Information notification device, notification pattern setting method, program, and notification pattern setting system

Abstract

To allow for notifying information tailored to the user's preferences without placing a significant burden on a user.SOLUTION: An information notification device comprises a controller and a communication unit. The controller generates a generation instruction prompt (630) by combining first text data (631) for instructing the generation of an information notification pattern, and second text data (632) for reflecting the preferences of a user (U1). Using the communication unit, the controller transmits a generation instruction prompt to an external device (2), which is a device provided outside the information notification device and includes a generative AI (5), and then receives AI output information (650) generated by the generative AI on the basis of the generation instruction prompt. The controller also sets an information notification pattern on the basis of the AI output information.SELECTED DRAWING: Figure 9

Description

[Technical Field]

[0001] The present invention relates to an information notification device, a notification pattern setting method, a program, and a notification pattern setting system. [Background technology]

[0002] There are devices that notify users of various types of information. Examples of information notification include left turn guidance notifications and right turn guidance notifications in car navigation systems. Information notification can be a warning to the user. It is often preferred to notify information in a pattern that matches the user's preferences, and a technology has been proposed that controls the warning method in consideration of the user's preferences (see Patent Document 1 below). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-160596 Summary of the Invention [Problem to be solved by the invention]

[0004] In order to have a device provide information notifications tailored to the user's preferences, it has been considered that the user himself / herself can set various information notification patterns for the device. However, such setting operations are complicated and place a heavy burden on the user. Furthermore, if the user is unfamiliar with the settings, there is a problem that the user cannot set them as desired.

[0005] An object of the present invention is to provide a technology that enables notification of information in accordance with a user's preferences without imposing a heavy burden on the user. [Means for solving the problem]

[0006] The information notification device according to the present invention includes a controller and provides information notification to a user. The controller creates a generation instruction prompt by combining first text data instructing the generation of the information notification pattern with second text data according to the user's preferences. The controller transmits the generation instruction prompt to an external device that is external to the information notification device and includes a generation AI, and then receives AI output information generated by the generation AI based on the generation instruction prompt. The controller then sets the information notification pattern based on the AI ​​output information. [Effects of the Invention]

[0007] With the information notification device, it is possible to set an information notification pattern that suits the user's preferences through the use of a generation AI, for example, by simply communicating the user's preferences to the information notification device, without requiring the user to perform complex setting operations. By using an existing generation AI, the configuration of the information notification device can be simple (the processing load on the controller itself can be light). [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 2 is a diagram illustrating the relationship between a user and other components according to an embodiment of the present invention. [Figure 2] 1 is a schematic block diagram of an in-vehicle system according to an embodiment of the present invention. [Figure 3] 1 is a diagram illustrating an internal configuration of an information notification device according to an embodiment of the present invention. [Figure 4] 1 is a diagram illustrating an internal configuration of a vibration device according to an embodiment of the present invention. [Figure 5] 10A and 10B are diagrams illustrating examples of a plurality of notifications that can be performed by an information notification device according to an embodiment of the present invention. [Figure 6] 3A and 3B are diagrams illustrating configurations of a basic table and a custom table stored on a recording medium according to an embodiment of the present invention. [Figure 7] 3 is a flowchart illustrating an operation of the information notification device according to the embodiment of the present invention. [Figure 8] 3A to 3C are diagrams illustrating examples of basic notification processing and custom notification processing performed by an information notification device according to an embodiment of the present invention. [Figure 9] FIG. 10 is a diagram showing the flow of operations related to the generation and setting of a custom pattern according to an embodiment of the present invention. [Figure 10] 10 is a diagram showing a specific example of prompts and the like in the operational flow of FIG. 9 according to the first example belonging to the embodiment of the present invention. [Figure 11] 10 is a diagram showing a specific example of prompts and the like in the operational flow of FIG. 9 according to the first example belonging to the embodiment of the present invention. [Figure 12] 10 is a flowchart of a notification pattern adjustment process according to a second example belonging to an embodiment of the present invention. [Figure 13] FIG. 10 is a diagram showing a specific example of setting a notification pattern according to a fourth embodiment of the present invention. [Figure 14] FIG. 13 is a functional block diagram of a controller in an information notification device according to a seventh example of an embodiment of the present invention. [Figure 15] FIG. 13 is a diagram showing a modified configuration of an information notification device according to an eighth embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, examples of embodiments of the present invention will be described in detail with reference to the drawings. In each of the drawings, the same parts are designated by the same reference numerals, and duplicate descriptions of the same parts will be omitted as a general rule. In this specification, for the sake of simplicity, symbols or signs referring to information, signals, physical quantities, functional units, circuits, elements, or components may be used, and the names of the information, signals, physical quantities, functional units, circuits, elements, or components corresponding to the symbols or signs may be omitted or abbreviated.

[0010] FIG. 1 shows the relationship between user U1 and other components assumed in an embodiment of the present invention. User U1 is an occupant of vehicle V1. User U1 is the driver of vehicle V1. However, user U1 may also be an occupant other than the driver (i.e., a passenger in vehicle V1). Vehicle V1 is any type of vehicle. Here, vehicle V1 is assumed to be an automobile or the like that runs on a road. An in-vehicle system 1 is mounted on vehicle V1, and each component of the in-vehicle system 1 is installed in an appropriate location in vehicle V1.

[0011] A seat ST1 is installed in the cabin of the vehicle V1. A user U1 sits in the seat ST1. Since it is assumed that the user U1 is the driver, the seat ST1 is the driver's seat. Hereinafter, when simply referring to the cabin, unless otherwise specified, this refers to the cabin of the vehicle V1. Also, below, unless otherwise specified, the inside of the vehicle refers to the internal area of ​​the vehicle V1, and the outside of the vehicle refers to the external area of ​​the vehicle V1. Figure 1 shows an information terminal TM carried by the user U1 located inside the cabin. The information terminal TM is a portable information terminal, such as a smartphone or tablet.

[0012] 2 shows a schematic block diagram of the in-vehicle system 1. The in-vehicle system 1 includes an information notification device 10, a cruise control device 20, an actuator unit 30, a vehicle sensor unit 40, a biosensor 60, an exterior camera 71, an interior camera 72, and an HMI 80. The components of the in-vehicle system 1 can transmit and receive any signals and information to and from each other through an in-vehicle network formed in the vehicle V1. The in-vehicle network includes, for example, a CAN (Controller Area Network) and an AVCLAN (Audio Visual Communication Local Area Network).

[0013] Some of the components of the in-vehicle system 1 are wirelessly connected to a communication network NET (see FIG. 1). At least the information notification device 10 is wirelessly connected to the communication network NET. The communication network NET includes the Internet, an intranet, etc. A server device 2 is connected to the communication network NET by wire or wirelessly. The server device 2 is an example of an external device provided outside the information notification device 10, and the information notification device 10 and the server device 2 form a notification pattern setting system.

[0014] The server device 2 is provided with a management device 3 and an artificial intelligence device 4. The information notification device 10 is capable of two-way communication with the server device 2 via a communication network NET. Specifically, two-way communication is possible between the management device 3 and the artificial intelligence device 4 via the communication network NET. However, communication between the information notification device 10 and the artificial intelligence device 4 may be performed via the management device 3. The artificial intelligence device 4 has a generation AI 5. In this specification, AI is an abbreviation for artificial intelligence (AI). The server device 2 is also provided with a database 6, which is a large-capacity recording medium. The database 6 may be a database connected to the communication network NET and provided external to the server device 2. The server device 2 is formed by any one or more computers connected to the communication network NET. The server device 2 may be formed using cloud computing. The computer constituting the management device 3 and the computer constituting the artificial intelligence device 4 may be separate computers. The management device 3 and the artificial intelligence device 4 may be included in a single computer. In the following description, any operation indicated as being performed by the server device 2 may be an operation performed by the management device 3 or the artificial intelligence device 4.

[0015] Each component shown in Fig. 2 will be explained. The information notification device 10 provides various notifications to the user U1 at the necessary timing. The driving control device 20 controls the driving of the vehicle V1 using an actuator unit 30. The actuator unit 30 has various driving components such as a motor that realizes the driving of the vehicle V1. Specifically, the actuator unit 30 includes an engine and a motor that generate driving force for the vehicle V1, a steering actuator that drives the steering of the vehicle V1, and a brake actuator that drives the brakes of the vehicle V1.

[0016] The vehicle sensor unit 40 has sensors that detect the details of the driving operation of the vehicle V1 by the driver of the vehicle V1 and sensors that detect various states of the vehicle V1. The vehicle sensor unit 40 outputs vehicle sensor information containing the detection results. The driving control device 20 realizes driving control of the vehicle V1 by driving and controlling the actuator unit 30 in accordance with the vehicle sensor information.

[0017] The biosensor 60 detects biometric information of the user U1. It is possible to estimate the emotion of the user U1 based on the detection result of the biometric information by the biosensor 60. Note that the biosensor 60 may be omitted from the in-vehicle system 1.

[0018] The exterior camera 71 consists of one or more cameras that capture images of the scene outside the vehicle V1. The exterior camera 71 has a capture area set outside the vehicle V1 and generates an exterior camera image by capturing images of the scene within the capture area. The exterior camera image is an image captured in the capture area of ​​the exterior camera 71. Image information showing the exterior camera image is referred to as exterior image information. Note that, in this specification, image information may be interpreted as an image signal or image data. The exterior camera 71 captures images at a predetermined frame rate. The capture area of ​​the exterior camera 71 includes all or part of the front area, rear area, right side area, and left side area of ​​the vehicle V1. The front area, rear area, right side area, and left side area of ​​the vehicle V1 are areas located in front, rear, right side, and left side of the vehicle V1, respectively, within the external area of ​​the vehicle V1.

