US 11,687,319 B2
Speech recognition method and apparatus with activation word based on operating environment of the apparatus
Sung-ja Choi, Seoul (KR); Eun-kyoung Kim, Suwon-si (KR); Ji-sang Yu, Seoul (KR); Ji-yeon Hong, Anyang-si (KR); Jong-youb Ryu, Hwaseong-si (KR); and Jae-won Lee, Seoul (KR)
Assigned to Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed by Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed on Mar. 29, 2021, as Appl. No. 17/215,409.
Application 17/215,409 is a continuation of application No. 15/783,476, filed on Oct. 13, 2017, granted, now 11,003,417.
Claims priority of application No. 10-2016-0171670 (KR), filed on Dec. 15, 2016; and application No. 10-2017-0054513 (KR), filed on Apr. 27, 2017.
Prior Publication US 2021/0216276 A1, Jul. 15, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G10L 15/22 (2006.01); G10L 15/26 (2006.01); G10L 15/08 (2006.01); G10L 17/24 (2013.01); G10L 15/30 (2013.01); G06F 3/16 (2006.01); G06N 20/00 (2019.01)
CPC G06F 3/167 (2013.01) [G06N 20/00 (2019.01); G10L 15/08 (2013.01); G10L 15/22 (2013.01); G10L 15/26 (2013.01); G10L 15/30 (2013.01); G10L 17/24 (2013.01); G10L 2015/228 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A speech recognition method comprising:
determining at least one activation word among a plurality of activation words based on information related to an operating environment in which a speech recognition apparatus is operating;
receiving an input audio signal;
performing speech recognition on the input audio signal, based on whether the input audio signal includes a speech signal of an utterance of an activation word included in the determined at least one activation word; and
outputting a result of the performing of the speech recognition,
wherein the performing of the speech recognition comprises:
extracting text of an utterance of a user by performing speech recognition on the input audio signal,
determining whether a speech command included in the input audio signal is a direct command or an indirect command based on natural language understanding and sentence structure analysis of the extracted text, wherein the direct command is speech uttered by the user with an intent for the speech recognition apparatus to output the result of the performing of the speech recognition, and wherein the indirect command is speech uttered by the user such that the speech recognition apparatus is unable to determine that the user intends for the speech recognition apparatus to output the result of the performing of the speech recognition,
when it is determined that the speech command is the direct command, performing an operation of responding to the speech command, and
when it is determined that the speech command is the indirect command:
determining whether a confirmation command is detected, and
in response to detecting the confirmation command from the user, performing the operation of responding to the speech command,
wherein the at least one activation word corresponds to executable functions of the speech recognition apparatus.