VISION-BASED PRESENCE-AWARE VOICE-ENABLED DEVICE

A method and apparatus for voice search. A voice search system for a voice-enabled device captures one or more images of a scene, detects a user in the one or more images, determines whether the position of the user satisfies an attention-based trigger condition for initiating a voice search operation, and selectively transmits a voice query to a network resource based at least in part on the determination. The voice query may include audio recorded from the scene and/or the one or more images captured of the scene. The voice search system may further determine whether the trigger condition is satisfied as a result of a false trigger and disable the voice-enabled device from transmitting the voice query to the network resource based at least in part on the trigger condition being satisfied as the result of a false trigger.