公开(公告)号：US20240371364A1

公开(公告)日：2024-11-07

申请号：US18642606

申请日：2024-04-22

Applicant: Meta Platforms Technologies, LLC

Inventor： Mohammad Keshavarzi ,  Biqiao Zhang ,  Yun Wang ,  Zhaojun Yang ,  Yangyang Shi ,  Varun Kumar Nagaraja ,  Gael Le Lan ,  Xin Lei ,  Sangeeta Srivastava ,  Ming Sun ,  Jason Dong Uk Kim ,  Li Wan ,  Yuguan Li ,  Qian Zhao ,  Liang Zhang ,  Peter Roberts

IPC: G10L15/08

Abstract: A system uses a machine learning model trained using multitask learning for processing audio signals of different types. The system may be used in a device such as a headset, smart glasses, or a smart watch, and may be used to make predictions based on different types of audio signals, for example, for classifying acoustic events based on audio signal or for keyword spotting to detect wake words. The use of a single machine learning model for analyzing different types of audio signals improves storage efficiency as well as energy efficiency of devices compared to systems that use a different machine learning model for processing each type of audio signal.