Method, information processing device and system

Abstract

The invention relates to a method, an information processing apparatus, and a system. A technique for providing a driving environment corresponding to a user is improved. An information processing device (10) detects the occurrence of any event among one or more events on the basis of vehicle information acquired from a vehicle. Next, the information processing device (10) acquires user information and presentation generation information corresponding to the detected event. Next, the information processing device (10) uses the acquired user information and the cue generation information to generate a cue for causing the language model server (20) to output an action for the detected event. Next, the information processing device (10) inputs the presentation to the language model server (20). The information processing device (10) then executes an action based on the output of the language model server (20) for the presentation.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a method, an information processing apparatus, and a system. BACKGROUND

[0002] Conventionally, a technology of providing a driving environment corresponding to a user is known. For example, in Patent Literature 1, an invention of performing an action related to a vehicle for each user in a vehicle is disclosed.

[0003] PRIOR ART DOCUMENTS

[0004] PATENT LITERATURE

[0005] Patent Literature 1: Japanese Patent Application Publication No. 2023-055148 SUMMARY

[0006] PROBLEMS TO BE SOLVED BY THE INVENTION

[0007] There is room for improvement in the technology of providing a driving environment corresponding to a user.

[0008] The present disclosure was achieved in view of this situation, and aims to improve the technology of providing a driving environment corresponding to a user.

[0009] SOLUTION TO PROBLEM

[0010] The method of one embodiment of the present disclosure is a method executed by an information processing apparatus, in which the method includes:

[0011] detecting occurrence of any event of one or more events based on vehicle information acquired from a vehicle,

[0012] acquiring user information and prompt generation information corresponding to the detected event,

[0013] generating a prompt for causing a first language model to output an action for the detected event using the acquired user information and prompt generation information,

[0014] inputting the prompt to the first language model, and

[0015] performing an action based on output of the first language model for the prompt.

[0016] The information processing apparatus of one embodiment of the present disclosure includes a control unit that performs the following actions:

[0017] detects occurrence of any event of one or more events based on vehicle information acquired from a vehicle,

[0018] acquires user information and prompt generation information corresponding to the detected event,

[0019] generate a prompt for causing the first language model to output an action for the detected event using the acquired user information and the prompt generation information,

[0020] input the prompt to the first language model,

[0021] perform an action based on the output of the first language model for the prompt.

[0022] The system of one embodiment of the present disclosure is a system including an information processing device and a language model server storing a first language model, in which

[0023] The information processing device detects occurrence of any event of one or more events based on vehicle information acquired from a vehicle, acquires user information and prompt generation information corresponding to the detected event, generates a prompt for causing a first language model to output an action for the detected event using the acquired user information and the prompt generation information, and inputs the prompt to the first language model,

[0024] The language model server performs output of the first language model for the prompt,

[0025] The information processing device performs an action based on the output.

[0026] Effects of the Invention

[0027] According to one embodiment of the present disclosure, a technology of providing a driving environment corresponding to a user is improved. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 is a block diagram illustrating a schematic structure of a system of one embodiment of the present disclosure.

[0029] Figure 2 is a block diagram illustrating a schematic structure of an information processing device.

[0030] Figure 3 is a flowchart illustrating an action of an information processing device.

[0031] REFERENCE NUMERALS

[0032] 1, system; 10, information processing device; 11, communication unit; 12, output unit; 13, input unit; 14, storage unit; 15, control unit; 16, sensor unit; 17, imaging unit; 20, language model server; 30, content server; 40, network. DETAILED DESCRIPTION

[0033] Hereinafter, one embodiment of the present disclosure will be described.

[0034] (SUMMARY OF EMBODIMENTS)

[0035] REFERENCE Figure 1 A summary of the system 1 of the embodiment of the present disclosure will be described. The system 1 is provided with an information processing device 10, a language model server 20, and a content server 30. The information processing device 10, the language model server 20, and the content server 30 are connected to a network 40 such as the Internet and a mobile communication network in a communicable manner.

