公开(公告)号：US20230343351A1

公开(公告)日：2023-10-26

申请号：US17728334

申请日：2022-04-25

Applicant: Cisco Technology, Inc.

Inventor： Eric Y. Chen ,  Shamim S. Pirzada ,  Cullen Frishman Jennings

IPC: G10L21/0364 ,  G10L15/18 ,  G10L15/187 ,  G10L15/22 ,  G10L15/06 ,  G06V40/10 ,  G10L15/30 ,  G10L21/034 ,  G06V20/60 ,  G10L15/25

CPC classification number: G10L21/0364 ,  G10L15/1807 ,  G10L15/187 ,  G10L15/22 ,  G10L15/063 ,  G06V40/10 ,  G10L15/30 ,  G10L21/034 ,  G06V20/60 ,  G10L15/25 ,  G10L2021/03646

Abstract: In one example embodiment, audio characteristics of audio signals are adjusted by a first machine learning model to reduce effects of a facial covering and produce adjusted audio signals. The audio signals correspond to resulting voice signals produced from the facial covering affecting original voice signals. Speech characteristics are predicted for the adjusted audio signals by a second machine learning model. Transformed audio signals corresponding to the original voice signals are produced based on the adjusted audio signals and predicted speech characteristics.