# US 11,308,942 B2 · Method and apparatus for operating smart terminal > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,308,942 B2 - Document ID: US11308942B2 - Status: Granted - Kind: B2 - Application number: 16/505,667 - Filed: 2019-07-08 - Published: 2020-03-12 - Granted: 2022-04-19 - Earliest claimed: 2018-09-10 - Assignees: Baidu Online Network Technology (Beijing) Co., Ltd.; Shanghai Xiaodu Technology Co. Ltd. - Applicants: Baidu Online Network Technology (Beijing) Co., Ltd. - Inventors: Qin Xiong; Gaofei Cheng; Leilei Gao; Fei Wang; Yue Liu; Wenlin Dong - Web page: https://patentgrep.com/patent/US11308942B2 ## CPC - G10L 15/1822: {Parsing for meaning understanding} - G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue - G10L 15/083: {Recognition networks (G10L15/142, G10L15/16 take precedence)} - G10L 15/26: Speech to text systems (G10L15/08 takes precedence) - G10L 15/05: Word boundary detection - G10L 2015/088: {Word spotting} - G10L 2015/223: {Execution procedure of a spoken command} - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract Embodiments of the present disclosure relate to a method and apparatus for operating a smart terminal. The method for operating a smart terminal includes: parsing a currently inputted user voice command to a current text command; querying a historical text command record matching the current text command from a historical text command record set corresponding to a historical user voice command set; parsing the queried historical text command record to obtain a user's intention; sending a user request generated based on the user's intention to a resource database; and commanding an action component to take an action based on a response returned from the resource database and the user's intention. ## Claims ### Claim 1 A method for operating a smart terminal, comprising: parsing a currently inputted user voice command to a current text command; querying a historical text command record matching the current text command from a historical text command record set corresponding to a historical user voice command set by: performing word segmentation on the current text command to obtain a word sequence corresponding to the current text command; calculating a similarity between the word sequence and each historical text command record in the historical text command record set corresponding to the historical user voice command set; and acquiring a historical text command record having a highest similarity based on similarity in descending order, for use as the queried historical text command record; parsing the queried historical text command record to obtain a user's intention; sending a user request generated based on the user's intention to a resource database; and commanding an action component to take an action based on a response returned from the resource database, and the user's intention. ### Claim 2 (depends on 1) The method according to claim 1, wherein the querying a historical text command record matching the current text command from a historical text command record set corresponding to a historical user voice command set comprises: querying the historical text command record matching the current text command from the historical text command record set corresponding to the historical user voice command set using a maximum matching algorithm. ### Claim 3 (depends on 1) The method according to claim 1, wherein the historical user voice command is a user voice command stored within preset historical time. ### Claim 4 (depends on 1) The method according to claim 1, wherein the commanding an action component to take an action based on a response returned from the resource database and the user's intention comprises: prompting, if the response returned from the resource database indicates failure to parse a correct command, failure to parse the correct command to a user; and presenting, if the response returned from the resource database indicates the correct command resolved, the user with an operation indicated by the user's intention based on the resolved correct command. ### Claim 5 (depends on 1) The method according to claim 1, wherein the commanding an action component to take an action based on a response returned from the resource database and the user's intention comprises: prompting, if the response returned from the resource database indicates failure to parse a correct command, failure to parse the correct command to a user; and presenting, if the response returned from the resource database indicates the correct command resolved, the user with an operation indicated by the user's intention based on the resolved correct command. ### Claim 6 (depends on 2) The method according to claim 2 wherein the commanding an action component to take an action based on a response returned from the resource database and the user's intention comprises: prompting, if the response returned from the resource database indicates failure to parse a correct command, failure to parse the correct command to a user; and presenting, if the response returned from the resource database indicates the correct command resolved, the user with an operation indicated by the user's intention based on the resolved correct command. ### Claim 7 (depends on 3) The method according to claim 3, wherein the commanding an action component to take an action based on a response returned from the resource database and the user's intention comprises: prompting, if the response returned from the resource database indicates failure to parse a correct command, failure to parse the correct command to a user; and presenting, if the response returned from the resource database indicates the correct command resolved, the user with an operation indicated by the user's intention based on the resolved correct command. ### Claim 8 An apparatus for operating a smart terminal, comprising: at least one processor; and a memory storing instructions, wherein the instructions