US 12,069,425 B2
Separation of self-voice signal from a background signal using a speech generative network on a wearable device
Lae-Hoon Kim, San Diego, CA (US); Dongmei Wang, Bellevue, WA (US); Fatemeh Saki, San Diego, CA (US); Taher Shahbazi Mirzahasanloo, San Diego, CA (US); Erik Visser, San Diego, CA (US); and Rogerio Guedes Alves, Macomb Township, MI (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by Qualcomm Incorporated, San Diego, CA (US)
Filed on Jul. 10, 2023, as Appl. No. 18/349,920.
Application 18/349,920 is a continuation of application No. 18/063,493, filed on Dec. 8, 2022, granted, now 11,743,631.
Application 18/063,493 is a continuation of application No. 17/201,998, filed on Mar. 15, 2021, granted, now 11,589,153, issued on Feb. 21, 2023.
Application 17/201,998 is a continuation of application No. 16/896,010, filed on Jun. 8, 2020, granted, now 10,951,975, issued on Mar. 16, 2021.
Application 16/896,010 is a continuation of application No. 16/285,923, filed on Feb. 26, 2019, granted, now 10,681,452, issued on Jun. 9, 2020.
Prior Publication US 2023/0353929 A1, Nov. 2, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. A61F 11/06 (2006.01); G10K 11/16 (2006.01); H03B 29/00 (2006.01); H04R 1/10 (2006.01)
CPC H04R 1/1083 (2013.01) [H04R 1/1075 (2013.01); H04R 2420/07 (2013.01); H04R 2460/01 (2013.01); H04R 2460/13 (2013.01)] 27 Claims
OG exemplary drawing
 
1. A wearable device, the wearable device comprising:
a memory configured to store a self-voice signal via one or more transducers; and
a processor coupled to the memory, configured to:
detect the self-voice signal, based on the one or more transducers;
separate the self-voice signal from a background signal in an external audio signal based on using a multi-microphone speech generative network;
apply a first filter to the external audio signal, detected by at least one external microphone on the wearable device, during a listen through operation based on an activation of an audio zoom feature to generate a first listen-through signal that includes the external audio signal; and
produce an output audio signal that is based on at least the first listen-through signal that includes the external audio signal, and is based on the detected self-voice signal.