[0019] The in-vehicle camera 72 consists of one or more cameras that capture images of the interior of the vehicle V1. The in-vehicle camera 72 has a capture area set inside the vehicle V1 (i.e., the interior of the vehicle V1), and generates an in-vehicle camera image by capturing images of the capture area. The in-vehicle camera image is an image captured in the capture area by the in-vehicle camera 72. Image information showing the in-vehicle camera image is called in-vehicle image information. The in-vehicle camera 72 captures images at a predetermined frame rate. User U1 is located within the capture area of ​​the in-vehicle camera 72, and therefore the in-vehicle camera image includes an image of user U1. Here, it is assumed that the face of user U1 is included in the capture area of ​​the in-vehicle camera 72, and therefore the in-vehicle camera image includes a facial image of user U1 (a captured image of user U1's face).

[0020] The HMI 80 is a human machine interface and includes a display device 81, a speaker 82, a microphone 83, an operation input unit 84, and a vibration device 85.

[0021] The display device 81 has a display screen such as a liquid crystal display panel, and displays any video (image) under the control of the information notification device 10 or a display control device (not shown). The display device 81 is installed in an appropriate location in the cabin of the vehicle V1 so that each occupant of the vehicle V1 can see the display content of the display device 81. Multiple display devices 81 may be installed in the cabin of the vehicle V1. The display device 81 may be a component of a car navigation system installed in the vehicle V1. The car navigation system may be included in the in-vehicle system 1. The display device 81 may be a display device provided in the information terminal TM.

[0022] The speaker 82 outputs any sound (message, warning sound, music, etc.) under the control of the information notification device 10 or an audio device (not shown). The speaker 82 is installed at an appropriate location in the cabin of the vehicle V1 so that each occupant of the vehicle V1 can hear the sound output from the speaker 82. Multiple speakers 82 may be installed in the cabin of the vehicle V1. The speaker 82 may be a speaker provided in the information terminal TM.

[0023] The microphone 83 picks up ambient sounds around its installation position and converts them into an electrical audio signal. The audio signal obtained by the conversion of the microphone 83 is called a microphone signal. The microphone 83 is installed at an appropriate location in the cabin of the vehicle V1 so that the signal components of the speech sounds of each occupant of the vehicle V1 are included in the microphone signal (i.e., so that the speech sounds are included in the content of the sound picked up by the microphone 83). Multiple microphones 83 may be installed in the cabin of the vehicle V1. The microphone 83 may be a microphone provided in the information terminal TM.

[0024] The operation input unit 84 receives arbitrary operations from each occupant of the vehicle V1. The operation input unit 84 can be configured with operation buttons, a touch panel, or the like. In this embodiment, attention is focused on a user U1 who is an occupant of the vehicle V1, and it is assumed that the user U1 inputs operations to the operation input unit 84. The operation input unit 84 may be an operation input unit provided in the information terminal TM.

[0025] The vibration device 85 generates vibrations under the control of the information notification device 10. The vibrations generated by the vibration device 85 are applied to the user U1. The installation position of the vibration device 85 is arbitrary as long as it can provide the desired vibration to the user U1, but here it is assumed that the vibration device 85 is installed in the seat ST1. Therefore, under the control of the information notification device 10, the vibration device 85 vibrates the seat ST1, and the vibrations are perceived by the user U1.

[0026] 3 shows the internal configuration of the information notification device 10. The information notification device 10 includes a controller 11, a memory 12, a communication unit 13, and a recording medium 14.

[0027] The controller 11 includes, as hardware resources, a processing unit including a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit), etc. The controller 11 may implement any function, operation, or process that should be implemented by the controller 11 by executing a program recorded in the memory 12 or any other recording medium.

[0028] The memory 12 is configured to include a non-volatile memory such as a ROM (Read Only Memory) or a flash memory, and a volatile memory such as a RAM (Random Access Memory). The memory 12 stores various data referenced by the controller 11 as well as various programs to be executed by the controller 11.

[0029] The communication unit 13 is a communication circuit (communication module) that transmits and receives any signal between the information notification device 10 and a counterpart device different from the information notification device 10. The communication unit 13 may be a communication circuit (communication module) provided outside the in-vehicle device 10. The communication circuit provided outside the in-vehicle device 10 may be a communication circuit shared by multiple components that make up the in-vehicle system 1. The communication unit 13 may be a communication circuit (communication module) provided in an information terminal (smartphone, etc.) brought into the vehicle V1. The counterpart device for the communication unit 13 includes components other than the information notification device 10 among the components of the in-vehicle system 1 shown in FIG. 2. The communication unit 13 can communicate with the counterpart device via an in-vehicle network formed in the vehicle V1. The counterpart device for the communication unit 13 includes an external device connected to the communication network NET, and therefore includes the server device 2. Note that the controller 11 can transmit and receive any information to and from the counterpart device using the communication unit 13, but the description of the communication unit 13 may be omitted below.

[0030] The recording medium 14 is a non-volatile recording medium made up of a magnetic disk, a flash memory, or the like, and stores (records) any information in a non-volatile manner. The controller 11 is capable of recording any information on the recording medium 14 and is also capable of reading any information recorded on the recording medium 14. The recording medium 14 may be external to the information notification device 10 and installed within the in-vehicle system 1. The recording medium 14 may also be installed external to the vehicle V1. In this case, the recording medium 14 is connected to a communication network NET, and the controller 11 accesses the recording medium 14 via the communication network NET.

[0031] As shown in FIG. 4, the vibration device 85 includes multiple vibration units 85U, each having a drive circuit 85a and a vibrator 85b. The multiple vibration units 85U have the same configuration. However, the total number of vibration units 85U may be one. The information notification device 10 can supply an acoustic signal Ain to each vibration unit 85U. The acoustic signal Ain supplied to the vibration unit 85U is hereinafter referred to as the input acoustic signal Ain. In each vibration unit U, the drive circuit 85a amplifies the input acoustic signal Ain and supplies the amplified input acoustic signal Ain, i.e., a drive acoustic signal Adrv, to the vibrator 85b. The vibrator 85b is an electro-acoustic transducer (vibration converter) formed of an electromagnetic circuit such as a magnet and a coil or a piezoelectric element, and converts the drive acoustic signal Adrv, which is an electrical signal supplied thereto, into mechanical vibration. That is, the vibrator 85b receives the drive acoustic signal Adrv and generates vibrations corresponding to the drive acoustic signal Adrv (and therefore generates vibrations corresponding to the input acoustic signal Ain). The vibration generated by the vibration device 85 (specifically, the vibration generated by the vibrator 85b) is transmitted to the user U1.

[0032] The information notification device 10 (controller 11) can adjust the vibration intensity of the vibrator 85b by adjusting the amplitude of the input acoustic signal Ain or the amplification factor of the drive circuit 85a for each vibration unit 85U. The information notification device 10 (controller 11) can selectively supply the acoustic signal Ain to only some of the vibration units 85U among the multiple vibration units 85U. The installation position of the vibrator 85b is arbitrary as long as it can provide the desired vibration to the user U1. Here, it is assumed that the vibrator 85b is installed in the seat ST1 to vibrate the seat ST1. The installation positions of the multiple vibrators 85b in the multiple vibration units 85U in the seat ST1 are different from each other. Therefore, the information notification device 10 (controller 11) can selectively vibrate any of multiple positions in the seat ST1 (any of the installation positions of the multiple vibrators 85b) using the multiple vibrators 85b. In this example, vibrations are output from the vibration unit 85U using the input acoustic signal Ain. However, the vibration unit 85U may be driven by a vibration signal obtained by performing appropriate signal processing on the input acoustic signal Ain (i.e., the vibration signal may be input to the vibration unit 85U). The signal processing may include, for example, a filter process that limits the frequency band of the input acoustic signal Ain, a frequency division process that reduces the frequency of the input acoustic signal Ain, or a frequency shift process. Specifically, the filter process may be a low-pass filter process that extracts low-frequency components of the input acoustic signal Ain (a frequency band in which people are more likely to perceive vibrations). The signal processing may provide a signal (drive signal) suitable for transmitting vibrations to the user U1.

[0033] In the information notification device 10, the controller 11 can perform information notification processing. In the information notification processing, the controller 11 provides any one of n types of notifications to the user U1. n is any integer equal to or greater than 1, but may normally be around several tens. Hereinafter, n is considered to represent an integer equal to or greater than 2. The first to nth types of notifications are referred to as the first to nth notifications. In the first to nth notifications, the first to nth information are notified to the user U1, respectively. That is, the i-th information is the information of the i-th notification (information notified to the user U1 in the i-th notification). i represents any integer. The first to n-th information are different from one another. Each notification in the first to n-th notifications is a notification such as a guidance, a warning, or a warning.

[0034] FIG. 5 shows a specific example of information to be notified to the user U1 in the information notification process. In this embodiment, for the sake of concrete explanation, the first to third notifications and the first to third information are as follows: That is, the first information is information that the vehicle V1 is approaching an intersection where it should turn left. The first notification is a left turn guidance notification that guides the user U1 that the vehicle V1 should turn left at the next intersection while the vehicle V1 is traveling toward its destination. The second information is information that the vehicle V1 is approaching an intersection where it should turn right. The second notification is a right turn guidance notification that guides the user U1 that the vehicle V1 should turn right at the next intersection while the vehicle V1 is traveling toward its destination. The third information is information that caution should be exercised against speeding. The third notification is a speeding notification that warns or calls attention to the fact that the speed of the vehicle V1 exceeds the speed limit or legal speed limit set for the road on which the vehicle V1 is traveling. In addition to the above, various other information can be notified to the user U1 in the information notification process.

[0035] Notification formats include audio output format, image output format, and vibration output format.

[0036] If the i-th notification is a notification in the form of audio output, the controller 11 performs the i-th notification to the user U1 by outputting a sound corresponding to the i-th notification from the speaker 82. That is, if the i-th notification is a notification in the form of audio output, the controller 11 conveys the information of the i-th notification to the user U1 through the hearing of the user U1. The sound corresponding to the i-th notification is the sound of words expressing the information of the i-th notification, or a sound (such as an instrument sound or a beep) that is different from words and is associated with the information of the i-th notification.

