Information processing device

Abstract

Display an appropriate message to the vehicle's occupants. [Solution] The information processing device comprises: message acquisition means for acquiring a predetermined message to be output to the occupants of a vehicle; estimation means for estimating situation variables that indicate a situation suitable for outputting the predetermined message using a machine learning-based text generation model; index means for storing the situation variables estimated by the estimation means in association with the predetermined message; situation acquisition means for acquiring situation information indicating the current situation; and output means for outputting the predetermined message, which is stored in association with situation variables that match the situation information, to the occupants of the vehicle.

Description

Technical Field

[0001] This disclosure relates to the technical field of information processing apparatuses.

Background Art

[0002] As this type of apparatus, there is known one that generates a message for a user by a machine-learned text generation model. For example, Patent Document 1 discloses a technique for notifying at least a part of answers inferred by a learning model in a ranking format with respect to an inquiry from a user.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When outputting a message to a vehicle occupant, if trying to newly generate a message from the current situation, the content is likely to become stale, and it is also difficult to guarantee that there is no lie in the message. Also, in the method of preparing a message in advance, it is difficult to appropriately determine under what circumstances to present the message.

[0005] This disclosure has been made in view of the above problems, and an object thereof is to provide an information processing apparatus capable of outputting an appropriate message to a vehicle occupant.

Means for Solving the Problems

[0006] An information processing device according to one aspect of this disclosure includes: message acquisition means for acquiring a predetermined message to be output to a passenger of a vehicle; estimation means for estimating situation variables indicating a situation suitable for outputting the predetermined message using a machine learning-based text generation model; index means for storing the situation variables estimated by the estimation means in association with the predetermined message; situation acquisition means for acquiring situation information indicating the current situation; and output means for outputting the predetermined message, which is stored in association with the situation variables that match the situation information, to the passenger of the vehicle. [Brief explanation of the drawing]

[0007] [Figure 1] This is a block diagram showing the configuration of the information processing device according to the embodiment. [Figure 2] This is a flowchart showing the flow of index operation by the information processing device according to the embodiment. [Figure 3] This is a flowchart showing the flow of message output operation by the information processing device according to the embodiment. [Modes for carrying out the invention]

[0008] The following describes an embodiment of the information processing device with reference to the drawings. In the following description, the information processing device is assumed to be configured as a device mounted on a vehicle.

[0009] (Device configuration) First, the configuration of the information processing apparatus according to the embodiment will be described with reference to Figure 1. Figure 1 is a block diagram showing the configuration of the information processing apparatus according to the embodiment.

[0010] In Figure 1, the information processing device 10 according to the embodiment is configured as a device that outputs a message according to the occupants of a vehicle. The information processing device 10 may be configured to include an arithmetic unit (e.g., a processor), a storage device (e.g., RAM (Random Access Memory), ROM (Read Only Memory), a hard disk drive, etc.), a communication device (e.g., a wired communication device or a wireless communication device), an input device (e.g., a keyboard, mouse, touch panel, etc.), and an output device (e.g., a display, speaker, etc.).

[0011] The information processing device 10 includes, as components for realizing its functions, a message acquisition unit 110, a situation variable estimation unit 120, a message index unit 130, a situation information acquisition unit 140, and a message output unit 150. Note that the message acquisition unit 110, the situation variable estimation unit 120, the situation information acquisition unit 140, and the message output unit 150 may each be processing blocks implemented by an arithmetic unit. Furthermore, the message index unit 130 may be a database implemented by a storage device.

[0012] The message acquisition unit 110 is configured to acquire messages to be output to the vehicle's occupants from outside the device. Preferably, the messages acquired by the message acquisition unit 110 are highly reliable messages (for example, messages guaranteed not to contain factual inaccuracies). For this reason, the message acquisition unit 110 may be configured to acquire messages from highly reliable sources.

[0013] The situation variable acquisition unit 120 is configured to estimate situation variables from messages acquired by the message acquisition unit 110. Situation variables are variables that indicate a situation suitable for outputting a message. Situation variables may include multiple values ​​corresponding to each of multiple types of situations. Situation variables may include variables related to, for example, season, time of day, weather, etc. Alternatively, situation variables may include variables related to the vehicle's condition (for example, the vehicle's driving range, etc.).

[0014] The situational variable acquisition unit 120 is configured to estimate situational variables using a pre-trained text generation model 200. The text generation model 200 may be, for example, a model trained on general Japanese language knowledge. The text generation model 200 may be configured to take a message and candidate situational variables as input and output situational variables appropriate to the message. For example, suppose the message acquisition unit has acquired the message "The weather is likely to worsen this afternoon, so please don't forget your umbrella." In this case, the situational variable acquisition unit 120 inputs the following sentence to the text generation model 200: "Which season does the following message fit? 'The weather is likely to worsen this afternoon, so please don't forget your umbrella.' 1. Spring, 2. Summer, 3. Autumn, 4. Winter, 5. Season-independent." The text generation model then outputs, for example, "5. Season-independent." When estimating multiple types of situational variables, the situational variable acquisition unit 120 can input sentences like the above into the text generation model 200 for each type of situational variable.