when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising: parsing a currently inputted user voice command to a current text command; querying a historical text command record matching the current text command from a historical text command record set corresponding to a historical user voice command set by: performing word segmentation on the current text command to obtain a word sequence corresponding to the current text command; calculating a similarity between the word sequence and each historical text command record in the historical text command record set corresponding to the historical user voice command set; and acquiring a historical text command record having a highest similarity based on similarity in descending order, for use as the queried historical text command record; parsing the queried historical text command record to obtain a user's intention; sending a user request generated based on the user's intention to a resource database; and commanding an action component to take an action based on a response returned from the resource database, and the user's intention. ### Claim 9 (depends on 8) The apparatus according to claim 8 wherein the querying a historical text command record matching the current text command from a historical text command record set corresponding to a historical user voice command set comprises: querying the historical text command record matching the current text command from the historical text command record set corresponding to the historical user voice command set using a maximum matching algorithm. ### Claim 10 (depends on 8) The apparatus according to claim 8 wherein the historical user voice command is a user voice command stored within preset historical time. ### Claim 11 (depends on 8) The apparatus according to claim 8 wherein the commanding an action component to take an action based on a response returned from the resource database and the user's intention comprises: prompting, if the response returned from the resource database indicates failure to parse a correct command, failure to parse the correct command to a user; and presenting, if the response returned from the resource database indicates the correct command resolved, the user with an operation indicated by the user's intention based on the resolved correct command. ### Claim 12 (depends on 8) The apparatus according to claim 8, wherein the commanding an action component to take an action based on a response returned from the resource database and the user's intention comprises: prompting, if the response returned from the resource database indicates failure to parse a correct command, failure to parse the correct command to a user; and presenting, if the response returned from the resource database indicates the correct command resolved, the user with an operation indicated by the user's intention based on the resolved correct command. ### Claim 13 (depends on 9) The apparatus according to claim 9, wherein the commanding an action component to take an action based on a response returned from the resource database and the user's intention comprises: prompting, if the response returned from the resource database indicates failure to parse a correct command, failure to parse the correct command to a user; and presenting, if the response returned from the resource database indicates the correct command resolved, the user with an operation indicated by the user's intention based on the resolved correct command. ### Claim 14 A non-transitory computer readable medium, storing a computer program thereon, wherein the program, when executed by a processor, implements operations, the operations comprising: parsing a currently inputted user voice command to a current text command; querying a historical text command record matching the current text command from a historical text command record set corresponding to a historical user voice command set by: performing word segmentation on the current text command to obtain a word sequence corresponding to the current text command: calculating a similarity between the word sequence and each historical text command record in the historical text command record set corresponding to the historical user voice command set; and acquiring a historical text command record having a highest similarity based on similarity in descending order, for use as the queried historical text command record; parsing the queried historical text command record to obtain a user's intention; sending a user request generated based on the user's intention to a resource database; and commanding an action component to take an action based on a response returned from the resource database, and the user's intention. ### Claim 15 (depends on 14) The non-transitory computer readable medium according to claim 14, wherein the querying a historical text command record matching the current text command from a historical text command record set corresponding to a historical user voice command set comprises: querying the historical text command record matching the current text command from the historical text command record set corresponding to the historical user voice command set using a maximum matching algorithm. ### Claim 16 (depends on 14) The non-transitory computer readable medium according to claim 14, wherein the historical user voice command is a user voice command stored within preset historical time. ### Claim 17 (depends on 14) The non-transitory computer readable medium according to claim 14, wherein the commanding an action component to take an action based on a response returned from the resource database and the user's intention comprises: prompting, if the response returned from the resource database indicates failure to parse a correct command, failure to parse the correct command to a user; and presenting, if the response returned from the resource database indicates the correct command resolved, the user with an operation indicated by the user's intention based on the resolved correct command.