US 11,670,290 B2
Speech signal processing method and apparatus
Tae Gyoon Kang, Seoul (KR)
Assigned to Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed by Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed on Nov. 30, 2020, as Appl. No. 17/106,599.
Claims priority of application No. 10-2020-0088919 (KR), filed on Jul. 17, 2020.
Prior Publication US 2022/0020362 A1, Jan. 20, 2022
Int. Cl. G10L 15/197 (2013.01); G10L 15/22 (2006.01); G06N 3/08 (2023.01); G10L 15/16 (2006.01); G06N 3/045 (2023.01)
CPC G10L 15/197 (2013.01) [G06N 3/045 (2023.01); G06N 3/08 (2013.01); G10L 15/16 (2013.01); G10L 15/22 (2013.01)] 23 Claims
OG exemplary drawing
 
1. A speech signal processing method, comprising:
receiving an input token that is based on a speech signal, the input token being a portion of a speech;
calculating, based on the input token, a plurality of first probability values respectively corresponding to a plurality of candidate output tokens, each first probability value indicates a probability that a respective candidate output token being a subsequent token to the input token;
adjusting, based on a priority of each first probability value, at least one of the first probability values; and
processing, based on the adjusted probability value, the speech signal.