[0037] If the i-th notification is a notification in an image output format, the controller 11 performs the i-th notification to the user U1 by displaying an image corresponding to the i-th notification on the display device 81. In other words, if the i-th notification is a notification in an image output format, the controller 11 conveys the information of the i-th notification to the user U1 through the user U1's vision. The image corresponding to the i-th notification is an image of text representing the information of the i-th notification, or an image of a graphic associated with the information of the i-th notification. Graphics include illustrations and pictograms.

[0038] When the i-th notification is a notification in the form of vibration output, the controller 11 performs the i-th notification to the user U1 by causing the vibration device 85 to generate a vibration corresponding to the i-th notification. That is, when the i-th notification is a notification in the form of vibration output, the controller 11 transmits the information of the i-th notification to the user U1 through the user U1's sense of touch. The vibration corresponding to the i-th notification is a vibration associated with the information of the i-th notification.

[0039] Each notification is associated with pattern data, which defines the pattern of the corresponding notification.

[0040] When the i-th notification is in the form of audio output, the pattern of the i-th notification is the pattern of the sound (content of the sound) output from speaker 82 in the i-th notification, and the pattern data of the i-th notification is data indicating the pattern of the sound (content of the sound) output from speaker 82 in the i-th notification. When the first notification is in the form of audio output, the pattern data of the first notification is, for example, an acoustic signal of a sound representing the words "Please turn left at the next intersection."

[0041] When the ith notification is in the image output format, the pattern of the ith notification is the pattern of the image (image content) displayed on the display device 81 in the ith notification, and the pattern data of the ith notification is data indicating the pattern of the image (image content) displayed on the display device 81 in the ith notification. When the first notification is in the image output format, the pattern data of the first notification is, for example, an image signal of an image representing the words "Turn left at the next intersection."

[0042] When the i-th notification is in the form of vibration output, the pattern of the i-th notification is the vibration pattern (vibration content) of vibration device 85 in the i-th notification, and the pattern data of the i-th notification is data indicating the vibration pattern (vibration content) of vibration device 85 in the i-th notification. When the first notification is in the form of vibration output, the pattern data of the first notification is, for example, an acoustic signal for vibrating the left side portion of seat ST1 using vibration device 85 at a predetermined vibration intensity and vibration frequency.

[0043] 6(a) and 6(b), a basic table TBL1 and a custom table TBL2 are stored and set in the recording medium 14. Of these, the basic table TBL1 is already stored and set in the recording medium 14 in the initial state of the information notification device 10. That is, the basic table TBL1 is stored in the recording medium 14 in advance. The basic table TBL1 stores pattern data for each of the first to nth notifications. The pattern data stored in the basic table TBL1 may be particularly referred to as basic pattern data. Furthermore, a notification pattern based on the basic pattern data may be referred to as a basic pattern. The basic pattern data stored in the basic table TBL1 corresponding to the i-th notification will be referred to by the symbol "BS[i]." Hereinafter, the basic table TBL1 may be abbreviated as table TBL1, and the basic pattern data BS[i] may be abbreviated as pattern data BS[i] or data BS[i]. The initial state of the information notification device 10 corresponds to a state in which the notification pattern adjustment process described below has never been performed, and in the initial state of the information notification device 10, each notification is made using the basic pattern data.

[0044] By executing the notification pattern adjustment process described below, one or more pattern data are stored in the custom table TBL2. FIG. 6(b) shows how all pattern data for the first to nth notifications are stored in the custom table TBL2. The pattern data stored in the custom table TBL2 may be particularly referred to as custom pattern data. Furthermore, the notification pattern based on the custom pattern data may be referred to as a custom pattern. The custom pattern data stored in the custom table TBL2 corresponding to the i-th notification will be referenced by the symbol "CST[i]". Hereinafter, the custom table TBL2 may be abbreviated as table TBL2, and the custom pattern data CST[i] may be abbreviated as pattern data CST[i] or data CST[i].

[0045] In table TBL1 or TBL2, the pattern data corresponding to the i-th notification is an audio signal or an image signal that matches the format of the i-th notification. The audio signal as pattern data is stored in table TBL1 or TBL2 in the form of an audio file. The image signal as pattern data is stored in table TBL1 or TBL2 in the form of an image file. The custom pattern data is created according to the preferences or requests of user U1.

[0046] FIG. 7 shows an operation flowchart of the information notification device 10. Each process of steps S1 to S6 shown in FIG. 7 is executed by the controller 11. The controller 11 also starts when the vehicle V1 starts, and the operation of the controller 11 begins with the process of step S1. In step S1, the controller 11 determines whether an adjustment instruction has been input from the user U1 to the information notification device 10. The adjustment instruction is an instruction requesting adjustment of one or more notification patterns. Adjustment of a certain notification pattern involves changing, creating, or setting the notification pattern. The input of the adjustment instruction is realized by the user U1 inputting an adjustment instruction operation to the operation input unit 84. The input of the adjustment instruction may also be a voice operation input. In this case, the input of the adjustment instruction is realized by the user U1's speech sound indicating the content of the adjustment instruction being picked up by the microphone 83. When the user U1 inputs an adjustment instruction operation to the operation input unit 84, the controller 11 determines that an adjustment instruction has been input (Y in step S1) and causes a transition from step S1 to step S6. In step S6, the controller 11 performs a notification pattern adjustment process, and then returns to step S1. The notification pattern adjustment process will be described in detail later.

[0047] If it is determined in step S1 that an adjustment instruction has not been input to the information notification device 10, the process moves from step S1 to step S2. In step S2, the controller 11 determines which notification to output (in other words, execute) from among the first to n-th notifications.

[0048] The navigation operation performed by the car navigation system installed in the vehicle V1 determines whether a first notification (left turn guidance notification) or a second notification (right turn guidance notification) is necessary. Furthermore, the speed of the vehicle V1 is monitored during the navigation operation, and whether a third notification (speeding notification) is necessary is determined based on the result of comparing the speed of the vehicle V1 with the speed limit or legal speed set for the road on which the vehicle V1 is traveling. The navigation operation provides route guidance to guide the vehicle V1 to the destination set by the user U1. The navigation operation may be performed by the controller 11. Regarding the fourth to n-th notifications, whether each notification is necessary is also determined according to the content of each notification. The determination of whether each notification is necessary may be made by a device installed in the in-vehicle system 1 and other than the information notification device 10, but hereinafter, it is assumed that the determination is made by the controller 11.

[0049] In step S2, if the controller 11 determines that any notification is to be output (executed) (Y in step S2), it transitions to step S3. A notification to be output, i.e., a notification that has been determined to be output, is referred to as an output target notification. If it is determined that no notification is to be output (executed) in step S2 (N in step S2), it returns to step S1.

[0050] In step S3, the controller 11 determines whether custom pattern data is set for the notification to be output. Note that setting custom pattern data for the notification to be output and setting a custom pattern for the notification to be output are synonymous. If custom pattern data is set for the notification to be output (Y in step S3), the process proceeds to step S5. If custom pattern data is not set for the notification to be output (N in step S3), the process proceeds to step S4. Therefore, if the notification to be output is the i-th notification, if custom pattern data CST[i] is stored in the custom table TBL2, the process proceeds to step S5; if not, the process proceeds to step S4. Since no custom pattern data exists in the initial state of the information notification device 10, the process always proceeds from step S3 to step S4. After the notification pattern adjustment process is executed and custom pattern data is stored in the custom table TBL2, the process can proceed from step S3 to step S5.

[0051] In step S4, the controller 11 executes information notification processing. The information notification processing in step S4 is information notification processing using basic pattern data, and may hereinafter be referred to as basic notification processing. In the basic notification processing, the controller 11 executes an output target notification to the user U1 using basic pattern data corresponding to the output target notification. Therefore, if the output target notification is the i-th notification, in the basic notification processing, the controller 11 executes the i-th notification to the user U1 using the basic pattern data BS[i].

[0052] In step S5, the controller 11 executes information notification processing. The information notification processing in step S5 is information notification processing using custom pattern data, and may hereinafter be referred to as custom notification processing. In the custom notification processing, the controller 11 executes an output target notification to the user U1 using custom pattern data corresponding to the output target notification. Therefore, if the output target notification is the i-th notification, in the custom notification processing, the controller 11 executes the i-th notification to the user U1 using the custom pattern data CST[i].

[0053] After the information notification process in step S4 or S5, the process returns to step S1, and the processes from step S1 onwards are repeated.

[0054] FIG. 8(a) shows a specific example of notification content in the basic notification process, and FIGS. 8(b) and 8(c) show specific examples of notification content in the custom notification process. In FIGS. 8(a) to 8(c), it is assumed that the notification to be output is the first notification (left-turn guidance notification). In the example of FIG. 8(a), the basic pattern data BS[1] is an audio signal that expresses the words 601, "Please turn left at the next intersection." Therefore, the words 601 are output as sound from the speaker 82 in the basic notification process. In the example of FIG. 8(b), the custom pattern data CST[1] is an audio signal that expresses the words 602, "Please turn left at the next intersection." Therefore, the words 602 are output as sound from the speaker 82 in the custom notification process. When the user U1 inputs an adjustment instruction to the information notification device 10 requesting that the first notification be changed from a notification using basic pattern data BS[1] to a notification using Kansai dialect, the custom pattern data CST[1] corresponding to FIG. 8(b) is generated. In the example of Fig. 8(c), the custom pattern data CST[1] is an acoustic signal of a sound representing the words 603 "Please turn left at the next intersection," and therefore, in the custom notification process, the words 603 are output as sound from the speaker 82. When user U1 inputs a request to change the first notification from a notification using basic pattern data BS[1] to a notification using Hakata dialect as an adjustment instruction to the information notification device 10, the custom pattern data CST[1] corresponding to Fig. 8(c) is generated.

