A scientist from Cornell University created special glasses that can “listen” to you without even saying a word. The eyeglass attachment has tiny microphones and speakers built in, which is used to detect the words your mouth silently speaks when you want it to pause or skip songs on your device, enter passcodes without having to touch your phone, or even build CAD models without using a keyboard.
Ruidong Zhang, a student at Cornell University, designed a system which links to earlier projects done with camera and wireless earbud. With this new glasses form factor you don’t need face the camera anymore or wear something uncomfortable in your ear.
Cheng Zhang, an assistant professor of information science at Cornell, mentioned that previous technologies for silent-speech recognition only accept a certain set of commands and requiring users to face cameras which could be annoying. This time they are using sonar technology on the body instead.
Researchers have created a special system that can understand your speech patterns after only a few minutes of training (like reading some numbers). It sends and receives sound from your face to detect what your mouth is doing. Then, an advanced type of computer program called deep learning is used to understand the echoes with 95% accuracy.
This new system can transfer data wirelessly and make the accessory small so it is not noticeable. It can be used for up to 10 hours before needing a charge, and no information can be seen by anyone other than you. Cheng Zhang said they are very excited since this system is great at performance, size, and privacy. With this technology, it will let people use wearable items in everyday life.
Type in a correct email address.
Choose the newsletter you want to subscribe to.
Voice synthesis technology may be used in a variety of real-world applications. For instance, you could use it to control music playback without needing to use your hands or eyes in a quiet library. In addition, people with speech disabilities can use this technology by feeding their dialogue into it, and the system will speak out loud for them. Lastly, voice synthesis can also help you send messages at a loud concert where other methods don’t work.
The team at Cornell’s SciFi Lab is trying to make it possible for you to get your hands on a special kind of technology in the future. They’re using an official program from Cornell and are even testing smart-glasses that can recognize facial expressions, eye movements, and even upper body movements. According to Cheng Zhang, “We believe that glass will be an effective way to track people’s activities in everyday life”.