公开(公告)号：US11948559B2

公开(公告)日：2024-04-02

申请号：US17700378

申请日：2022-03-21

Applicant: QUALCOMM Incorporated

Inventor： Yang Yang ,  Anusha Lalitha ,  Jin Won Lee ,  Christopher Lott

IPC: G10L15/187 ,  G06F3/16 ,  G10L15/02 ,  G10L15/06 ,  G10L15/07 ,  G10L15/19 ,  G10L15/22

CPC classification number: G10L15/187 ,  G06F3/167 ,  G10L15/02 ,  G10L15/063 ,  G10L15/07 ,  G10L15/19 ,  G10L15/22

Abstract: Various embodiments include methods and devices for implementing automatic grammar augmentation for improving voice command recognition accuracy in systems with a small footprint acoustic model. Alternative expressions that may capture acoustic model decoding variations may be added to a grammar set. An acoustic model-specific statistical pronunciation dictionary may be derived by running the acoustic model through a large general speech dataset and constructing a command-specific candidate set containing potential grammar expressions. Greedy based and cross-entropy-method (CEM) based algorithms may be utilized to search the candidate set for augmentations with improved recognition accuracy.

公开(公告)号：US20220335929A1

公开(公告)日：2022-10-20

申请号：US17700378

申请日：2022-03-21

Applicant: QUALCOMM Incorporated

Inventor： Yang YANG ,  Anusha Lalitha ,  Jin Won LEE ,  Christopher LOTT

IPC: G10L15/187 ,  G10L15/06 ,  G10L15/22 ,  G10L15/19 ,  G10L15/07 ,  G06F3/16 ,  G10L15/02

Abstract: Various embodiments include methods and devices for implementing automatic grammar augmentation for improving voice command recognition accuracy in systems with a small footprint acoustic model. Alternative expressions that may capture acoustic model decoding variations may be added to a grammar set. An acoustic model-specific statistical pronunciation dictionary may be derived by running the acoustic model through a large general speech dataset and constructing a command-specific candidate set containing potential grammar expressions. Greedy based and cross-entropy-method (CEM) based algorithms may be utilized to search the candidate set for augmentations with improved recognition accuracy.

公开(公告)号：US11282512B2

公开(公告)日：2022-03-22

申请号：US16665177

申请日：2019-10-28

Applicant: QUALCOMM Incorporated

Inventor： Yang Yang ,  Anusha Lalitha ,  Jin Won Lee ,  Christopher Lott

IPC: G10L15/187 ,  G10L15/06 ,  G10L15/22 ,  G10L15/19 ,  G10L15/07 ,  G06F3/16 ,  G10L15/02

Abstract: Various embodiments include methods and devices for implementing automatic grammar augmentation for improving voice command recognition accuracy in systems with a small footprint acoustic model. Alternative expressions that may capture acoustic model decoding variations may be added to a grammar set. An acoustic model-specific statistical pronunciation dictionary may be derived by running the acoustic model through a large general speech dataset and constructing a command-specific candidate set containing potential grammar expressions. Greedy based and cross-entropy-method (CEM) based algorithms may be utilized to search the candidate set for augmentations with improved recognition accuracy.