A newly granted patent (number 11,290834) shows that Siri, running on various Apple devices, may one day respond to a speaker’s head pose. The patent is dubbed “determining head pose based on room reverberation.”
About the patent
The patent is designed to deal with the problem of Siri not being sure which person is talking to her/him. Or which Siri-equipped device is being addressed. As you’ve probably noticed, in some circumstances, it’s not clear to a digital assistant such as Siri if the user is intending to interact with the digital assistant or is rather speaking to another person or device.
For example, the user may speak an utterance in a room with an electronic device and multiple other people that could be directed at the digital assistant or one of the other people. In this situation, the digital assistant may be unable to determine from the utterance alone if it should respond to the user through the electronic device.
Apple says it would be “advantageous for the digital assistant to determine the user’s head pose (e.g., if the user is facing the electronic device) during the utterance and thus is more or less likely addressing the digital assistant.”
Summary of the patent
Here’s Apple’s abstract of the patent with some technical details: “Systems and processes for operating an intelligent automated assistant are provided. An examples process of operating an intelligent automated assistant includes, at an electronic device with one or more processors and memory, receiving audio input, determining a direct-to-reverberant energy ratio based on the audio input, and determining a head pose of a user based on the direct-to-reverberant energy ratio.”