[0036] The information processing device 10 is, for example, an in-vehicle device mounted on a vehicle or a server that instructs a vehicle via the network 40. The information processing device 10 is capable of communicating with the language model server 20 and the content server 30 via the network 40.

[0037] The language model server 20 is, for example, a server that exists on a cloud and provides a first language model. The first language model is, for example, a large-scale language model (LLM) such as Chatgpt and Megatron-Turing Natural Language Generation (MT-NLG), but is not limited thereto, and any language model can be used. The language model server 20 receives a prompt input from the information processing device 10 via the network 40, and transmits an output for the prompt to the information processing device 10. In addition, the language model server 20 can be the information processing device 10.

[0038] The content server 30 is a server that provides a user with content such as music and video. The content server 30 is, for example, a streaming service such as YouTube (registered trademark), TikTok, Netflix, and Spotify, but is not limited thereto, and any streaming service can be used. In addition, the information processing device 10 can function as the content server 30.

[0039] First, a summary of the present embodiment will be described, and the details will be described later. The information processing device 10 detects occurrence of any event of one or more events based on vehicle information acquired from a vehicle. Next, the information processing device 10 acquires user information and prompt generation information corresponding to the detected event. Next, the information processing device 10 generates a prompt for causing the language model server 20 to output an action for the detected event using the acquired user information and prompt generation information. Next, the information processing device 10 inputs the prompt to the language model server 20. Then, the information processing device 10 performs an action based on an output of the language model server 20 for the prompt.

[0040] In conventional technologies that provide a user-appropriate driving environment, uniform actions are executed, making it difficult to consider various factors such as the user's state and driving conditions. In contrast, according to this embodiment, user information and prompt generation information corresponding to detected events are used to generate prompts, which are then input into a first language model. Actions based on the output of the first language model in response to these prompts are then executed. Therefore, for example, if there is a child among the passengers in the vehicle and the child is crying, the possibility of executing actions that consider the user's state, characteristics, or preferences in response to the event, such as playing a popular video that the child likes, is increased. Thus, according to this embodiment, the technology for providing a user-appropriate driving environment is improved in terms of increasing the possibility of executing actions that consider the user's state, characteristics, or preferences in response to an event.

[0041] Next, the structures of System 1 will be described in detail.

[0042] (Structure of an information processing device)

[0043] like Figure 2 As shown, the information processing device 10 includes a communication unit 11, an output unit 12, an input unit 13, a storage unit 14, a control unit 15, a sensor unit 16, and an imaging unit 17. It should be noted that in this embodiment, the information processing device 10 may also be composed of an in-vehicle device and a server for instructing the vehicle. For example, the output unit 12, input unit 13, sensor unit 16, and imaging unit 17 of the information processing device 10 may be located in the in-vehicle device, while the storage unit 14 and control unit 15 may be located in the server.

[0044] The communication unit 11 includes one or more communication interfaces connected to the network 40. These communication interfaces may correspond to, for example, mobile communication standards, wired LAN (Local Area Network) standards, or wireless LAN standards, but are not limited to these; they may correspond to any communication standard. In this embodiment, the information processing device 10 communicates with the language model server 20 and the content server 30 via the communication unit 11 and the network 40.

[0045] The output unit 12 includes one or more output devices for outputting information to the vehicle. These output devices may be, for example, a display that outputs image information, or a speaker that outputs sound information, but are not limited to these. Alternatively, the output unit 12 may also include an interface for connecting an external output device.

[0046] The input section 13 includes one or more input devices mounted on the vehicle that detect input operations by the user. The input devices are, for example, physical keys, electrostatic capacitance keys, a mouse, a touch panel, a touch screen provided integrally with the display of the output section 12, or a microphone, but are not limited thereto. Alternatively, the input section 13 can include an interface for connecting an external input device. In the present embodiment, the input section 13 can also be used to acquire the user's voice required to acquire vehicle information or user information.

