Lip-reading involves tracking facial movements to determine what someone is saying. Many lip-reading devices point a camera at the user's face. Others rely on sensors stuck around the speaker's mouth. Neither approach is suitable for daily use, says Ruidong Zhang of Cornell University in Ithaca, NY. His team built a new lip-reading tech on a pair of eyeglasses, which uses acoustics—sound—to recognize silent speech. Zhang presented this work on April 19, 2023 at the ACM Conference in Hamburg, Germany.
"Imagine the sonar system whales use, " says Zhang. They send a sound into their environment and listen for echoes (回声). From those echoes, they locate objects in their surroundings. "Our approach is similar, but not exactly the same," Zhang explains. "We're not just interested in locating something. Instead, we're trying to track moving patterns."
Zhang calls the new tech EchoSpeech, consisting of two small speakers under one lens of a pair of glasses, two small microphones under the other lens, and a circuit board attached to one of the side arms. When EchoSpeech is switched on, its speakers play high-pitched sounds. People can't hear them. The sound waves move in every direction. Some travel around the user's lips and mouth. While speaking, the user's facial movements change the paths of those sound waves. That, in turn, changes the echo patterns picked up by the microphones. These patterns are sent to the wearer's smartphone over Bluetooth. Using AI, an EchoSpeech app then interprets the echo patterns. It matches each pattern to commands the smartphone then follows.
EchoSpeech currently recognizes 31 voice commands and recognizes numbers that are three to six digits long. According to Zhang, future versions could recognize a much larger vocabulary. If so, users could write personal text messages via silent speech. In a noisy restaurant, they could use that approach to send messages to friends who're hard of hearing.
People who've had their vocal cords (声带) removed have been contacting Zhang's team. They want to know if this interface could read their lips and then speak out loud for them. Zhang is now exploring whether EchoSpeech could do this in a person's own voice.
1. Why does Zhang mention the sonar system used by whales in paragraph 2?A.To make their new tech more convincing. |
B.To highlight the advantages of their new tech. |
C.To explain the inspiration for their new tech. |
D.To make their new tech more understandable. |
A.Sound travels in a straight line. |
B.One's facial movements are unique. |
C.Sound waves are reflected from an object. |
D.Microphones are sensitive to loud sound. |
A.People can hear sounds created by EchoSpeech. |
B.EchoSpeech will be put into use before long. |
C.EchoSpeech can enhance interpersonal communication. |
D.Users can use EchoSpeech to write personal messages. |
A.New glasses read your lips and tell your phone. |
B.Lip-reading devices can understand voice commands. |
C.Lip-reading techniques are springing up worldwide. |
D.Recognizing silent speech has already become possible. |
相似题推荐
【推荐1】Most teens are trying to find their purposes in life. However, Gitanjali Rao, a 15-year-old girl has been coming up with creative solutions to worldwide problems since she was 10. It is, therefore, not surprising that the teen has won the honor of “America’s Top Young Scientist”.
In the 3rd grade, Rao was inspired to do something after seeing the city’s primary water supply in her hometown was polluted by a factory nearby and that high levels of lead (铅)made its way into people’s drinking water.
After 2 months’ research, Rao designed a tool that used sensors to instantly discover lead in water. Called Tethys, after the Greek Goddess (女神)of freshwater, it informs the residents via an app if their drinking water contains lead. The design earned her the 3M Young Scientist Challenge in 2017. She is now working with scientists and medical professionals to test Tethys’ potential and hopes the device will be ready for market in the near future.
Later, Rao took on another social issue — drug (毒品)addiction. Her app, called Epione, which won the Health Pillar Prize in May 2019, is designed to catch drug addiction in young adults before it’s too late.
More recently, the teen has developed an app named Kindly, which uses AI technology to find possible signs of cyberbullying (网络欺凌). When users type in a word or phrase, Kindly is able to pick it up if it’s bullying, and then it gives the choice to edit it or send it the way it is. It gives them the chance to rethink what they are saying so that they know what to do next time.
Gitanjali Rao has been selected from 5, 000 equally impressive nominees (被提名人)for TIME’s first-ever “Kid of the Year”.
1. What led to Rao’s inventing Tethys?A.The accident of lead pollution. |
B.The issue with drug addiction. |
C.The shortage of water supplies. |
D.The high cost of making water clean. |
A.It’ll be fitted to cellphones. |
B.It’ll win her a higher prize. |
C.It’ll be put on the market soon. |
D.It’ll remove lead from water. |
A.Receive warning signals of threat. |
B.Input words into a computer directly. |
C.Choose from safe social networking sites. |
D.Consider their words before posting online. |
A.Professional and positive. | B.Creative and productive. |
C.Inspiring and adventurous | D.Determined and amazing |
【推荐2】The history of writing instruments, with which humans have recorded and conveyed thoughts and feelings, is the history of civilization itself. This is how we know about our ancestors and their life.