[0055] The custom pattern data is generated by the generation AI5 (see FIG. 1). The generation AI5 is formed by machine learning a DNN (deep neural network) using training data in a machine learning device (not shown). A generation AI already in existence at the time of filing the present application can be used as the generation AI5. A generation AI developed after the time of filing the present application may also be used as the generation AI5. The generation AI5 receives a prompt as input. The prompt is text data indicating an instruction (command) to the generation AI5. The generation AI5 generates and outputs a creative work in accordance with the content of the input prompt. In this embodiment, the generation AI5 can generate text data as a creative work, as well as audio signals and image signals. When the generation AI5 generates an audio signal, it can generate an audio file storing the generated audio signal in a required file format. When the generation AI5 generates an image signal, it can generate an image file storing the generated image signal in a required file format. The generation AI5 may include a generation AI capable of generating text data as a creative work, a generation AI capable of generating an audio signal as a creative work, and a generation AI capable of generating an image signal as a creative work, all of which may be included separately.

[0056] Fig. 9 shows the flow of operations related to the generation and setting of a custom pattern. Of the processes shown in Fig. 9, the processes performed by the controller 11 (information notification device 10) constitute the notification pattern adjustment process in step S6 of Fig. 7. First, the user U1 inputs an adjustment instruction 610 to the information notification device 10. The input of the adjustment instruction 610 is realized by an adjustment instruction operation on the operation input unit 84 or a voice operation using the microphone 83. The contents of the adjustment instruction operation on the operation input unit 84 and the contents of the voice operation using the microphone 83 are transmitted to the information notification device 10.

[0057] The adjustment instruction 610 includes information specifying the notification to be adjusted and preference information 611 indicating the preferences of user U1. The notification to be adjusted refers to the notification whose pattern is to be adjusted. Therefore, for example, when generating and setting custom pattern data CST[i] for the i-th notification, the notification to be adjusted is the i-th notification. There may be multiple notifications to be adjusted, but here it is considered that there is only one notification to be adjusted. The adjustment instruction 610 is an adjustment instruction for the notification to be adjusted.

[0058] The preference information 611 indicates the preference of the user U1, indicating what kind of pattern the user U1 wants the notification to be adjusted to have. In the adjustment instruction 610, the user U1 inputs text data of the request statement REQ1 into the operation input unit 84 or utters the request statement REQ1. At this time, the preference information 611 is indicated by the request statement REQ1. For example, in the adjustment instruction 610, the user U1 inputs text data of the request statement REQ1, "I want the notification to be in Kansai dialect," into the operation input unit 84 as the request statement REQ1 for the first notification, or utters an utterance indicating that "I want the notification to be in Kansai dialect." However, the controller 11 may also acquire the preference information 611 in a manner that does not require the user U1 to input or utter the request statement REQ1. That is, for example, the controller 11 may display multiple options for investigating the preferences of the user U1 on the display device 81, and create and acquire the preference information 611 based on which option the user U1 selects from the multiple options. The controller 11 may display a questionnaire for investigating the preferences of the user U1 on the display device 81 and acquire the preference information 611 based on the results of the questionnaire. An interactive questionnaire may be conducted using the speaker 82 and microphone 83. The controller 11 may estimate the preferences of the user U1 from the search history of the user U1 using the information terminal TM, and include the estimated preferences in the preference information 611.

[0059] In response to the input of the adjustment instruction 610, the controller 11 performs a prompt creation process 620. In the prompt creation process 620, the controller 11 creates a generation instruction prompt 630, which is a prompt based on the preference information 611. In the prompt creation process 620, the controller 11 creates the generation instruction prompt 630 by combining instruction text data 631 that instructs the generation of a pattern of the notification to be adjusted with preference text data 632 that corresponds to the preference information 611 (and therefore the preferences of the user U1). Therefore, the generation instruction prompt 630 includes the instruction text data 631 and the preference text data 632. The instruction text data 631 includes a basic pattern (basic pattern data) of the notification to be adjusted. The instruction text data 631 may be formed using fixed text data that does not depend on the preference information 611. The controller 11 transmits the created generation instruction prompt 630 to the server device 2 using the communication unit 13. The generation instruction prompt 630 received by the server device 2 is input to the generation AI5.

[0060] When the generation AI 5 receives the input of the generation instruction prompt 630, it executes a creation generation process 640 that generates a creation 650 in accordance with the generation instruction prompt 630. The creation 650 is output information (AI output information) of the generation AI 5. When the creation 650 is generated, the server device 2 transmits the creation 650 to the information notification device 10. The creation 650 is, for example, an acoustic signal of a sound representing the words 602, "Turn left at the next intersection" (see FIG. 8(b)). However, more specifically, the acoustic signal stored in an audio file is transmitted to the information notification device 10 as the creation 650. In the information notification device 10 (controller 11), the creation 650 is received by the communication unit 13. When the creation 650 is received by the communication unit 13, the controller 11 executes a test notification process 660.

[0061] In the test notification process 660, the controller 11 sets a test pattern 671 based on the creative work 650 and issues a notification to be adjusted using the test pattern 671 to the user U1. Basically, the creative work 650 itself is set as the test pattern 671, but the creative work 650 may be set as the test pattern 671 after undergoing necessary file conversion, etc. If the generation AI 5 cannot generate data in the file format or format requested by the controller 11, data in a file format or format that the generation AI 5 can generate is generated as the creative work 650. In this case, the controller 11 can generate the test pattern 671 by performing the necessary file conversion or format conversion on the creative work 650. The notification to be adjusted using the test pattern 671 is referred to as a test notification 670. Therefore, in the test notification process 660, the test notification 670 is issued to the user U1. The test pattern 671 corresponds to a candidate pattern (a candidate custom pattern) for the notification to be adjusted.

[0062] After receiving the test notification 670, the user U1 makes a test notification response 680. The test notification response 680 is the user U1's response (reaction) to receiving the test notification 670, and indicates whether the notification to be adjusted by the test pattern 671 should be approved by the user U1. The test notification response 680 is input to the information notification device 10. The input of the test notification response 680 is realized by the user U1 operating the operation input unit 84 or by a voice operation using the microphone 83. The contents of the operation on the operation input unit 84 and the contents of the voice operation using the microphone 83 are transmitted to the information notification device 10. Depending on the contents of the test notification response 680, the following operation branches into either an approval case or a non-approval case.

[0063] If the user U1 indicates a positive response (response) in the test notification response 680, the process branches to the approval case. That is, the approval case occurs when approval information indicating that the notification to be adjusted using the test pattern 671 is approved by the user U1 in the test notification response 680 is input to the information notification device 10. In this case, the approval information indicates a positive response (response). In the approval case, the controller 11 executes a registration process 690. The registration process 690 is a custom pattern registration process. In the registration process 690, the controller 11 sets (registers) the test pattern 671 based on the creative work 650 as the pattern of the notification to be adjusted. This setting (registration) corresponds to setting the test pattern 671 as a custom pattern for the notification to be adjusted, and is realized by storing the data of the test pattern 671 in the custom table TBL2 as custom pattern data corresponding to the notification to be adjusted. Therefore, for example, if the notification to be adjusted is the first notification, in the registration process 690, the controller 11 stores the data of the test pattern 671 in the custom table TBL2 as custom pattern data CST[1]. If custom pattern data CST[1] different from the test pattern 671 has already been stored in the custom table TBL2 before the registration process 690, the controller 11 updates the custom pattern data CST[1] with the data of the test pattern 671 in the registration process 690. When the registration process 690 is performed, the operation shown in Fig. 9, which starts from the adjustment instruction 610, ends without executing the pattern correction process 700 described below.

[0064] When the user U1 indicates a negative response (response) in the test notification response 680, the process branches to the non-approval case. That is, the case in which non-approval information indicating that the notification to be adjusted using the test pattern 671 is not approved by the user U1 in the test notification response 680 is input to the information notification device 10 is the non-approval case. In this case, the non-approval information indicates a negative response (response). In the non-approval case, the controller 11 executes the pattern correction process 700. In the pattern correction process 700, the controller 11 works with the generation AI 5 to correct the test pattern 671, which is a candidate for the pattern of the notification to be adjusted. This correction is achieved by having the generation AI 5 generate a new test pattern. That is, in the pattern correction process 700, the controller 11 works with the generation AI 5 to set a new test pattern (test pattern 771, described below) as a new candidate for the pattern of the notification to be adjusted. Setting a new test pattern here corresponds to creating a new test pattern (test pattern 771, described below) in collaboration with the generation AI 5.

[0065] In the pattern correction process 700, the controller 11 receives a readjustment instruction 710 from the user U11. The readjustment instruction 710 is input to the information notification device 10. The readjustment instruction 710 is input by the user U1 operating the operation input unit 84 or by voice operation using the microphone 83. The readjustment instruction 710 is an instruction to readjust the notification to be adjusted. The readjustment instruction 710 includes corrected preference information 711 indicating the preferences of the user U1. Similar to the preference information 611 described above, the corrected preference information 711 also indicates the preferences of the user U1, indicating what kind of pattern the notification to be adjusted should have. However, the corrected preference information 711 reflects the result of the user U1 receiving the test notification 670 (the notification to be adjusted using the test pattern 671). For example, the corrected preference information 711 indicates how the user U1 wants the pattern of the notification to be adjusted to be changed from the test pattern 671.

[0066] In the readjustment instruction 710, the user U1 inputs the text data of the request statement REQ2 into the operation input unit 84 or utters the request statement REQ2. At this time, the request statement REQ2 indicates the corrected preference information 711. For example, after receiving the test notification 670 in Kansai dialect, the user U1 inputs the text data of the request statement REQ2, requesting that the language of the notification to be adjusted be changed from Kansai dialect to Hakata dialect, into the operation input unit 84 or utters the request statement REQ2 in the readjustment instruction 710. However, the controller 11 may also acquire the corrected preference information 711 in a manner that does not require the user U1 to input or utter the request statement REQ2. That is, for example, the controller 11 may display multiple options for investigating the preferences of the user U1 on the display device 81, and create and acquire the corrected preference information 711 based on the option selected by the user U1 from the multiple options. The controller 11 may display a questionnaire for investigating the preferences of the user U1 on the display device 81 and acquire the revised preference information 711 based on the results of the questionnaire. The questionnaire may be conducted in an interactive format using the speaker 82 and the microphone 83.