[0047] Note that, in the present embodiment, the "user" includes not only the driver of the vehicle but also a passenger, such as a family member of the driver, such as a child. The passenger can also include a pet.

[0048] The storage section 14 includes one or more memories. Each memory included in the storage section 14 can function, for example, as a main storage device, an auxiliary storage device, or a cache memory. The storage section 14 stores arbitrary information for the operation of the information processing device 10. For example, the storage section 14 can store a system program, an application program, and embedded software, and the like. Note that, in the present embodiment, the storage section 14 stores user information. The storage section 14 can be provided in the in-vehicle device or in the server, and the entire storage section 14 can be constituted by the in-vehicle device and the server. In the case where the information processing device 10 functions as the content server 30, programs and data required for the content server 30 can also be stored. The storage section 14 can also store a second language model different from the first language model.

[0049] The control section 15 includes one or more processors, one or more programmable circuits, one or more dedicated circuits, or a combination thereof. The processor is, for example, a general-purpose processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), or a dedicated processor dedicated to a specific process, but is not limited thereto. The programmable circuit is, for example, an FPGA (Field-Programmable Gate Array), but is not limited thereto. The dedicated circuit is, for example, an ASIC (Application Specific Integrated Circuit), but is not limited thereto. The control section 15 controls the operation of the entire information processing device 10.

[0050] The sensor section 16 includes, for example, a steering angle sensor, an acceleration sensor, a vehicle speed sensor, a position sensor, a gyro sensor, and the like, which are provided in the vehicle to acquire vehicle information. The sensor section 16 can also correspond to, for example, a GPS (Global Positioning System).

[0051] The imaging section 17 includes one or more cameras mounted on the vehicle, which can image the surroundings or the interior of the vehicle. Note that in the present embodiment, the imaging section 17 can also be used to acquire an image of the user, which is required to acquire vehicle information or user information.

[0052] (Action flow of information processing device)

[0053] Reference Figure 3 The operation of the information processing device 10 of the present embodiment will be described.

[0054] S100: The control section 15 of the information processing device 10 detects the occurrence of any of one or more events on the basis of vehicle information acquired from the vehicle.

[0055] The "one or more events" refers to one or more arbitrary events that can occur from when the user gets on the vehicle to when the user gets off the vehicle. The one or more events can include, for example, crying of a fellow passenger (e.g., a child), distraction of the driver, and fatigue of the driver.

[0056] The "vehicle information" is arbitrary information that can be acquired by the vehicle-mounted devices such as the input section 13, the sensor section 16, and the imaging section 17 at predetermined intervals or in real time. The vehicle information can include, for example, an image, a sound, various states (position, vehicle speed, shift position, and the like) of the vehicle, and route information of the user, but is not limited thereto. Note that which vehicle information is acquired in each of the one or more events can be determined in advance.

[0057] S101: The control section 15 acquires user information and prompt generation information corresponding to the detected event.

[0058] The "user information" can include, for example, a state, a schedule, a driving history, a family structure, an age, a gender, and a hobby of the user, but is not limited thereto, and can include arbitrary information related to the user. In addition, the control section 15 can acquire the user information by referring to information managed by another application, such as schedule information of Microsoft Outlook. Note that which user information is acquired in each of the one or more events can be determined in advance.

[0059] In the "information of the state of the user", information of the state of the user at the current time point, such as the user is crying, is tired, is absent-minded, has a cold, and the like, information of the health state of the user, such as the user is ill, and the like, but also any information related to the state of the user other than the above can be included.

[0060] In the "information of the schedule of the user", information of the action, the sleeping time, and the like of the previous day, and the like, but also any information indicating the action of the user other than the above can be included.

[0061] In the "information of the driving history of the user", information of the driving time, the destination, and the like of the previous day, and the like of the user, but also any information indicating the driving history of the user other than the above can be included. Note that the information of the driving history of the user can also be a history of any information acquired by the in-vehicle device using the input section 13, the sensor section 16, the imaging section 17, and the like.

