US 12,277,928 B2
Automatic hotword threshold tuning
Aishanee Shah, Sunnyvale, CA (US); Alexander H. Gruenstein, Mountain View, CA (US); and Ian C. McGraw, Menlo Park, CA (US)
Assigned to Google LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Mar. 10, 2023, as Appl. No. 18/181,895.
Application 18/181,895 is a continuation of application No. 16/898,278, filed on Jun. 10, 2020, granted, now 11,610,578.
Prior Publication US 2023/0206908 A1, Jun. 29, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G10L 15/065 (2013.01); G10L 15/22 (2006.01)
CPC G10L 15/065 (2013.01) [G10L 15/22 (2013.01); G10L 2015/223 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A computer-implemented method executed on data processing hardware that causes the data processing hardware to perform operations comprising:
receiving a near-miss indication and audio data characterizing a hotword detected by a hotword detector in streaming audio captured by a user device, the near-miss indication indicating that the hotword detector detected the hotword in the streaming audio within a threshold period of time after generating a previous probability score that failed to satisfy a hotword detection threshold of the hotword detector, wherein the hotword detector is configured to:
generate a probability score indicating a presence of the hotword in audio features of the streaming audio captured by the user device; and
detect the hotword in the streaming audio when the probability score satisfies a hotword detection threshold, wherein the previous probability score indicates the presence of the hotword in previous audio features of the streaming audio captured by the user device;
identifying a false rejection instance at the hotword detector indicating that the hotword detector failed to detect the hotword in the previous audio features of the streaming audio; and
determining whether a false rejection rate associated with the hotword detector satisfies a false rejection rate threshold;
processing the audio data to confirm that the hotword was correctly detected by the hotword detector in the streaming audio; and
based on the near-miss indication, the confirmation that the hotword was correctly detected by the hotword detector in the streaming audio, and determining the false rejection rate associated with the hotword detector satisfies the false rejection rate threshold, adjusting the hotword detection threshold of the hotword detector.