"[T]he current capabilities of many AI systems closely match some of the specialized needs of disabled people.... Fortunately, there is a growing interest in applying the scientific knowledge and engineering experience developed by AI researchers to the domain of assistive technology and in investigating new methods and techniques that are required within the assistive technology domain."
– Bruce G. Buchanan; from his Foreword to Assistive Technology and Artificial Intelligence: Applications in Robotics, User Interfaces and Natural Language Processing
Students have one day to create prototype assistive devices to suit client needs. Students had access to a wide range of resources, including working space, machinery, and building materials, within Beaver Works and technical assistance from several mentors: John Vivilecchia, Kurt Krueger, and Richard Landry of MIT Lincoln Laboratory; Don Fredette of The Boston Home; Michael Buchman of the MIT Department of Mechanical Engineering; and Mary Ziegler of the MIT Office of Digital Learning. The team decided to hack a universal remote that communicates via Wi-Fi with a web interface from which Dan could control television power, volume, and channels. Once the build time was over, several judges, including the ATHack organizers, David Crandelle and David Binder of the Spaulding Rehabilitation Network; John Vivilecchia; Don Fredette; and Mary Ziegler evaluated each team's device.
Through coordination of the patient's prosthetic limb, existing nerves, and muscle grafts, amputees would be able to sense where their limbs are in space and to feel how much force is being applied to them. For example, when you bend your elbow, the biceps muscle contracts, causing the triceps to stretch, and that triceps stretch sends sensory information related to position, velocity, and force back to the brain. Without these intact muscle pairs, persons with limb amputation have no way of sensing where their artificial limbs are, nor can they sense the forces applied to those limbs. When the brain sends signals instructing a limb to move, one of the grafted muscles will contract, and its agonist will extend.
Chinese tech giant Baidu's text-to-speech system, Deep Voice, is making a lot of progress toward sounding more human. Baidu says that unlike previous text-to-speech systems, Deep Voice 2 finds shared qualities between the training voices entirely on its own, and without any previous guidance. "Deep voice 2 can learn from hundreds of voices and imitate them perfectly," a blog post says. In a research paper (PDF), Baidu concludes that its neural network can create voice pretty effectively even from small voice samples from hundreds of different speakers.
TL;DR Baidu's TTS system now supports multi-speaker conditioning, and can learn new speakers with very little data (a la LyreBird). I'm really excited about the recent influx of neural-net TTS systems, but all of the them seem to be too slow for real time dialog, or not publicly available, or both. Hoping that one of them gets a high quality open-source implementation soon!
Derrick Campana kneels beside Angel Marie, a three-legged mini horse who wears a prosthetic leg made by Campana. Campana made the jump to the animal field 12 years ago when few, if any, people created artificial limbs for dogs and other pets. Derrick Campana holds the prosthetic paw he made for Kenna, a three year-old golden retriever born without a front paw. Derrick Campana holds the molds for prosthetic legs he made for two Thai elephants who lost limbs in landmine explosions.
"It's not safe," an airline representative allegedly told Seward during an exchange that was not captured on video. Seward then asks the Asiana Airlines agent why he is being asked to change seats. As ABC pointed out, the Federal Aviation Administration's website clearly states that "physical ability" to perform the necessary duties required of exit row seats are used to determine whether a person with a prosthetic can sit there -- not the prosthetic alone. "If a passenger with a prosthesis is being evaluated for assignment to an exit seat, the presence of the prosthesis would not be the determinant for being able to meet the criteria but rather the physical ability to perform the exit seat duties," reads the FAA website.
Tim Seward said that he paid more for a seat on an Asiana Airlines flight with extra leg room, but a flight attendant told him Sunday that he had to get up or face ejection. "They threatened me that they were going to kick me off the plane if I didn't move," Seward told news station KGO-TV. In the video, the flight attendant says he's concerned that Seward won't be able to perform the exit seat duties. The Federal Aviation Administration's guidelines say that "presence of the prosthesis would not be the determinant" for performing exit seat duties, but rather physical ability.
With most contemporary limbs, the signal to grip is detected using myoelectric sensors -- which read muscle activity from the skin. That's the idea being developed by biomedical researchers from Newcastle University, who have developed a prototype prosthetic limb with a AI-powered camera mounted on top. When the wearer of the limb moves to grab, say, a mug, the camera takes a picture of the object, and moves the hand into a suitable "grasp type." The user then confirms the grip action with a myoelectric signal.
Most of us take for granted how well our brains can relay instructions to our limbs. Doctors and engineers have been trying for years to grant that same surety to those with prosthetic limbs. But interfacing the biological and technological is tricky. There have been some impressive advances in this area of research, but they usually require the prosthetic to be directly wired into the patient's brain--not exactly practical.
A new technique where the robotic arm clicks directly to the bone, however, is showing promise. The "mind control" is enabled by connecting a patient's nerves to the socket, with a special Bluetooth bracelet to receive the signals. It takes three surgeries to make it happen, though, including one to insert the a metal rod into a patient's bone marrow, one to implant the piece that will eventually connect the robot arm, and a third -- performed by a specialized plastic surgeon -- to connect all the nerves that used to control the patient's hand muscles to the upper arm stump. The entire process, including surgery and rehab, is strenuous and takes a while to complete.