[0062] The "information of the family structure of the user" can include information of the wife or husband, the child, the parent, and the like of the user. In addition, information of a pet can also be included in the information of the family structure of the user.

[0063] In the "information of the hobby of the user", information of the food, the music, the interest, and the like that the user likes, but also any information related to the preference of the user other than the above can be included. In addition, the information of the hobby can include information related to things that the user dislikes. For example, information of the food, the music, and the like that the user dislikes can also be included in the information of the hobby.

[0064] In the "prompt generation information", an instruction to output the action for the event and an instruction to consider the acquired user information when deciding the action can also be included as a template of the prompt. The template of the prompt can also be stored in the storage section 14 in advance. Alternatively, the template of the prompt can also be created using the second language model stored in the storage section 14. Specifically, the control section 15 can input the information of the detected event to the second language model, and output from the second language model for the input can be used as the template of the prompt.

[0065] S102: The control section 15 generates a prompt for causing the language model server 20 to output the action for the detected event using the acquired user information and the prompt generation information.

[0066] Specifically, the control section 15 generates the prompt by embedding the acquired user information in the part considered when deciding the action of the template of the prompt. Alternatively, the control section 15 can input the information of the detected event and the acquired user information to the second language model, and output from the second language model for the input can be used as the prompt.

[0067] S103: The control section 15 inputs the prompt to the language model server 20.

[0068] S104: The control section 15 executes an action based on the output of the language model server 20 for the prompt.

[0069] The execution of the "action" refers to causing each device of the information processing apparatus 10 or the content server 30 to perform an action corresponding to the output of the language model server 20 for the prompt or outputting an execution command to each device of the information processing apparatus 10 or the content server 30. In the "action", for example, playing a specific music, playing a specific video, and showing a route guidance to a rest place such as a service area can be included, but any action that can be performed using each device of the information processing apparatus 10 or the content server 30 can be included without being limited thereto.

[0070] By executing the above steps, the action of the present embodiment is performed, but for a more specific action, a specific example is described below.

[0071] A first example in which the event is a child crying is described. In this case, the vehicle information used in S100 can also be an image and a sound of the child. The control section 15 detects that the child is crying from the image of the child taken by the imaging section 17. Alternatively, the control section 15 can detect the sound of the child crying using the microphone of the input section 13. Next, in S101, the control section 15 acquires the prompt generation information and the user information corresponding to the event that the child has cried, that is, the information of the state, the age, the gender, and the hobby of the child. Next, in S102, the control section 15 generates a prompt by adding the information of the state, the age, the gender, and the hobby of the child to the prompt generation information, for example, in the case where the prompt generation information is "The child is crying. Please consider the following matters to study countermeasures and execute." Next, in S103, the control section 15 inputs the prompt to the language model server 20. At this time, the various information added to the prompt generation information is set to the state of the child crying, the age of the child being 5 years old, the gender being male, and the hobby being very fond of animation programs. Then, the output of the language model server 20 for the prompt is, for example, "How about trying to play a video of an animation program A for children?". In this case, in S104, the control section 15 can connect to the content server 30 or search the storage section 14, acquire the video, and play the video using the display and the speaker of the output section 12, thereby executing the action.