[0015] The message index unit 130 is configured to store messages acquired by the message acquisition unit 110 and situation variables estimated by the situation variable acquisition unit 120 in a linked manner. The message index unit 130 is configured to store multiple pairs of messages and situation variables.

[0016] The status information acquisition unit 140 is configured to acquire status information indicating the current situation. The status information is information that can determine the degree of match with the status variables stored in the message index unit 130. For example, if the status variables stored in the message index unit 130 relate to season, time of day, and weather, the status information acquisition unit 140 acquires status information relating to at least one of the season, time of day, and weather.

[0017] The message output unit 150 is configured to be able to search for a message whose associated stored status variable matches the status information acquired by the status information acquisition unit 140 from the messages stored in the message index unit 130. Further, the message output unit 150 is configured to be able to output to the vehicle occupant a message stored in association with a status variable that matches the status information acquired by the status information acquisition unit 140. For example, the message output unit 150 may output a message stored in association with the status variable having the highest degree of match. Alternatively, the message output unit 150 may output a message stored in association with a status variable whose degree of match is equal to or higher than a predetermined threshold value. In this case, the message output unit 150 may output a plurality of messages. The message output unit 150 may output a message using, for example, a display, a speaker, or the like.

[0018] (Index operation) Next, referring to FIG. 2, the flow of the index operation (that is, the operation when associating and storing a message and a status variable) by the information processing apparatus 10 according to the embodiment will be described. FIG. 2 is a flowchart showing the flow of the index operation by the information processing apparatus according to the embodiment.

[0019] As shown in FIG. 2, when the index operation by the information processing apparatus 10 according to the embodiment is started, first, the message acquisition unit 110 acquires a message to be output to the vehicle occupant (step S101). Note that the message acquisition unit 110 may acquire a plurality of messages collectively.

[0020] Subsequently, the status variable estimation unit 120 estimates a status variable suitable for outputting the message acquired by the message acquisition unit 110 (step S102). That is, the status variable estimation unit 120 inputs the message acquired by the message acquisition unit 110 to the text generation model 200 and acquires the status variable output from the text generation model 200.

[0021] Subsequently, the message index unit 130 associates and stores the message acquired by the message acquisition unit 110 with the situation variable estimated by the situation variable estimation unit 120 (step S103).

[0022] The above-described indexing operation may be repeatedly executed. By repeatedly executing the indexing operation, a plurality of messages (specifically, messages associated with situation variables) are accumulated in the message index unit 130. The indexing operation may be executed even when the vehicle is not in use.

[0023] (Message output operation) Next, referring to FIG. 3, the flow of the message output operation (that is, the operation when outputting a message that matches the situation information) by the information processing apparatus 10 according to the embodiment will be described. FIG. 3 is a flowchart showing the flow of the message output operation by the information processing apparatus according to the embodiment.

[0024] As shown in FIG. 3, when the message output operation by the information processing apparatus 10 according to the embodiment is started, first, the situation information acquisition unit 140 acquires situation information indicating the current situation (step S201).

[0025] Subsequently, the message output unit 150 searches for a message stored in association with a situation variable that matches the situation information acquired by the situation information acquisition unit 140 (step S202). Then, the message output unit 150 outputs the message stored in association with the situation variable that matches the situation information to the vehicle occupants (step S203).

[0026] Note that the above-described message output operation may be automatically started when there is an occupant in the vehicle. Alternatively, the message output operation may be started in response to a user's query or input operation or the like.

[0027] (Technical effect) Next, the technical effect obtained by the information processing apparatus 10 according to the embodiment will be described.

[0028] As explained in Figures 1 to 3, the information processing device 10 according to this embodiment can output appropriate messages to the vehicle occupants according to their current situation. In this embodiment, in particular, the situation variables corresponding to the messages are estimated by a machine learning-based text generation model 200. Therefore, compared to determining the situation variables manually, costs, effort, and the occurrence of omissions can be reduced.

[0029] This disclosure is not limited to the embodiments described above and can be modified as appropriate without contradicting the gist or idea of ​​the invention as can be inferred from the claims and the specification as a whole. Information processing devices with such modifications are also included within the technical scope of the present invention. [Explanation of Symbols]

[0030] 10... Information processing device, 110... Message acquisition unit, 120... Situation variable estimation unit, 130... Message index unit, 140... Situation information acquisition unit, 150... Message output unit, 200... Text generation model

Claims

[Claim 1] A message acquisition means for acquiring a predetermined message to be output to the occupants of a vehicle, An estimation means for estimating situational variables that indicate a situation suitable for outputting the predetermined message, using a machine learning-based text generation model, An index means for storing the situation variables estimated by the estimation means in association with the predetermined message, A means for acquiring situational information that shows the current situation, Output means for outputting to the occupant of the vehicle a predetermined message, which is stored in association with the situation variable that matches the situation information, An information processing device equipped with the following features.

Images