公开(公告)号：US20240071408A1

公开(公告)日：2024-02-29

申请号：US18243804

申请日：2023-09-08

Applicant: Amazon Technologies, Inc.

Inventor： Qingming Tang ,  Chieh-Chi Kao ,  Qin Zhang ,  Ming Sun ,  Chao Wang ,  Sumit Garg ,  Rong Chen ,  James Garnet Droppo ,  Chia-Jung Chang

IPC: G10L25/51 ,  G06N3/045 ,  G06N3/08 ,  G10L25/21 ,  G10L25/30

CPC classification number: G10L25/51 ,  G06N3/045 ,  G06N3/08 ,  G10L25/21 ,  G10L25/30 ,  G10L15/08

Abstract: A system may include a first acoustic event detection (AED) component configured to detect a predetermined set of acoustic events, and include a second AED component configured to detect custom acoustic events that a user configures a device to detect. The first and second AED components are configured to perform task-specific processing, and may receive as input the same acoustic feature data corresponding to audio data that potentially represents occurrence of one or more events. Based on processing by the first and second AED components, a device may output data indicating that one or more acoustic events occurred, where the acoustic events may be a predetermined acoustic event and/or a custom acoustic event.

公开(公告)号：US11790932B2

公开(公告)日：2023-10-17

申请号：US17547644

申请日：2021-12-10

Applicant: Amazon Technologies, Inc.

Inventor： Qingming Tang ,  Chieh-Chi Kao ,  Qin Zhang ,  Ming Sun ,  Chao Wang ,  Sumit Garg ,  Rong Chen ,  James Garnet Droppo ,  Chia-Jung Chang

IPC: G10L25/51 ,  G10L25/21 ,  G10L25/30 ,  G06N3/08 ,  G06N3/045 ,  G10L15/08 ,  G10L15/22

CPC classification number: G10L25/51 ,  G06N3/045 ,  G06N3/08 ,  G10L25/21 ,  G10L25/30 ,  G10L15/08 ,  G10L15/22 ,  G10L2015/088 ,  G10L2015/223

Abstract: A system may include a first acoustic event detection (AED) component configured to detect a predetermined set of acoustic events, and include a second AED component configured to detect custom acoustic events that a user configures a device to detect. The first and second AED components are configured to perform task-specific processing, and may receive as input the same acoustic feature data corresponding to audio data that potentially represents occurrence of one or more events. Based on processing by the first and second AED components, a device may output data indicating that one or more acoustic events occurred, where the acoustic events may be a predetermined acoustic event and/or a custom acoustic event.

公开(公告)号：US11961514B1

公开(公告)日：2024-04-16

申请号：US17547610

申请日：2021-12-10

Applicant: Amazon Technologies, Inc.

Inventor： Chia-Jung Chang ,  Qingming Tang ,  Ming Sun ,  Chao Wang

IPC: G10L15/16 ,  G10L15/14 ,  G10L17/16

CPC classification number: G10L15/16

Abstract: An acoustic event detection system may employ one or more recurrent neural networks (RNNs) to extract features from audio data, and use the extracted features to determine the presence of an acoustic event. The system may use self-attention to emphasize features extracted from portions of audio data that may include features more useful for detecting acoustic events. The system may perform self-attention in an iterative manner to reduce the amount of memory used to store hidden states of the RNN while processing successive portions of the audio data. The system may process the portions of the audio data using the RNN to generate a hidden state for each portion. The system may calculate an interim embedding for each hidden state. An interim embedding calculated for the last hidden state may be normalized to determine a final embedding representing features extracted from the input data by the RNN.

公开(公告)号：US20230186939A1

公开(公告)日：2023-06-15

申请号：US17547644

申请日：2021-12-10

Applicant: Amazon Technologies, Inc.

Inventor： Qingming Tang ,  Chieh-Chi Kao ,  Qin Zhang ,  Ming Sun ,  Chao Wang ,  Sumit Garg ,  Rong Chen ,  James Garnet Droppo ,  Chia-Jung Chang

IPC: G10L25/51 ,  G10L25/21 ,  G10L25/30 ,  G06N3/08 ,  G06N3/04

CPC classification number: G10L25/51 ,  G10L25/21 ,  G10L25/30 ,  G06N3/08 ,  G06N3/0454 ,  G10L15/22

Abstract: A system may include a first acoustic event detection (AED) component configured to detect a predetermined set of acoustic events, and include a second AED component configured to detect custom acoustic events that a user configures a device to detect. The first and second AED components are configured to perform task-specific processing, and may receive as input the same acoustic feature data corresponding to audio data that potentially represents occurrence of one or more events. Based on processing by the first and second AED components, a device may output data indicating that one or more acoustic events occurred, where the acoustic events may be a predetermined acoustic event and/or a custom acoustic event.