[0072] Next, a case 2 of the event of the driver's fatigue is explained. In this case, the vehicle information used in S100 can also be the image of the driver, various states of the vehicle (position, vehicle speed, shift position, and the like), and the route information. For example, the control section 15 can also detect the driver's fatigue from the movement of the driver's body or eyes in the image of the driver taken by the imaging section 17 or from various states of the vehicle acquired by the sensor section 16. Next, in S101, the control section 15 acquires the prompt generation information and the user information corresponding to the event of the driver's fatigue, that is, the information of the driver's schedule and driving history. Next, in S102, the control section 15 generates a prompt by adding, for example, in the case where the prompt generation information is "The driver's fatigue. Please consider the following matters, and inform the position information of the rest point if you rest or continue driving without resting," the information of the driver's schedule, driving history, and the position and route information of the vehicle included in the vehicle information. Next, in S103, the control section 15 inputs the prompt to the language model server 20. At this time, the various information added in the prompt generation information is set to: the driver's schedule is 3 hours of sleep time yesterday, the driving history is a round trip from Tokyo to Osaka yesterday, the position of the vehicle is Yako, and the route information is the remaining distance of 340 km in the route to Nagoya. Also, it is assumed that the output of the language model server 20 with respect to the prompt is, for example, the position information of the Ikaruga parking area. In this case, in S104, the control section 15 can also perform the action by displaying the route from the current position of the vehicle to the Ikaruga parking area on the display of the output section 12. In addition, the control section 15 can also output a sound suggesting rest at the parking area using the speaker of the output section 12.

[0073] Alternatively, it is assumed that the various information added in the prompt generation information is: the driver's schedule is 7 hours of sleep time yesterday, the driving history is no driving history yesterday, the position of the vehicle is Yako, and the route information is the remaining distance of 7 km in the route to Takao Mountain. Also, it is assumed that the output of the language model server 20 with respect to the prompt is, for example, the content of the meaning that it is not a problem to continue driving without resting, but the driver needs to be reminded to pay attention. In this case, in S104, the control section 15 can also perform the action by outputting a message display or a sound reminding the driver of the lack of attention in driving using the display or the speaker of the output section 12.

[0074] As described above, the information processing apparatus 10 of the present embodiment detects the occurrence of any of one or more events on the basis of vehicle information acquired from the vehicle. Next, the information processing apparatus 10 acquires user information and prompt generation information corresponding to the detected event. Next, the information processing apparatus 10 generates a prompt for causing the language model server 20 to output an action with respect to the detected event using the acquired user information and prompt generation information. Next, the information processing apparatus 10 inputs the prompt to the language model server 20. Then, the information processing apparatus 10 performs an action on the basis of an output of the language model server 20 with respect to the prompt.

[0075] According to this structure, the user information and the prompt generation information corresponding to the detected event are used, the made prompt is input to the first language model, and an action on the basis of an output of the first language model with respect to the prompt is performed. Therefore, for example, in a case where a child of a passenger of the vehicle is crying, the possibility of performing an action such as playing a popular video of a content that the child likes with respect to the occurred event, which takes into account the state, characteristics, or preferences of the user and the like, is increased. Thus, according to the present embodiment, the technology of providing a driving environment corresponding to the user is improved in that the possibility of performing an action that takes into account the state, characteristics, or preferences of the user and the like with respect to the occurred event is increased.

[0076] The present disclosure is described on the basis of the drawings and the embodiments, but it should be noted by those skilled in the art that various modifications and changes can be made on the basis of the present disclosure. Thus, it should be noted that these modifications and changes are included in the scope of the present disclosure. For example, the functions and the like included in each of the constituent parts or each of the steps and the like can be rearranged in a logically non-contradictory manner, a plurality of constituent parts or steps and the like can be combined into one, or can be divided.

[0077] For example, in the above-described embodiments, an embodiment in which the structure and the action of the information processing apparatus 10 are dispersed to a plurality of computers capable of communicating with each other can also be possible.

[0078] In addition, for example, in the above-described embodiments, the control section 15 can also acquire a reaction of the user to the action performed in S104. In this case, the control section 15 can also update the user information on the basis of the reaction of the user.

[0079] Specifically, the control section 15 acquires a change in the in-vehicle environment related to the reaction of the user after the execution of S104 using any device of the information processing apparatus 10. For example, assume that the event is the above-described first example. In this case, after S104, the control section 15 acquires, for example, any change in the environment that is generally considered to be a negative reaction of the child, including but not limited to whether or not the child is crying, whether or not the driver or the fellow passenger has muted the volume of the speaker, or whether or not the driver or the fellow passenger has stopped the playback of the video, and the like. Then, in a case where a change in the environment that is considered to be negative is acquired, the control section 15 can also update the information indicating that the animation program A is not an object of preference in the information of the preference of the child stored in the storage section 14. On the other hand, in a case where a change in the environment that is considered to be negative is not acquired, the control section 15 can also update the information indicating that the animation program A is a program that is liked in the information of the preference of the child stored in the storage section 14.

