Apple can teach Siri to read lips without using a camera
A recently disclosed Apple patent shows that the company is actively exploring how to add lip-reading to Siri, for example. And you don’t have to use a camera to do it.
The patent is titled “Keyword detection using a motion sensor”. The publication notes that adding the ability to recognize commands using gestures could improve Siri’s accuracy. Because often the voice assistant can pick up additional sounds, such as ambient noise or other people talking. While the new system will help to better recognize the user’s request and save the phone’s battery.
“When the user speaks, their mouth, face, head, and neck move and vibrate,” the patent says. “Motion sensors, such as accelerometers and gyroscopes, can detect these movements while using relatively little energy compared to audio sensors such as microphones.”
It is noted that this will work well with simple commands, such as “Hello, Siri” or “Next track”. If the user speaks a more complex phrase, the system can turn on the microphone in advance for better recognition.
Lipreading does not require the use of a camera, so it is likely that the technology will be integrated not only into iPhones but also into other Apple devices.
Earlier, we reported on other interesting patents of the tech company. For example, the Apple team has developed a “friction-resistant” surface that allows gadgets to be used without protective covers. Users can also customize the color of the Apple Watch band depending on their mood, clothing, weather, or other preferences.
As a reminder, Apple presented the Vision Pro augmented reality headset. This is the first “big” product created without Steve Jobs’ involvement. Its price unpleasantly surprised users, but the company is already working on a cheaper version.
