAIL and a co-author on the paper describing the new system.

The current methods that people use to train up speech recognizers are very supervised. You get an utterance, and you're told what's said. And you do this for a large body of data. "Big advances have been made -- Siri, Google -- but it's expensive to get those annotations, and people have thus focused on, really, the major languages of the world. There are 7,000 languages, and I think less than 2 percent have ASR [automatic speech recognition] capability, and probably nothing is going to be done to address the others. So if you're trying to think about how technology can be beneficial for society at large, it's interesting to think about what we need to do to change the current situation. And the approach we've been taking through the years is looking at what we can learn with less supervision." Joining Glass on the paper are first author David Harwath, a graduate student in electrical engineering and computer science (EECS) at MIT; and Antonio Torralba, an EECS professor.

The version of the system reported in the new paper doesn't correlate recorded speech with written text; instead, it correlates speech with groups of thematically related images. But that correlation could serve as the basis for others. If, for instance, an utterance is associated with a particular class of images, and the images have text terms associated with them, it should be possible to find a likely transcription of the utterance, all without human intervention. Similarly, a class of images with associated text terms in different languages could provide a way to do automatic translation.

Conversely, text terms associated with similar clusters of images, such as, say, "storm" and "clouds," could be inferred to have related meanings. Because the system in some sense learns words' meanings -- the images associated with them -- and not just their sounds, it has a wider range of potential applications than a standard speech recognition system. To test their system, the researchers used a database of 1,000 images, each of which had a recording of a free-form verbal description associated with it. They would feed their system one of the recordings and ask it to retrieve the 10 images that best matched it. That set of 10 images would contain the correct one 31 percent of the time.

"I always emphasize that we're just taking baby steps here and have a long way to go," Glass says. "But it's an encouraging start."

The researchers trained their system on images from a huge database built by Torralba; Aude Oliva, a principal research scientist at CSAIL; and their students. Through Amazon's Mechanical Turk crowdsourcing site, they hired people to describe the images verbally, using whatever phrasing came to mind, for about 10 to 20 seconds. For an initial demonstration of the researchers' approach, that kind of tailored data was necessary to ensure good results. But the ultimate aim is to train the system using digital video, with minimal human involvement.

"I think this will extrapolate naturally to video," Glass says.

To build their system, the researchers used neural networks, machine-learning systems that approximately mimic the structure of the brain. Neural networks are composed of processing nodes that, like individual neurons, are capable of only very simple computations but are connected to each other in dense networks. Data is fed to a network's input nodes, which modify it and feed it to other nodes, which modify it and feed it to still other nodes, and so on. When a neural network is being trained, it constantly modifies the operations executed by its nodes in order to improve its performance on a specified task. The researchers' network is, in effect, two separate networks: one that takes images as input and one that takes spectrograms, which represent audio signals as changes of amplitude, over time, in their component frequencies. The output of the top layer of each network is a 1,024-dimensional vector -- a sequence of 1,024 numbers.

The final node in the network takes the dot product of the two vectors. That is, it multiplies the corresponding terms in the vectors together and adds them all up to produce a single number. During training, the networks had to try to maximize the dot product when the audio signal corresponded to an image and minimize it when it didn't. For every spectrogram that the researchers' system analyzes, it can identify the points at which the dot-product peaks. In experiments, those peaks reliably picked out words that provided accurate image labels -- "baseball," for instance, in a photo of a baseball pitcher in action, or "grassy" and "field" for an image of a grassy field.

In ongoing work, the researchers have refined the system so that it can pick out spectrograms of individual words and identify just those regions of an image that correspond to them.

Source: Materials provided by Massachusetts Institute of Technology. Original written by Larry Hardesty.

Author of this post :