New patent shows Apple developing sound wave sensors to verify user voices

      New research shows that Apple is studying how to enable Siri to detect specific sounds and determine their location just by vibration. Two newly exposed patent applications show that Apple is studying different ways for devices to detect people or interact with people. The most important one is to enable Siri to recognize individual people and their spoken commands, and the device does not need a conventional microphone.

　　Apple's patent, titled "Self-mixing interferometry sensor for sensing vibrations of device external surface structures or housing components," involves the use of self-mixing interferometry (SMI). SMI involves the device detecting signals generated by reflection or backscattering of emitted light. Apple stated in the patent that with the improvement and increasing use of voice recognition, microphones have become increasingly important as input devices for interacting with devices. In traditional microphones, sound waves are converted into sound wave vibrations on the membrane of the microphone, which requires a port for air to flow in and out of the device under the microphone. The port makes the device susceptible to water damage, clogging, and moisture, and interferes with the appearance.

　　Therefore, Apple suggests using an SMI sensor array due to its higher sensitivity. SMI sensors can sense vibrations caused by sound and/or tapping a surface. Unlike traditional diaphragm-based microphones, SMI sensors can work in a closed (or sealed) environment. Details in the patent show how SMI sensors can also be used on the back of the Apple Watch. The device can be configured to sense one or more types of parameters, such as but not limited to vibration; light; touch; force; heat; motion; relative motion; biometric data of the user (such as biometric parameters); air quality; proximity; location; connectivity; and so on.

　　Apple describes how devices such as the Apple Watch can determine their location and nearby things through this patent. For example, after recognizing a specific human voice carried in a vibration waveform, its electronic display can be transitioned from a low-power or no-power state to an operating power state. So you can walk into the living room and ask your watch to turn on the TV. Even though the watch doesn't have a traditional microphone, it will recognize a verbal command. It will also specifically identify you. Knowing that you are authorized to use the TV and which TV is nearby, the device can turn on that TV.

　　Apple's proposal is to use a mix of different ways to detect a user's request, and even calculate the probability that the vibration is coming from a person. Such a device, whether a wearable or a static device like the Apple TV, will determine that the source of the vibration waveform is likely to be a person. It does this based on the information contained in the vibration waveform, including the determined direction or distance of the source. Such information will also include any changes in location, such as footsteps indicating that a person is moving to the intended viewing or listening position.