US 11,790,912 B2
Phoneme recognizer customizable keyword spotting system with keyword adaptation
Lakshmish Kaushik, San Mateo, CA (US); Zhenhao Ge, San Mateo, CA (US); and Xiaoyu Liu, Dublin, CA (US)
Assigned to Sony Interactive Entertainment Inc., Tokyo (JP)
Filed by Sony Interactive Entertainment Inc., Tokyo (JP)
Filed on Jan. 3, 2022, as Appl. No. 17/567,873.
Application 17/567,873 is a division of application No. 16/555,616, filed on Aug. 29, 2019, granted, now 11,217,245.
Prior Publication US 2022/0130384 A1, Apr. 28, 2022
Int. Cl. G10L 15/22 (2006.01); G10L 15/02 (2006.01); G10L 15/16 (2006.01); G10L 15/06 (2013.01); G06F 40/242 (2020.01); G10L 15/08 (2006.01)
CPC G10L 15/22 (2013.01) [G06F 40/242 (2020.01); G10L 15/02 (2013.01); G10L 15/063 (2013.01); G10L 15/16 (2013.01); G10L 2015/025 (2013.01); G10L 2015/088 (2013.01)] 18 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
at least one processor; and
at least one computer storage that is not a transitory signal and that comprises instructions executable by the at least one processor to:
register, using a first phoneme recognizer model, a wake-up word for a digital assistant based on recordings of a person speaking the wake-up word at least in part by adding first phoneme sequences to a dictionary accessible to the first phoneme recognizer model, the first phoneme sequences being derived from the recordings;
train the first phoneme recognizer model using the recordings of the person speaking the wake-up word to render a second phoneme recognizer model;
replace the first phoneme recognizer model with the second phoneme recognizer model;
again, register the wake-up word for the digital assistant based on the recordings but using the second phoneme recognizer model; and
update the dictionary by adding second phoneme sequences to the dictionary that are derived from the recordings using the second phoneme recognizer model.