[0067] In response to the input of the readjustment instruction 710, the controller 11 performs a prompt creation process 720. Note that the readjustment instruction 710 may be included in the test response notification 680, and in this case, in response to the input of the test response notification 680 including the readjustment instruction 710, the controller 11 performs the prompt creation process 720.

[0068] In the prompt creation process 720, the controller 11 creates a correction instruction prompt 730, which is a prompt based on the correction preference information 711. In the prompt creation process 720, the controller 11 creates the creation instruction prompt 730 by combining instruction text data 731 that instructs the generation of a pattern of a notification to be adjusted with response text data 732 based on the user's response (reaction) to the test notification 670. Therefore, the creation instruction prompt 730 includes the instruction text data 731 and the response text data 732. The response text data 732 reflects and includes the correction preference information 711. The readjustment instruction 710 corresponds to the user U1's response to the test notification 670 (a response that the test notification 670 cannot be approved and therefore the user wishes to readjust it), and the content of this response is reflected in the response text data 732. The instruction text data 731 may be formed using fixed text data that does not depend on the correction preference information 711. However, in order to allow the generation AI 5 to recognize the pattern to be corrected, it is advisable to include the test pattern 671 (data indicating the test pattern 671) used in the executed test notification 670 in the instruction text data 731. In cases where the test pattern 671 completely matches the creative work 650 and the server device 2 or the artificial intelligence device 4 has the creative work 650 stored therein, the test pattern 671 does not have to be included in the instruction text data 731. The controller 11 transmits the created generation instruction prompt 730 to the server device 2 using the communication unit 13. The generation instruction prompt 730 received by the server device 2 is input to the generation AI 5.

[0069] When the generation AI 5 receives the input of the generation instruction prompt 730, it executes a creation generation process 740 that generates a creation 750 in accordance with the generation instruction prompt 730. The creation 750 is output information (AI output information) of the generation AI 5. When the creation 750 is generated, the server device 2 transmits the creation 750 to the information notification device 10. The creation 750 is, for example, an acoustic signal of a sound representing the phrase 603, "Please turn left at the next intersection" (see FIG. 8(c)). However, more specifically, the acoustic signal stored in an audio file is transmitted to the information notification device 10 as the creation 750. In the information notification device 10 (controller 11), the creation 750 is received by the communication unit 13. When the creation 750 is received by the communication unit 13, the controller 11 executes a test notification process 760.

[0070] In the test notification process 760, the controller 11 sets a test pattern 771 based on the creative work 750 and issues a notification to be adjusted using the test pattern 771 to the user U1. Basically, the creative work 750 itself is set as the test pattern 771, but the creative work 750 may be set as the test pattern 771 after undergoing necessary file conversion, etc. If the generation AI 5 cannot generate data in the file format or format requested by the controller 11, data in a file format or format that the generation AI 5 can generate is generated as the creative work 750. In this case, the controller 11 can generate the test pattern 771 by performing the necessary file conversion or format conversion on the creative work 750. The notification to be adjusted using the test pattern 771 is referred to as a test notification 770. Therefore, in the test notification process 760, the test notification 770 is issued to the user U1. The test pattern 771 corresponds to a candidate pattern (a candidate custom pattern) for the notification to be adjusted.

[0071] After receiving the test notification 770, the user U1 makes a test notification response 780. The test notification response 780 is the user U1's response (reaction) to receiving the test notification 770, and indicates whether the notification to be adjusted by the test pattern 771 should be approved by the user U1. The test notification response 780 is input to the information notification device 10. The input of the test notification response 780 is realized by the user U1 operating the operation input unit 84 or by a voice operation using the microphone 83. The contents of the operation on the operation input unit 84 and the contents of the voice operation using the microphone 83 are transmitted to the information notification device 10. Depending on the contents of the test notification response 780, the following operation branches into either a correction approval case or a re-disapproval case.

[0072] If the user U1 indicates a positive response (response) in the test notification response 780, the process branches to the modification approval case. That is, the modification approval case occurs when approval information indicating that the notification to be adjusted using the test pattern 771 is approved by the user U1 in the test notification response 780 is input to the information notification device 10. In this case, the approval information indicates a positive response (response). In the modification approval case, the controller 11 executes a registration process 790. The registration process 790 is a custom pattern registration process. In the registration process 790, the controller 11 sets (registers) the test pattern 771 based on the creative work 750 as the pattern of the notification to be adjusted. This setting (registration) corresponds to setting the test pattern 771 as a custom pattern for the notification to be adjusted, and is realized by storing the data of the test pattern 771 in the custom table TBL2 as custom pattern data corresponding to the notification to be adjusted. Therefore, for example, if the notification to be adjusted is the first notification, in the registration process 790, the controller 11 stores the data of the test pattern 771 in the custom table TBL2 as custom pattern data CST[1]. If custom pattern data CST[1] different from the test pattern 771 has already been stored in the custom table TBL2 before the registration process 790, the controller 11 updates the custom pattern data CST[1] with the data of the test pattern 771 in the registration process 790. When the registration process 790 is performed, the operation shown in Fig. 9, which starts from the adjustment instruction 610, ends without executing the pattern correction process 700 again.

[0073] If the user U1 indicates a negative response (response) in the test notification response 780, the process branches to the re-non-approval case. That is, the re-non-approval case occurs when non-approval information indicating that the notification to be adjusted using the test pattern 771 is not approved by the user U1 in the test notification response 780 is input to the information notification device 10. In this case, the non-approval information indicates a negative response (response). In the re-non-approval case, the controller 11 executes the pattern correction process 700 again. In the re-pattern correction process 700, the controller 11 works with the generation AI 5 to correct the test pattern 771, which is a candidate for the pattern of the notification to be adjusted. This correction is achieved by having the generation AI 5 generate a new test pattern. That is, in the re-pattern correction process 700, the controller 11 works with the generation AI 5 to set a new test pattern different from the test patterns 671 and 771 as a new candidate for the pattern of the notification to be adjusted. Setting a new test pattern here corresponds to working with generation AI5 to create a new test pattern different from test patterns 671 and 771 as a new candidate for the pattern of the notification to be adjusted.

[0074] The pattern correction process 700 is a series of steps that starts with input of a readjustment instruction 710 to the information notification device 10 and ends with input of a test notification response 780 to the information notification device 10. In the case of re-rejection, this series of steps is executed again. This series of steps is repeated until registration processing 790 is performed (i.e., until user U1 is satisfied) or until a predetermined cancellation instruction is input from user U1 to the information notification device 10.

[0075] Below, several specific operational examples, application techniques, modified techniques, etc. related to the technology of this embodiment will be described in multiple examples. The matters described above in this embodiment are applied to each of the following examples unless otherwise specified and unless there is a contradiction. If there are any matters in each example that contradict the matters described above, the description in each example may take precedence. Furthermore, unless there is a contradiction, matters described in any of the multiple examples shown below can also be applied to any other example (i.e., any two or more of the multiple examples can be combined).

[0076] <<First Example>> A first embodiment will be described. Figures 10 and 11 show specific examples of prompts and the like in the operational flow of Figure 9. In the examples of Figures 10 and 11, the notification to be adjusted is the first notification (left turn guidance notification). Adjustment instruction 610a, generation instruction prompt 630a, creative work 650a, and test notification 670a in Figure 10 are adjustment instruction 610, generation instruction prompt 630, creative work 650, and test notification 670 (see Figure 9), respectively, according to the first embodiment.

[0077] In response to the adjustment instruction 610a, the user U1 inputs text data of a request sentence REQ1a "I would like notifications to be in Kansai dialect" into the operation input unit 84 or utters the request sentence REQ1a. The request sentence REQ1a is an example of the request sentence REQ1 (see FIG. 9). In a prompt creation process 620 in response to the input of the adjustment instruction 610a (see also FIG. 9), the controller 11 creates a generation instruction prompt 630a based on the request sentence REQ1a. The request sentence REQ1a is included in the adjustment instruction 610a and corresponds to the preference information 611 (see FIG. 9) according to the first embodiment.

[0078] The generation instruction prompt 630a is synthesized text data obtained by combining text data 631a_1, 632a, and 631a_2 in this order. The text data 631a_1 and 631a_2 are examples of instruction text data 631, and the text data 632a is an example of preference text data 632 (see FIG. 9). It is assumed that the vehicle V1 is a passenger car.

[0079] In the example of FIG. 10, the text data 631a_1 is "I am an in-vehicle device that provides multiple types of notifications to the user, the driver of a passenger vehicle. -Multiple types of notifications include left turn guidance notifications. When notifying drivers of left turns, I generally output the following message to them via voice: "Please turn left at the next intersection." The user has the following preferences regarding left turn notification: This is text data showing: In the example of FIG. 10, the text data 632a is "Users' request is to receive notifications in the Kansai dialect." This is text data showing: In the example of FIG. 10, the text data 631a_2 is "- Using the standard left turn notification "Turn left at the next intersection" as a reference, please generate new voice data for left turn notification in accordance with the user's wishes. - The length of the audio data must be between 1 and 3 seconds. The audio data bit rate is 96kbps. -Output the audio file that contains the generated audio data. The audio file format is mp3. This is text data showing:

[0080] The controller 11 creates text data 632a, which is an example of the preference text data 632, based on the preference information 611 of the user U1 (see FIG. 9). The preference information 611 here is the request sentence REQ1a itself or is extracted from the request sentence REQ1a by the controller 11. The controller 11 forms the text data 632a by applying the request sentence REQ1a to fixed text data for preference text data.

