Abstract

Many devices can be networked to a central device (e.g., a so-called “smart speaker”) such that a user can control the devices via interaction with the central device. In one example, a user can interact with the central device using a voice command to request a certain action to be performed by a connected device. For example, the user can speak a hotword such as “Ok Assistant” and then continue with the command such as “turn on the downstairs lights.” However, the use of voice controls can be problematic due to word misrecognition and/or imprecise controls. In another example, certain devices may infer or otherwise estimate when certain actions should be taken. For example, machine learning is popularly used to infer when certain actions or operations should be performed by a device. However, machine learning or other algorithmic approaches are not always correct. Incorrect actions by a device can be frustrating for the user and, in some scenarios, users may simply prefer the ease and simplicity of manual control over device activities.

Creative Commons License

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

Share

COinS