[0080] In addition, for example, in the above-described modified example, in a case where the reaction of the user is negative, the control section 15 can also further use the reaction of the user to generate the prompt in S102 and execute S103 and S104 again.

[0081] For example, assume that the event is the above-described first example and a change in the environment that is considered to be negative is acquired. In this case, the control section 15 can also generate a prompt in S102 that not only appends the state of the child as crying, the age of the child as 5 years old, the gender as male, and the preference as very liking of animation programs in the prompt generation information, but also appends a prompt that dislikes the animation program A. Or, the control section 15 can also generate a prompt that includes an instruction to output an alternative.

[0082] In addition, for example, in the above-described embodiment, a priority level can also be included in each of one or more events, and in a case where two or more of the one or more events occur at the same time, the control section 15 can also further generate a prompt based on the priority level of each of the two or more events.

[0083] For example, assume that the above-described first example and the second example occur at the same time and the priority level of the second example is higher than that of the first example. In this case, the control section 15 can also give priority to the event of the second example and execute the process.

[0084] For example, the priority level can also be set in a numerical value form of an integer in an arbitrary range of 5 to 1 or the like in order from high to low. In addition, for example, the priority level can also not be a number, but can be set in an alphabetical-based ranking form of S, A, B, C, D, or the like in order from high to low, or can be a word of "high", "medium", "low", or the like. In addition, the priority level can also be changeable.

[0085] Further, for example, an embodiment in which a general-purpose computer functions as the information processing apparatus 10 of the above-described embodiment can also be possible. Specifically, a program that describes the processing content of each function of the information processing apparatus 10 of the above-described embodiment is stored in the memory of a general-purpose computer, and the program is read and executed by the processor. Thus, the present disclosure can also be realized as a program that can be executed by the processor or a non-volatile computer-readable medium that stores the program.

[0086] Hereinafter, a part of the embodiments of the present disclosure will be exemplified. However, it should be noted that the embodiments of the present disclosure are not limited to this.

[0087] [Postscript 1]

[0088] A method executed by an information processing apparatus, wherein the method comprises:

[0089] detecting occurrence of any event of one or more events based on vehicle information acquired from a vehicle;

[0090] acquiring user information and prompt generation information corresponding to the detected event;

[0091] generating a prompt for causing a first language model to output an action for the detected event using the acquired user information and prompt generation information;

[0092] inputting the prompt to the first language model; and

[0093] performing an action based on an output of the first language model for the prompt.

[0094] [Postscript 2]

[0095] The method according to Postscript 1, wherein the method further comprises:

[0096] making the prompt generation information corresponding to the detected event using a second language model stored in the information processing apparatus.

[0097] [Postscript 3]

[0098] The method according to Postscript 1 or 2, wherein,

[0099] the vehicle information includes an image of the user, a sound of the user, a state of the vehicle, and path information of the vehicle.

[0100] [Postscript 4]

[0101] The method according to any one of Postscripts 1 to 3, wherein,

[0102] the action includes:

[0103] playing specific music;

[0104] playing specific video; and

[0105] showing a path guide to a rest place.

[0106] [Para 5]

[0107] The method according to any one of Paras 1 to 4, wherein the method further comprises:

[0108] acquiring a reaction of the user to the performed action; and

[0109] updating the user information based on the reaction of the user.

[0110] [Para 6]

[0111] The method according to any one of Paras 1 to 5, wherein the method further comprises:

[0112] acquiring a reaction of the user to the performed action,

[0113] in a case where the reaction of the user is a negative reaction, the information processing apparatus further generates the prompt using the reaction of the user.