The handy sharpened-stone was adapted into the first writing instrument. Around 24,000 BC, the cave man started drawing pictures with the stone onto the walls of his cave.
Before paper came along, people used clay or wax tablets on which they wrote with sharp objects such as metal sticks or bones. Around 6000 years ago, the Egyptians invented the first paperlike material called papyrus. The word “paper” actually comes from the word “papyrus”.
Another writing instrument that remained active in history for a long period was the quill (鹅毛) pen. Introduced around 700 AD, the quill was a pen made from a bird feather. Goose feathers were most common.
When writers had better inks and paper and handwriting had developed into both an art form and an everyday occurrence, man’s inventive nature once again turned to improving the writing instrument.
A.Now something was needed to write upon the papyrus. |
B.The papyrus became the most popular material at that time. |
C.Swan feathers were of a high quality, being rare and most expensive. |
D.Bamboo stems were better and much more expensive than goose feathers. |
E.This led to the development of the modern fountain pen in the 19th century. |
F.These were mostly the stems of grasses, especially from the bamboo plant. |
G.These drawings showed events in daily life like the planting of crops or hunting victories. |
【推荐3】Fewer than 400 North Atlantic right whales remain in the wild, and not even 100 of them are breeding females. Their biggest survival threats are boat strikes and getting caught in fishing gear. Protecting these whales, such as by turning boats from dangerous encounters, requires positioning them more reliably -- and new technology, described in the Journal of the Acoustical Society of America, could help make that possible.
To listen for marine life, researchers often install underwater microphones called hydrophones on buoys (浮标) and robotic gliders (滑翔机). The recorded audio is changed into spectrograms -- visual representations of sound used to precisely recognize, for instance, specific whale species’ calls. But those distinctive sounds are often drowned out by other noise. In recent years researchers have used a machine-learning technique called deep learning to automate this analysis, but background sounds still block reliability.
Now researchers have trained two deep-learning models specifically to cut through the noise. They started by giving the models thousands of “clean” spectrograms with only North Atlantic right whale calls. Then they slowly added in thousands of spectrograms mixed with typical background sounds, such as tanker engines. The program can successfully turn noisy spectrograms into clean ones, reducing false alarms and helping spot whales before they reach dangerous areas, the scientists say.
Shyam Madhusudhana, a Cornell University data engineer, who was not involved in the study, says he would want to see if such models could be used to locate other marine mammals (哺乳动物), too. “Humpback whales and dolphins have much more complex speech pathways than the right whale,” he notes. And University of East Anglia machine-learning researcher Ben Milner, one of the study’s authors, wants to take this technology above water as well -- to Ukrainian forests, where he hopes to identify animals near the site of the 1986 Chernobyl disaster.
University of St. Andrews behavioral ecologist Peter Tyack, who was not involved in the study, says this new system should be used to figure out where whales are throughout the year, so that these areas can be protected. “In terms of estimating the density and the number of these whales in places where it’s hard to see them,” Peter says, “this technology could be fantastic.”
But he warns that it should not be the only approach to preventing ship strikes. In his work, Tyack has found that North Atlantic right whales can be silent for hours at a time -- so passive sound monitoring could easily miss one. And killing just a few, he adds, “could lead to extinction of the population.”
1. What does the new technology do to help protect the North Atlantic right whales?A.Locating the right whales precisely. |
B.Turning boats to encounter the right whales. |
C.Guiding the right whales away from fishing boats. |
D.Positioning fishing boats reliably and thus deadly strikes. |
A.They turn off the nearby alarms that may create background noises. |
B.They can eliminate disturbances shown on a spectrogram sheet. |
C.They add thousands of mixed sounds such as vehicle engines. |
D.They can prevent whales from reaching dangerous areas. |
A.Shyam and Ben are both scientists at different universities who know each other well |
B.all of the three scientists consider the extended use of the technology could face challenges |
C.Shyam and Ben both hope that the deep learning model can go beyond its current application |
D.Peter thinks the technology should have been used to figure out where it’s hard to see the whales |
A.Latest Techs on Marine Lifesaving |
B.Save the Right Whale by Noise-cutting |
C.See Whales’ Noise through Human’s Eyes |
D.Non-professional Scientists’ Role in Saving Whales |
【推荐1】The olinguito(小型犬浣熊) is new to science, although it has been living in the cloud forests of South America for some time, unknown to researchers.
The olinguito is a hairy orange-brown creature with a sweet face and big eyes. The animal has small, rounded ears and lives in the trees. An adult weighs one kilogram and measures about 75 centimeters, with half of those centimeters taken up by its ringed tail.
Most of the time, it likes to eat fruit, although it also eats meat. Active at night, the animal has lived in Colombia and Ecuador for a long time. But the olinguito did not exist in science books before.
