This publication describes a computing device that captures a user-initiated sound by incorporating a microphone array. The computing device analyzes the captured sound from each microphone by using a model capable of determining a variation in the sound power of the captured sound signal by each microphone. To conserve the resources of the computing device, machine learning trains the model remotely or off-line. The ability of the computing device to determine the user-initiated sound power variation captured by each microphone enables the device to determine the location of the user in respect to the device and, thus, allows the device to respond accordingly (e.g., auto-rotate the screen towards to the user).

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.