[0114] [Para 7]

[0115] The method according to any one of Paras 1 to 6, wherein,

[0116] the one or more events each include a priority,

[0117] in a case where two or more of the one or more events occur simultaneously, the information processing apparatus further generates the prompt based on the priority of each of the two or more events.

[0118] [Para 8]

[0119] An information processing apparatus, wherein the information processing apparatus comprises a control unit that performs the following actions:

[0120] detecting occurrence of any event of one or more events based on vehicle information acquired from a vehicle,

[0121] acquiring user information and prompt generation information corresponding to the detected event,

[0122] generating a prompt for causing a first language model to output an action for the detected event using the acquired user information and prompt generation information,

[0123] inputting the prompt to the first language model,

[0124] perform an action based on the output of the prompt based on the first language model.

[0125] [Para 9]

[0126] The information processing apparatus according to Para 8, wherein

[0127] The information processing apparatus further includes a storage unit that stores a second language model,

[0128] The control unit generates the prompt generation information corresponding to the detected event using the second language model.

[0129] [Para 10]

[0130] The information processing apparatus according to any one of Paras 8 to 9, wherein

[0131] The vehicle information includes an image of the user, a sound of the user, a state of the vehicle, and path information of the vehicle.

[0132] [Para 11]

[0133] The information processing apparatus according to any one of Paras 8 to 10, wherein

[0134] The action includes:

[0135] playing specific music;

[0136] playing specific video; and

[0137] showing a path guide to a rest place.

[0138] [Para 12]

[0139] The information processing apparatus according to any one of Paras 8 to 11, wherein

[0140] The control unit acquires a reaction of the user to the performed action,

[0141] The control unit updates the user information based on the reaction of the user.

[0142] [Para 13]

[0143] The information processing apparatus according to any one of Paras 8 to 12, wherein

[0144] The control unit acquires a reaction of the user to the performed action,

[0145] In a case where the reaction of the user is a negative reaction, the control section further generates the prompt for causing the first language model to output the action for the detected event using the reaction of the user.

[0146] [Para 14]

[0147] The information processing apparatus according to any one of Paras 8 to 13,

[0148] The one or more events each include a priority,

[0149] In a case where two or more of the one or more events occur simultaneously, the control section further generates the prompt based on the priority of each of the two or more events.

[0150] [Para 15]

[0151] A system including an information processing apparatus and a language model server storing a first language model,

[0152] The information processing apparatus detects occurrence of any event of one or more events based on vehicle information acquired from a vehicle, acquires user information and prompt generation information corresponding to the detected event, generates a prompt for causing a first language model to output an action for the detected event using the acquired user information and prompt generation information, and inputs the prompt to the first language model,

[0153] The language model server causes the first language model to output the prompt,

[0154] The information processing apparatus performs the action based on the output.

[0155] [Para 16]

[0156] The system according to Para 15,

[0157] The information processing apparatus uses a second language model stored in the information processing apparatus to create the prompt generation information corresponding to the detected event.

[0158] [Para 17]

[0159] The system according to Para 15 or 16,

[0160] The vehicle information includes an image of the user, a sound of the user, a state of the vehicle, and path information of the vehicle.

[0161] [Para 18]

[0162] The system according to any one of appendices 15 to 17, wherein

[0163] The action includes:

[0164] playing specific music;

[0165] playing specific video; and

[0166] showing a path guide to a rest place.

[0167] [Appendix 19]

[0168] The system according to any one of appendices 15 to 18, wherein

[0169] The information processing apparatus acquires a reaction of the user to the action performed,

[0170] The information processing apparatus updates the user information based on the reaction of the user.

[0171] [Appendix 20]

[0172] The system according to any one of appendices 15 to 19, wherein

[0173] The information processing apparatus acquires a reaction of the user to the action performed,

[0174] In a case where the reaction of the user is a negative reaction, the information processing apparatus further generates, using the reaction of the user, the prompt for causing the first language model to output the action for the detected event.