[0081] The controller 11 forms text data 631a_1 and 631a_2, which are examples of instruction text data 631, using fixed text data that does not depend on the preference information 611 (request sentence REQ1a). The controller 11 forms the text data 631a_1 and 631a_2 by applying the phrase "left turn guidance notification" and text data of a basic pattern for left turn guidance notification, "Please turn left at the next intersection," to the fixed text data for instruction text data. The text data 631a_1 and 631a_2 may be dedicated fixed text data for left turn guidance notification, and in this case, the instruction text data 631 (see FIG. 9) is formed using only the dedicated fixed text data.

[0082] The generation AI5 generates a creation 650a in accordance with the generation instruction prompt 630a. The creation 650a is an audio file in accordance with the generation instruction prompt 630a, and the audio file as the creation 650a stores an acoustic signal of a sound representing the words 602, "Turn left at the next intersection." Therefore, in the test notification 670a in which the creation 650a is set as the test pattern 671a, the words 602, "Turn left at the next intersection," are output as sound from the speaker 82. The test pattern 671a is the test pattern 671 according to the first embodiment (see FIG. 9).

[0083] In the approval case where the user U1 inputs the above-described approval information into the information notification device 10 in the test notification response 680 after receiving the test notification 670a, a registration process 690 is executed (see also FIG. 9 ). In the registration process 690, the controller 11 sets (registers) a test pattern 671a based on the creative work 650a as a pattern for the notification to be adjusted. This setting (registration) corresponds to setting the test pattern 671a as a custom pattern for the notification to be adjusted, and is achieved by storing the data of the test pattern 671a in the custom table TBL2 as custom pattern data corresponding to the notification to be adjusted. Therefore, in the registration process 690 according to the first embodiment, the controller 11 stores the data of the test pattern 671a in the custom table TBL2 as custom pattern data CST[1]. If custom pattern data CST[1] different from the test pattern 671a has already been stored in the custom table TBL2 before the registration process 690, the controller 11 updates the custom pattern data CST[1] with the data of the test pattern 671a in the registration process 690. When the registration process 690 is performed, the operation starting from the adjustment instruction 610a ends without executing the pattern correction process 700.

[0084] In the case of non-approval, in which the non-approval information described above is input to the information notification device 10 in the test notification response 680 by user U1 after receiving the test notification 670a, a pattern correction step 700 is executed (see also FIG. 9). A specific example of the pattern correction step 700 following the example of FIG. 10 is shown in FIG. 11.

[0085] In FIG. 11, a readjustment instruction 710a, a correction instruction prompt 730a, a creation 750a, and a test notification 770a are respectively the readjustment instruction 710, the correction instruction prompt 730, the creation 750, and the test notification 770 (see FIG. 9) according to the first embodiment.

[0086] In response to the readjustment instruction 710a, the user U1 inputs text data of the request sentence REQ2a "After all, change the notification to Hakata dialect" into the operation input unit 84 or utters the request sentence REQ2a. The request sentence REQ2a is an example of the request sentence REQ2 (see FIG. 9). In a prompt creation process 720 in response to the input of the readjustment instruction 710a (see also FIG. 9), the controller 11 creates a correction instruction prompt 730a based on the request sentence REQ2a. The request sentence REQ2a is included in the readjustment instruction 710a and corresponds to the corrected preference information 711 (see FIG. 9) according to the first embodiment.

[0087] The correction instruction prompt 730a is synthesized text data obtained by combining text data 731a_1, 732a, and 731a_2 in this order. The text data 731a_1 and 731a_2 are examples of instruction text data 731, and the text data 732a is an example of response text data 732 (see FIG. 9). In the series of operations shown in FIG. 9 (and therefore the series of operations shown in FIGS. 10 and 11), information is exchanged between the controller 11 and the generation AI 5 in a continuous interactive format. Therefore, when the generation AI 5 outputs a creation 650a in response to the generation instruction prompt 630a to the other device of the generation AI 5 (here, the information notification device 10), the generation AI 5 recognizes that the correction instruction prompt 730a has been input from the other device.

[0088] In the example of FIG. 11, text data 731a_1 is text data indicating the sentence "The user has the following wishes regarding the audio file of the left turn guidance notification that you just generated." In the example of FIG. 11, text data 732a is text data indicating the sentence "The user's new wish is, 'After all, change the notification to Hakata dialect.'" In the example of FIG. 11, text data 731a_2 is text data indicating the sentence "The user's new wish is, 'After all, change the notification to Hakata dialect.'" "Please refer to the generated audio file for the left turn notification and regenerate the audio data for the left turn notification in line with the user's new wishes. - The length of the audio data must be between 1 and 3 seconds. The audio data bit rate is 96kbps. -Output the audio file that contains the generated audio data. The audio file format is mp3. This is text data showing:

[0089] The controller 11 creates text data 732a, which is an example of the response text data 732, based on the modified preference information 711 of the user U1 (see FIG. 9). The modified preference information 711 here is the request sentence REQ2a itself or is extracted from the request sentence REQ2a by the controller 11. The controller 11 forms the text data 732a by applying the request sentence REQ2a to fixed text data for preference text data.

[0090] The controller 11 forms text data 731a_1 and 731a_2, which are examples of instruction text data 731, using fixed text data that does not depend on the modified preference information 711 (request sentence REQ2a). The controller 11 forms the text data 731a_1 and 731a_2 by applying the phrase "left turn guidance notice" to the fixed text data for instruction text data. The text data 731a_1 and 731a_2 may be dedicated fixed text data for the left turn guidance notice, and in this case, the instruction text data 731 (see FIG. 9) is formed using only the dedicated fixed text data.

[0091] The generation AI5 generates a creation 750a in accordance with the correction instruction prompt 730a. The creation 750a is an audio file in accordance with the correction instruction prompt 730a, and the audio file as the creation 750a stores an acoustic signal of a sound representing the words 603, "Please turn left at the next intersection." Therefore, in the test notification 770a in which the creation 750a is set as the test pattern 771a, the words 603, "Please turn left at the next intersection," are output as sound from the speaker 82. The test pattern 771a is the test pattern 771 (see FIG. 9) according to the example of FIG. 11.

[0092] In the case of a correction approval, in which the user U1 inputs the above-described approval information into the information notification device 10 in a test notification response 780 after receiving the test notification 770a, a registration process 790 is executed (see also FIG. 9 ). In the registration process 790, the controller 11 sets (registers) a test pattern 771a based on the creative work 750a as a pattern for the notification to be adjusted. This setting (registration) corresponds to setting the test pattern 771a as a custom pattern for the notification to be adjusted, and is achieved by storing the data of the test pattern 771a in the custom table TBL2 as custom pattern data corresponding to the notification to be adjusted. Therefore, in the registration process 790 shown in FIG. 11 , the controller 11 stores the data of the test pattern 771a in the custom table TBL2 as custom pattern data CST[1]. If custom pattern data CST[1] different from the test pattern 771a has already been stored in the custom table TBL2 before the registration process 790, the controller 11 updates the custom pattern data CST[1] with the data of the test pattern 771a in the registration process 790. When the registration process 790 is performed, the pattern correction process 700 is not executed again, and the operations in FIGS. 10 and 11 starting from the adjustment instruction 610a are completed.

[0093] In the case of re-disapproval, in which the above-mentioned disapproval information is input to the information notification device 10 in the test notification response 780 by the user U1 after receiving the test notification 770a, the pattern correction process 700 is executed again.

[0094] Note that in the series of operations shown in FIG. 9 (and therefore the series of operations shown in FIGS. 10 and 11 ), it is assumed that information is exchanged between the controller 11 and the generation AI 5 in a continuous, interactive format. Under this assumption, the generation AI 5 outputs a creation 650a in response to a creation instruction prompt 630a to the other device (here, the information notification device 10) of the generation AI 5, and recognizes that a correction instruction prompt 730a has been input from the other device. However, the information exchange between the controller 11 and the generation AI 5 does not have to be continuous, interactive. In this case, the controller 11 may include the test pattern 671a (data indicating the test pattern 671a) used in the executed test notification 670a in the text data 731a_2 so that the generation AI 5 can recognize the pattern to be corrected. Even if the information exchange between the controller 11 and the generation AI 5 is continuous, interactive, it is possible to include the test pattern 671a (data indicating the test pattern 671a) in the text data 731a_2.

[0095] In the first embodiment, for the sake of concreteness, the method for setting a custom pattern for the notification to be adjusted was explained assuming that the notification to be adjusted is the first notification, but the same applies when the notification to be adjusted is any of the second to nth notifications (the same setting method applies).

[0096] <<Second Example>> A second embodiment will now be described. A flowchart of the notification pattern adjustment process (step S6 in FIG. 7) by the controller 11 is shown in FIG. 12. The notification pattern adjustment process consists of steps S11 to S22. The processes of steps S11 to S22 are executed by the controller 11. For the sake of concreteness, the flow of the notification pattern adjustment process will be described with reference to the reference numerals shown in FIG. 9.

[0097] The controller 11 involved in the notification pattern adjustment process first performs a prompt creation process 620 based on an adjustment instruction 610 from the user U1 in step S11. In the prompt creation process 620, a generation instruction prompt 630 is created based on the above-mentioned preference information 611. In step S12 following step S11, the controller 11 transmits the generation instruction prompt 630 to the server device 2 using the communication unit 13. The generation instruction prompt 630 received by the server device 2 is input to the generation AI5.

[0098] The generation AI 5 generates and outputs a creative product 650 in accordance with the generation instruction prompt 630, and the server device 2 transmits the creative product 650, which is AI output information, to the information notification device 10. After transmitting the generation instruction prompt 630, the controller 11 waits for reception of the creative product 650 in step S13, and when the communication unit 13 receives the creative product 650 (Y in step S13), it transitions to step S14.

