公开(公告)号：US12283272B1

公开(公告)日：2025-04-22

申请号：US17379382

申请日：2021-07-19

Applicant: Amazon Technologies, Inc.

Inventor： Henry Michael D Souza ,  Vladimir Adam ,  Ketan Ashok Kulkarni ,  Oliver Benjamin Hill ,  Ragini Rajendra Prasad

IPC: G10L15/22 ,  G10L15/08 ,  G10L21/0208

Abstract: Techniques for processing utterance audio are described. In an example, a computer system determines audio data representing an utterance detected by a device, and generates, based at least in part the audio data, first data representing at least one of portion of the utterance in a frequency domain. The first data specific is to a first frequency range. The computer system determines determining a second frequency range that is a subset of the first frequency range, the second frequency range meeting a frequency threshold, and generates, based at least in part on the first data, second data that represents the at least one portion in the frequency domain. The second data is specific to the second frequency range. The computer system determines, based at least in part on the second data, that additional audio data associated with the device is to be processed.