US 12,437,774 B2
Audio event analysis, classification, and detection system
Ajit Belsarkar, Lancaster, PA (US); Jacob A. Gallucci, Mechanicsburg, PA (US); and Irtsam Ghazi, Pittsburgh, PA (US)
Assigned to Robert Bosch GmbH, Stuttgart (DE)
Filed by Robert Bosch GmbH, Stuttgart (DE)
Filed on Nov. 9, 2022, as Appl. No. 18/054,015.
Prior Publication US 2024/0153526 A1, May 9, 2024
Int. Cl. G10L 25/57 (2013.01); H04N 7/18 (2006.01); H04N 23/69 (2023.01); H04R 1/02 (2006.01); H04R 1/04 (2006.01); H04R 1/40 (2006.01); H04R 3/00 (2006.01); H04R 29/00 (2006.01); G06V 20/40 (2022.01); G08B 21/18 (2006.01)
CPC G10L 25/57 (2013.01) [H04N 7/183 (2013.01); H04N 23/69 (2023.01); H04R 1/028 (2013.01); H04R 1/04 (2013.01); H04R 1/406 (2013.01); H04R 3/005 (2013.01); H04R 29/005 (2013.01); G06V 20/44 (2022.01); G06V 20/46 (2022.01); G06V 2201/10 (2022.01); G08B 21/18 (2013.01)] 18 Claims
OG exemplary drawing
 
1. An event detection system comprising:
a plurality of audio devices, each of the plurality of audio devices configured to be communicatively coupled to a server and including,
a first memory;
a first electronic processor configured to:
detect, via a microphone, audio;
determine an audio event within the audio;
receive an image from a camera; and
associate the image data with the audio event to generate event metadata; and
a parent device communicatively coupled to each of the audio devices, the parent device including a second memory, and a second electronic processor configured to:
receive the event metadata from each of the plurality of audio devices;
receive, from each audio device, a confidence level associated with the event metadata;
compare, for each event metadata, the confidence level associated with the event metadata to a threshold value;
add, for each event metadata and in response to the confidence level for the event metadata being greater than or equal to the threshold value, the event metadata to an aggregated event metadata;
ignore, for each event metadata and in response to the confidence level for the event metadata being less than the threshold value, the event metadata; and
transmit the aggregated event metadata to the server.