Claims

1. A method performed by an information processing device, wherein, The method includes: Based on vehicle information obtained from the vehicle, detect the occurrence of any one of more than one events; Acquire user information and generate prompts corresponding to the detected events; Using the acquired user information and the prompt generation information, a prompt is generated to cause the first language model to output an action in response to the detected event; Input the prompt into the first language model; and Perform actions based on the output of the first language model in response to the prompt.

2. The method according to claim 1, wherein, The method further includes: Using the second language model stored in the information processing device, the prompt generation information corresponding to the detected event is generated.

3. The method according to claim 1, wherein, The vehicle information includes the user's image, the user's voice, the vehicle's status, and the vehicle's route information.

4. The method according to claim 1, wherein, The actions include: Play specific music; Play specific videos; and The path to the rest area is shown.

5. The method according to claim 1, wherein, Also includes: Obtain the user's response to the action performed; as well as The user information is updated based on the user's response.

6. The method according to claim 1, wherein, Also includes: Obtain the user's response to the action performed. If the user's response is negative, the information processing device further uses the user's response to generate the prompt.

7. The method according to claim 1, wherein, Each of the one or more events has a priority. If two or more of the above events occur simultaneously, the information processing device further generates the prompt based on the priority of each of the above events.

8. An information processing apparatus, wherein, The information processing device includes a control unit that performs the following actions: Based on vehicle information obtained from the vehicle, detect the occurrence of any one of more than one events. Obtain user information and prompt generation information corresponding to the detected events. Using the acquired user information and the prompt generation information, a prompt is generated to cause the first language model to output an action in response to the detected event. The prompt is input into the first language model. Perform actions based on the output of the first language model in response to the prompt.

9. The information processing apparatus according to claim 8, wherein, It also has a storage unit for storing the second language model. The control unit uses the second language model to generate prompt information corresponding to the detected events.

10. The information processing apparatus according to claim 8, wherein, The vehicle information includes the user's image, the user's voice, the vehicle's status, and the vehicle's route information.

11. The information processing apparatus according to claim 8, wherein, The actions include: Play specific music; Play specific videos; and The path to the rest area is shown.

12. The information processing apparatus according to claim 8, wherein, The control unit acquires the user's response to the action performed. The control unit updates the user information based on the user's response.

13. The information processing apparatus according to claim 8, wherein, The control unit acquires the user's response to the action performed. If the user's response is negative, the control unit further uses the user's response to generate a prompt for the first language model to output the action in response to the detected event.

14. The information processing apparatus according to claim 8, wherein, Each of the one or more events has a priority. If two or more of the above events occur simultaneously, the control unit further generates the prompt based on the priority of each of the two or more events.

15. A system comprising an information processing device and a language model server storing a first language model, wherein, The information processing device, based on vehicle information obtained from the vehicle, detects the occurrence of any one of more than one events, acquires user information and prompt generation information corresponding to the detected event, uses the acquired user information and prompt generation information to generate a prompt for the first language model to output an action in response to the detected event, and inputs the prompt to the first language model. The language model server outputs the first language model in response to the prompt. The information processing device performs actions based on the output.

16. The system according to claim 15, wherein, The information processing device uses a second language model stored in the information processing device to generate prompt information corresponding to the detected event.

17. The system according to claim 15, wherein, The vehicle information includes the user's image, the user's voice, the vehicle's status, and the vehicle's route information.

18. The system according to claim 15, wherein, The actions include: Play specific music; Play specific videos; and The path to the rest area is shown.

19. The system according to claim 15, wherein, The information processing device acquires the user's response to the action performed. The information processing device updates the user information based on the user's response.

20. The system according to claim 15, wherein, The information processing device acquires the user's response to the action performed. If the user's response is negative, the information processing device further uses the user's response to generate a prompt for the first language model to output the action in response to the detected event.

Images