[0099] In step S14, the controller 11 executes a test notification process 660. In the test notification process 660, the controller 11 performs a test notification 670 for the user U1 using a test pattern 671. Executing the test notification 670 means executing a notification to be adjusted using the test pattern 671 based on the creative work 650.

[0100] After the test notification 670, in step S15 the controller 11 receives a test notification response 680 from the user U1 to the information notification device 10 and analyzes the contents of the test notification response 680. If the user U1 responds positively to the test pattern 671 (Y in step S15), the controller 11 causes the process to proceed to step S16. That is, if the test notification response 680 contains approval information indicating that the notification to be adjusted using the test pattern 671 is approved by the user U1, the controller 11 causes the process to proceed to step S16. In step S16, the controller 11 executes the above-mentioned registration process 690, and ends the notification pattern adjustment process upon completion of the registration process 690.

[0101] If the user U1's reaction to the test pattern 671 is negative (N in step S15), the controller 11 causes the process to proceed to step S17. That is, if the test notification response 680 contains non-approval information indicating that the adjustment target notification using the test pattern 671 is not approved by the user U1, the controller 11 causes the process to proceed to step S17. When the process proceeds from step S15 to step S17, a pattern correction process 700 is executed. The pattern correction process 700 is made up of the processes of steps S17 to S22.

[0102] In step S17, the controller 11 performs a prompt creation process 720 based on the readjustment instruction 710 from the user U1. In the prompt creation process 720, a correction instruction prompt 730 is created based on the above-mentioned correction preference information 711. In step S18 following step S17, the controller 11 transmits the correction instruction prompt 730 to the server device 2 using the communication unit 13. The correction instruction prompt 730 received by the server device 2 is input to the generation AI5.

[0103] The generation AI 5 generates and outputs a creation 750 in accordance with the correction instruction prompt 730, and the server device 2 transmits the creation 750, which is AI output information, to the information notification device 10. After transmitting the correction instruction prompt 730, the controller 11 waits for reception of the creation 750 in step S19, and when the creation 750 is received by the communication unit 13 (Y in step S19), a transition to step S20 occurs.

[0104] In step S20, the controller 11 executes a test notification process 760. In the test notification process 760, the controller 11 performs a test notification 770 for the user U1 using a test pattern 771. Executing the test notification 770 refers to executing a notification to be adjusted using the test pattern 771 based on the creative work 750.

[0105] After the test notification 770, in step S21 the controller 11 receives a test notification response 780 from the user U1 to the information notification device 10 and analyzes the contents of the test notification response 780. If the user U1 responds positively to the test pattern 771 (Y in step S21), the controller 11 causes the process to proceed to step S22. That is, if the test notification response 780 contains approval information indicating that the notification to be adjusted using the test pattern 771 is approved by the user U1, the controller 11 causes the process to proceed to step S22. In step S22, the controller 11 executes the above-mentioned registration process 790, and ends the notification pattern adjustment process upon completion of the registration process 790.

[0106] If the user U1's reaction to the test pattern 771 is negative (N in step S21), the controller 11 transitions to step S17. That is, if the test notification response 780 contains non-approval information indicating that the adjustment target notification using the test pattern 771 is not approved by the user U1, the controller 11 returns to step S17 and executes the pattern correction process 700 again. If the process proceeds from step S15 to step S17, the pattern correction process 700 is repeated until the registration process 790 is performed (i.e., until the user U1 is satisfied) or until a predetermined cancel instruction is input from the user U1 to the information notification device 10. Note that if the pattern correction process 700 is executed multiple times, the specific contents of the readjustment instruction 710, the correction instruction prompt 730, the creation 750, the test notification 770, etc. will differ among the multiple pattern correction processes 700.

[0107] ---Considerations--- The main operations of the information notification device 10 described so far are summarized below. The controller 11 creates a generation instruction prompt (630) by combining first text data (631) that instructs the generation of a pattern of information notification (notification to be adjusted) with second text data (632) according to the preferences of the user U1. The controller 11 transmits the generation instruction prompt (630) to the server device 2 including the generation AI5 using the communication unit 13, and then receives AI output information (650) generated by the generation AI5 based on the generation instruction prompt (630). The controller 11 can then set the information notification pattern based on the AI ​​output information (650).

[0108] This makes it possible to set an information notification pattern that suits the preferences of the user U1 using the generation AI 5, without requiring the user U1 to perform complex setting operations. By using an existing generation AI, the configuration of the information notification device 10 can be simplified (the processing load of the controller 11 itself can be light).

[0109] Specifically, after receiving the AI ​​output information (650), the controller 11 issues a test notification (670) to the user U1 using a test pattern (671) based on the AI ​​output information, and sets (registers) the test pattern (671) as the information notification pattern depending on the user U1's reaction (680) to the test notification.

[0110] This allows the user U1 to set (register) the test pattern as an information notification pattern when the test pattern desired by the user U1 is obtained (when the user U1 is satisfied with the test pattern). In other words, it is possible to set an information notification pattern that meets the user U1's wishes.

[0111] More specifically, if the user U1 responds affirmatively to the test notification (670), the controller 11 sets (registers) the test pattern (671) as the information notification pattern in the registration process (690). On the other hand, if the user U1 responds negatively to the test notification (670), the controller 11 cooperates with the generation AI 5 to set (in other words, create) a new test pattern (771) as a candidate for the information notification pattern.

[0112] When a test pattern that user U1 desires is obtained (when user U1 is satisfied), the test pattern can be set (registered) as an information notification pattern. If user U1 is not satisfied, a new test pattern can be set as a candidate for the information notification pattern, allowing user U1 to further search for a pattern that satisfies user U1.

[0113] More specifically, if the user U1 responds negatively to the test notification (670), the controller 11 creates a correction instruction prompt (730) including text data based on the user U1's response to the test notification. The controller 11 transmits the correction instruction prompt (730) to the server device 2 using the communication unit 13, and then receives new AI output information (750) generated by the generation AI based on the correction instruction prompt. The controller 11 then sets (in other words, creates) the new test pattern (771) based on the new AI output information (750).

[0114] This makes it possible to set (create) the new test pattern using the generation AI5 without requiring the user U1 to perform complicated setting operations.

[0115] After receiving the new AI output information (750), the controller 11 issues a new test notification (770) to the user U1 using a new test pattern (771), and sets (registers) the new test pattern (771) as the information notification pattern depending on the user U1's reaction (780) to the new test notification.

[0116] This makes it possible for the user U1 to set (register) the new test pattern as an information notification pattern if the new test pattern is one that the user U1 desires. In other words, it becomes possible to set an information notification pattern that meets the user U1's wishes.

[0117] Here, the controller 11 creates the second text data (631) based on the preference information (611) of the user U1, while creating the first text data as fixed text data that does not depend on the preference information.

[0118] Therefore, the controller 11 can easily (by simple processing) create a generation instruction prompt by simply obtaining the preference information of the user U1. By including the format of the information notification in the fixed text data, the generation AI 5 can generate a test pattern that is easy for the controller 11 to use.

[0119] <<Third Example>> A third embodiment will be described below. The controller 11 may be capable of executing an emotion estimation process for estimating the emotion of the user U1 based on the estimation information.

[0120] The controller 11 can perform emotion estimation processing using biometric data obtained from a biometric sensor 60 (see FIG. 2) as estimation information. The biometric sensor 60 may include an electroencephalogram (EEG) sensor and a heartbeat sensor attached to the user U1. The EEG sensor detects the user U1's brain waves and outputs brainwave data indicating the results of the brainwave detection. The heartbeat sensor detects the user U1's heartbeat and outputs heartbeat data indicating the results of the heartbeat detection. The EEG data and heartbeat data are biometric information of the user U1. The biometric data as estimation information includes the EEG data and heartbeat data, and may further include other biometric information (such as the amount of sweat, body temperature, and blood pressure of the user U1).

[0121] The controller 11 can use a known method for emotion estimation based on biometric data. For example, the method described in International Publication No. 2023 / 127930 proposed by the applicant of the present application may be used. In this case, the controller 11 derives the arousal level of the user U1's central nervous system based on the user U1's electroencephalogram data, while deriving the activity level of the user U1's autonomic nervous system based on the user U1's heart rate data. An emotion estimation model configured with a calculation formula or conversion data table for identifying emotions from the arousal level and activity level may be provided in the controller 11. The emotion of the user U1 can be estimated by inputting the arousal level and activity level derived as described above into the emotion estimation model. The controller 11 may also estimate the emotion of the user U1 in detail based on the user U1's facial expression in addition to the biometric data. The controller 11 can detect the user U1's facial expression using image recognition based on the facial image of the user U1 included in the in-vehicle camera image.

[0122] An AI model capable of estimating the brain wave data and heart rate data of user U1 from the facial image of user U1 may be provided in the controller 11. In this case, the brain wave data and heart rate data of user U1 can be estimated by inputting an in-car camera image (in-car image information) to the AI ​​model. Then, the controller 11 can estimate the emotions of user U1 based on the estimated brain wave data and heart rate data.

[0123] The emotion estimation method using the arousal level and activity level described above is merely one example of an emotion estimation method that the emotion estimation unit F2 can use. The estimation information includes at least one of biometric data of the user U1 during the estimation period and an in-vehicle camera image (a facial image of the user U1) during the estimation period. The controller 11 can estimate the emotion of the user U1 during the estimation period based on the estimation information.

[0124] The emotion of user U1 estimated by the emotion estimation process is referred to as an estimated emotion. The emotion estimation process enables the controller 11 to estimate the emotion of user U1 when the test notification (670, 770) was issued. The estimated emotion when the test notification 670 was issued is particularly referred to as an estimated emotion EM, and a method for using the estimated emotion EM will be described with reference to the symbols shown in FIG.

[0125] The controller 11 determines an estimated use period (hereinafter, particularly, the estimated use period P 670 The estimation period P 670 The estimation period P may be configured by a period in which the test notification 670 notifies the user of the adjustment target using the test pattern 671, and a period of a certain length of time that follows the period. 670 Based on the estimated use information in the estimated use period P 670 The estimated emotion EM is determined by estimating the emotion of the user U1 in the image.

