| CPC G06F 16/90332 (2019.01) [B60K 35/60 (2024.01); G06F 40/35 (2020.01); G10L 15/26 (2013.01); B60K 35/10 (2024.01); B60K 35/265 (2024.01); B60K 35/28 (2024.01); B60K 2360/148 (2024.01); B60K 2360/164 (2024.01)] | 11 Claims |
|
1. An agent device comprising:
a memory; and
a processor coupled to the memory, the processor being configured to:
receive state information regarding a state of a vehicle from the vehicle, and receive a question from a user as question information,
perform inference processing for the received question information and the received state information in order to infer an intent of the question, and
acquire a response to the question based on the inferred intent,
wherein the processor is configured to perform the inference processing by inputting the received question information and the received state information into a trained model generated by employing past state information relating to a past state of the vehicle prior to a reception timing at which the question information was received, information regarding a plurality of expressions relevant to a question asked by the user at a time at which the past state existed, and correct response information for the plurality of expressions,
wherein the trained model is data generated by performing machine learning employing training data based on past states of the vehicle prior to a reception timing when the processor received text information, and on past occupant questions, and
wherein input of the training data is the past state information of the vehicle, and text information relating to plural expressions relevant to questions during these past states, and an intent label relating to a correct response to each of the questions is output.
|