Kristofer Helgen is the curator (馆长) of mammals at the Smithsonian Institution’s National Museum of Natural History in Washington. He led the research team that confirmed the existence of the olinguito. It had been mistaken for a look-alike animal —the similar but larger olingo (尖吻浣熊).
“Think about these foggy and dense cloud forests. This animal, like so many animals in the rainforest, only comes out at night. It doesn’t come out of the trees. It’s wet. It’s dark. It’s hard to find these animals.”
About 10 years of work led Kristofer Helgen from museum storage areas to cloud forests in South America. He got lucky when he communicated with a zoologist in Ecuador. The animal expert there made a short video that showed what scientists say is an olinguito in the trees. It confirms that the olinguito is different from the olingo. Mr. Helgen says even people who live near the olinguitos’ forests do not see it different from similar animals.
1. What can we learn about the olinguito from the text?A.It has a preference for meat. | B.It is much larger than the olingo. |
C.It was described in some science books. | D.It and the olingo are very similar in appearance. |
A.It is too small. | B.It only comes out at night. |
C.It spends all its time eating in the trees. | D.People’s eyesight there is not very good. |
A.The tree. | B.The video. | C.The animal. | D.The study. |
A.The Olinguito Has Unique Living Habits | B.Scientists Discover a New Animal |
C.Researchers Are Searching for the Olinguito | D.The Olinguito Is an Endangered Animal |
【推荐2】Blood donations save lives. But blood can only be stored under refrigeration (冷藏) for up to six weeks. After that, it’s no longer usable for transfusions(输血).
“Because of that limitation, people have to continually donate blood to meet the needs. But also, in places where refrigeration may not be available, that can also be a challenge. It’s difficult to have blood available when needed.” University of Louisville bioengineer Jonathan Kopechek says. He also says disruptions to regular blood donations due to COVID-19 have put stress on the blood supply, and COVID-19 strengthens the need for more reliable long-term storage methods.
But it’s pretty rare because of all the challenges and complexities with that process. Instead Kopechek’s team has developed a method of preserving blood so it can be stored in a dehydrated(脱水的) state at room temperature. To do so, they turned to an unusual preservative(防腐剂): a sugar called trehalose, which is a common ingredient(成分) in donuts(甜甜圈)to help make them look fresh even when they might be months old, and people wouldn’t know the difference.
At that point, the blood could be dried and made into a powder(粉末). And then we can rehydrate(再水化) the blood and have it return back to normal. The team is still trying to improve yields but thinks the dried blood could be stored at room temperature for years. Kopechek says the technique could be ready for clinical trials in three to five years. If successful, it could be used to create stores of dried blood in case of future pandemics or natural disasters, and for first aid work, military operations or even missions to Mars. Maybe first aid kits on the Red Planet will include dried red blood cells.
1. Why we need a more reliable method to preserve blood?A.Refrigeration can’t be afforded. |
B.The cost to store blood is very high. |
C.People can’t donate their blood for COVID-19. |
D.Blood can only be stored under refrigeration for a short time. |
A.Trehalose. | B.Donuts. |
C.Powder. | D.The new technique. |
A.It is successful now. | B.It has been put into use. |
C.It will be widely used. | D.It has been researched for 3 to 4 years. |
A.How to store blood. |
B.Shortage of refrigeration . |
C.A new technique of storing blood. |
D.A new technique used in many fields. |
【推荐3】If sticking with an exercise habit is a challenge for you, there may be an app for that. In a University of Pennsylvania trial, smartphone games helped subjects become more active.
Participants were randomly assigned to one of three versions of a game that recorded how many steps they took each week. The first version placed them in direct competition with others. In the second version, they worked as a group to gain points. In the third, players earned points on their own but were asked to choose a friend or family member to be their supporter. This person received an e-mail each week reporting on the player's performance in order to help cheer her or him on.
On average, all three groups took more steps than usual. That said, certain versions of the game worked better for people with different personality traits. For instance, players who were more outgoing and more motivated to persevere with their goals tended to accumulate more steps in the competitive mode. In contrast, quiet players responded well to the game whether it was competitive, cooperative, or supportive. A third type, made up of those who were prone to take risks with their health and safety (by not wearing a seat belt, for example), was not helped by the game at all.
So though not all of us benefit, exercise gamification can work—and work especially well when you keep your personality in mind as you choose from among the many available apps.
1. What can we learn from paragraph 2?A.Each participant chose their version of a game. |
B.The second version got the participants to cooperate to earn points. |
C.Friends helped the player in the third version earn more points. |
D.The first version placed participants in indirect competition. |
A.Cooperative mode. | B.Competitive mode. |
C.Supportive mode. | D.No mode. |
A.were likely to | B.were unwilling to |
C.were afraid to | D.were hesitant to |
A.Challenge your body with different game versions. |
B.Change your personality to suit different game versions. |
C.Choose an exercise app with your personality in mind. |
D.Cooperate with your family to accumulate more steps. |