Dialogue system and dialogue processing method
AssigneesHyundai Motor Company, Kia Corporation
InventorsSungwang Kim, Jaemin Moon, Minjae Park
Life of this patent
Estimating the term…
Abstract
Provided is a dialogue system including: a speech recognizer module configured to convert a speech of a user into a plurality of candidate texts, and prioritize the plurality of candidate texts; a understanding module configured to determine a first action corresponding to a first candidate text with a highest priority among the plurality of candidate texts; and a controller configured to attempt to perform the determined first action, and when the first action is not performable, reprioritize other candidate texts of the plurality of candidate texts.
Claims16
A dialogue system, comprising:
- a speech recognizer module comprising at least one processor configured to execute a speech-to-text engine and convert a speech of a user into a plurality of candidate texts, and prioritize the plurality of candidate texts;
- an understanding module in communication with the speech recognizer module, the understanding module comprising the at least one processor configured to execute a natural language understanding engine and determine a first action corresponding to a first candidate text having a highest priority from among the plurality of candidate texts;
- a controller in communication with the understanding module and the speech recognizer module, the controller comprising the at least one processor configured to attempt to perform the first action as determined by the understanding module and generate a first action signal for performing the determined first action, and
- a communicator configured to transmit the generated first action signal to an external server or a vehicle,
- wherein if a failure signal is received from the external server or the vehicle indicating that an operation corresponding to the generated first action signal is not performable, the controller is further configured to reprioritize other candidate texts from among the plurality of candidate texts based on whether a completeness of a sentence of the speech of the user and whether an action corresponding with each of other candidate texts from among the plurality of candidate texts is performable.
A dialogue processing method, comprising:
- converting, by at least one processor, a speech of a user into a plurality of candidate texts;
- prioritizing, by the at least one processor, the plurality of candidate texts;
- determining, by the at least one processor, a first action corresponding to a first candidate text having a highest priority from among the plurality of candidate texts;
- attempting, by the at least one processor, to perform the determined first action;
- generating, by the at least one processor, a first action signal for performing the determined first action;
- transmitting, by a communicator, the generated first action signal to an external server or a vehicle;
- and
- reprioritizing, by the at least one processor, other candidate texts of the plurality of candidate texts if the first action is not performable,
- wherein, if a failure signal is received from the external server or the vehicle, the reprioritizing step further comprises reprioritizing the other candidate texts based on whether a completeness of a sentence of the speech of the user and whether an action corresponding with each of other candidate texts from among the plurality of candidate texts is performable, and
- wherein the failure signal indicates that an operation corresponding to the generated first action signal is not performable.
Dependencies are inferred from the claim text.