US 11,657,804 B2
Wake word detection modeling
Rohit Prasad, Acton, MA (US); Kenneth John Basye, Sutton, MA (US); Spyridon Matsoukas, Hopkinton, MA (US); Rajiv Ramachandran, Seattle, WA (US); Shiv Naga Prasad Vitaladevuni, Cambridge, MA (US); and Bjorn Hoffmeister, Seattle, WA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Nov. 5, 2020, as Appl. No. 17/90,716.
Application 17/090,716 is a continuation of application No. 15/641,169, filed on Jul. 3, 2017, granted, now 10,832,662.
Application 15/641,169 is a continuation of application No. 14/311,163, filed on Jun. 20, 2014, granted, now 9,697,828, issued on Jul. 4, 2017.
Prior Publication US 2021/0134276 A1, May 6, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G10L 15/18 (2013.01); G10L 15/08 (2006.01); G10L 15/30 (2013.01)
CPC G10L 15/18 (2013.01) [G10L 15/08 (2013.01); G10L 15/30 (2013.01); G10L 2015/088 (2013.01)] 20 Claims
OG exemplary drawing
1. A system comprising:
computer-readable memory storing executable instructions; and
one or more processors in communication with the computer-readable memory, wherein the one or more processors are programmed by the executable instructions to at least:
receive audio data from a computing device subsequent to the computing device determining a wake word was detected in a first portion of the audio data;
generate speech recognition result data using at least the first portion of the audio data;
generate acoustic data representing an acoustic property of a voice represented by the audio data;
generate feature data using the speech recognition result data and the acoustic data;
generate wake word detection data using a statistical wake word detection model trained to receive the feature data as input; and
determine, based on the wake word detection data, that the audio data fails to satisfy a detection criterion related to detecting a representation of the wake word.