View source for Amazon Technologies, Inc. (20240412728). AUDIO DETECTION

=AUDIO DETECTION=


==Organization Name==

[[:Category:Amazon Technologies, Inc.|Amazon Technologies, Inc.]]


[[Category:Amazon Technologies, Inc.]]

==Inventor(s)==

[[:Category:Michael Thomas Peterson of Watertown MA (US)|Michael Thomas Peterson of Watertown MA (US)]][[Category:Michael Thomas Peterson of Watertown MA (US)]]

[[:Category:Gengshen Fu of Sharon MA (US)|Gengshen Fu of Sharon MA (US)]][[Category:Gengshen Fu of Sharon MA (US)]]

[[:Category:Aaron Challenner of Melrose MA (US)|Aaron Challenner of Melrose MA (US)]][[Category:Aaron Challenner of Melrose MA (US)]]

[[:Category:Rong Chen of Boston MA (US)|Rong Chen of Boston MA (US)]][[Category:Rong Chen of Boston MA (US)]]

[[:Category:Cody Jacques of Westwood MA (US)|Cody Jacques of Westwood MA (US)]][[Category:Cody Jacques of Westwood MA (US)]]

[[:Category:Stefan M Bradstreet of Medford MA (US)|Stefan M Bradstreet of Medford MA (US)]][[Category:Stefan M Bradstreet of Medford MA (US)]]

==AUDIO DETECTION==


This abstract first appeared for US patent application 20240412728 titled 'AUDIO DETECTION




==Original Abstract Submitted==

a device is configured to detect multiple different wakewords. a device may operate a joint encoder that operates on audio data to determine encoded audio data. the device may operate multiple different decoders which process the encoded audio data to determine if a wakeword is detected. each decoder may correspond to a different wakeword. the decoders may use fewer computing resources than the joint encoder, allowing for the device to more easily perform multiple wakeword processing. enabling/disabling wakeword(s) may involve the reconfiguring of a wakeword detector to add/remove data for respective decoder(s). specific decoders may be activated/deactivated depending on device context, thereby efficiently managing device resources.



[[Category:G10L15/22]]
[[Category:G10L15/16]]
[[Category:CPC_G10L15/22]]