[0126] When the controller 11 according to the third embodiment executes the prompt creation process 720 in the pattern correction step 700, it includes text data indicating the estimated emotion EM (hereinafter referred to as emotion text data) in the correction instruction prompt 730. The emotion text data is included, for example, in the response text data 732. For example, if the estimated emotion EM is displeasure, the controller 11 sets the emotion text data to the text data "When a notification was sent to the user based on your previous output information, the user felt an emotion of 'displeasure'." Alternatively, if the estimated emotion EM is sadness, the controller 11 sets the emotion text data to the text data "When a notification was sent to the user based on your previous output information, the user felt an emotion of 'sadness'." These emotion text data are formed by combining the text data indicating the estimated emotion EM ('displeasure' or 'sad') with the fixed text data "When a notification was sent to the user based on your previous output information, the user felt an emotion of 'displeasure'."

[0127] When the correction instruction prompt 730 including the emotion text data is input to the generation AI 5, the generation AI 5 will take the emotion text data into consideration to generate the creation 750. Although it depends on how the generation AI 5 is made, the generation AI 5 is usually expected to generate the creation 750 with the aim of evoking positive emotions in the user U1, which is beneficial to the user U1 who is seeking a better notification pattern.

[0128] The text data "Aim to make the user feel positive emotions" may be included in the correction instruction prompt 730. Although the method for using the estimated emotions when the test notification 670 is sent has been described, the estimated emotions when the test notification 770 is sent can also be used in a similar manner.

[0129] The controller 11 may also suggest to the user U1 a change in the notification pattern based on the estimated emotion. That is, for example, when the controller 11 determines that the emotion of the user U1 has changed from positive to negative by performing the i-th notification, the controller 11 may suggest to the user U1 a change in the pattern of the i-th notification. That is, when the estimated emotion of the user U1 immediately before the execution of the i-th notification is positive and the estimated emotion of the user U1 immediately after the execution of the i-th notification is negative, the controller 11 may make the above suggestion. The suggestion is made using the display device 81 or the speaker 82. In response to the suggestion, the user U1 may input an adjustment instruction 610 to the information notification device 10.

[0130] <<Fourth Example>> A fourth embodiment will be described. After the user receives an attention notification that is one of the first to n-th notifications, the user U1 often desires to change the pattern of the attention notification based on the pattern of the attention notification that he or she received.

[0131] Referring to FIG. 13, at timing t A1Consider case CS4 in which user U1 wishes to change the pattern of attention notification after sound 910 is output from speaker 82 due to attention notification. In case CS4, user U1 inputs to the information notification device 10 an adjustment instruction 610 that specifies attention notification as the notification to be adjusted, and includes in the adjustment instruction 610 preference information 611 how the user U1 wishes to change the pattern of attention notification. In this case, the adjustment instruction 610 functions as an instruction requesting that the pattern of attention notification be changed from a specific pattern.

[0132] Here, the specific pattern is at timing t A1 The pattern of attention notification at timing t A1 Alternatively, the specific pattern may be a download pattern acquired by the controller 11 via a download process (described later). That is, at the timing t A1 The attention notification is performed in a specific pattern, and the specific pattern is at timing t A1 It is a basic pattern or a custom pattern of attention notification at timing t A1 is a concept having a time length equivalent to the execution period of the attention notification.

[0133] In case CS4, the controller 11 receives the adjustment instruction 610 and generates a generation instruction prompt 630 through a prompt generation process 620 (see FIG. 9 ). At this time, text data indicating the characteristics of the specific pattern is included in the generation instruction prompt 630. The text data indicating the characteristics of the specific pattern is referred to as characteristic text data. As described above (see FIG. 9 ), the generation instruction prompt 630 includes the instruction text data 631 and the preference text data 632. In addition to these, the generation instruction prompt 630 also includes characteristic text data. The characteristic text data may be part of the instruction text data 631 or part of the preference text data 632.

[0134] In the example of FIG. 13, the sound 910 is a voice uttered by a woman saying, "Please be careful not to exceed the speed limit," and it is assumed that at least the user U1 recognizes the sound 910 as a voice uttered by a woman. It is also assumed that the user U1 desires that the attention notification be given by a voice uttered by a man. In this case, the user U1 recognizes the sound 910 as a voice uttered by a man at the timing t A1 After receiving the attention notification at timing t, the user U1 issues an adjustment instruction 610 including information to the effect that the attention notification should be made in a male voice (information to the effect that the attention notification sound should be changed from a female voice to a male voice) in the preference information 611. Here, the user U1 issues an adjustment instruction 610 at timing t A1 After receiving the attention notification, the request statement 915 "Change the previous notification to a male voice notification" is input to the information notification device 10 via the operation input unit 84 or the microphone 83 in the adjustment instruction 610. The request statement 915 is an example of the request statement REQ1 (see FIG. 9).

[0135] In response to the adjustment instruction 610, a generation instruction prompt 630 including the characteristic text data is transmitted from the information notification device 10 to the server device 2, and is input to the generation AI 5. Then, the generation AI 5 generates a creation 650 in accordance with the generation instruction prompt 630 including the characteristic text data. The creation 650 is transmitted from the server device 2 to the information notification device 10 and received by the information notification device 10. In the example of FIG. 13 , the characteristic text data is, for example, “The notification that I, an in-vehicle device, just sent to the user is a voice output of the phrase ‘Please be careful not to exceed the speed limit,’ and this voice output is composed of a female voice.” Furthermore, the preference text data 632 is a sentence including a request sentence 915, “The user expressed a desire for the previous notification to be a male voice notification in response to the notification just sent.”

[0136] Then, the generation AI generates the creation 650 according to the wishes of the user U1, and as a result, at timing t A2 A test notification 670 is performed in which sound 920 is output from the speaker 82 at timing t A2is a concept having a time length equivalent to the execution period of the test notification 670. Sound 920 is an audio signal constituting the creative work 650 played back by the speaker 82, and is the sound of a man speaking the words "Please be careful not to exceed the speed limit" (a sound that makes it easy to hear as if the words were spoken by a man). Thereafter, depending on the reaction of user U1 to the test notification 670, the registration process 690 or the pattern correction process 700 is executed.

[0137] While different from the example shown in FIG. 13 , if sound 910 is a bell ringing sound with a fundamental frequency of 400 Hz and user U1 requests "a lower sound" in adjustment instruction 610, the result will be as follows: In this case, the characteristic text data will contain the sentence "The notification I, the in-vehicle device, just made to the user was a notification by sound output, and that sound was a bell ringing sound with a fundamental frequency of 400 Hz." Furthermore, preference text data 632 will contain the sentence "The user wishes the notification to be 'lower'." In this case, creation 650 representing an audio signal with a lower frequency than sound 910 will be generated by generation AI5, and as a result, for example, a 300 Hz bell ringing sound will be output from speaker 82 in test notification 670.

[0138] As shown in this embodiment, by including characteristic text data indicating the characteristics of a specific pattern in the generation instruction prompt 630, it becomes clear to the generation AI 5 what kind of pattern should be generated starting from the specific pattern. As a result, it becomes easier to generate the notification pattern desired by the user U1.

[0139] <<Fifth Example>> A fifth embodiment will now be described.

[0140] Any test pattern created after the adjustment instruction 610 and before the registration process 690 or 790 is performed is stored for a certain period in the memory 12 or recording medium 14 under the control of the controller 11. In response to a retest instruction input from the user U1 to the information notification device 10, the controller 11 can perform a test notification using any test pattern stored in the memory 12 or recording medium 14. In other words, the user U1 can also try a notification using a test pattern created in the past after the fact. The retest instruction is input by operating the operation input unit 84 or by voice operation using the microphone 83.

[0141] When the test pattern 671 is an acoustic signal, the information notification device 10 may provide a user interface that allows the user U1 to easily adjust the frequency characteristics of the test pattern 671 in response to the readjustment instruction 710.

[0142] For example, if the notification to be adjusted is a left-turn guidance notification, the left-turn guidance notification is issued when the vehicle V1 approaches a point where it should turn left. A request to advance or delay the timing of the left-turn guidance notification can also be included in the adjustment instruction 610 or 710. The controller 11 subsequently issues a left-turn guidance notification to meet the request. The controller 11 can independently respond to the request without the cooperation of the generation AI 5. That is, the controller 11 can issue a left-turn guidance notification when the distance between the vehicle V1 and the point where the vehicle V1 should turn left decreases to the notification output distance. In principle, the notification output distance is equal to the initial setting distance (e.g., 100 m). In this case, if the adjustment instruction 610 or 710 includes a request to advance the timing of the left-turn guidance notification, the controller 11 simply increases the notification output distance from the initial setting distance. Conversely, if the adjustment instruction 610 or 710 includes a request to delay the timing of the left-turn guidance notification, the controller 11 simply decreases the notification output distance from the initial setting distance. Alternatively, the above request can be met as follows by using a creative work generated by a generation AI for a left turn guidance notification. Suppose the creative work generated by a generation AI for a left turn guidance notification is an audio signal J for the left turn guidance notification. In this case, the controller 11 generates a pattern for the left turn guidance notification (a custom pattern or a test pattern) by combining a silent audio signal with the audio signal J. The silent audio signal is inserted before the audio signal J on the time axis. Even if the silent audio signal is played on the speaker 82, no sound is output from the speaker 82. The controller 11 can meet requests to advance or delay the timing of the left turn guidance notification by adjusting the length of the silent audio signal. The same applies to notifications other than left turn guidance notifications.

[0143] When the information terminal TM (see FIG. 1 ) and the information notification device 10 are connected via a short-range wireless communication line such as Bluetooth (registered trademark), the microphone and operation input unit provided in the information terminal TM can function as the microphone 83 and operation input unit 84. However, to ensure safety, the controller 11 locks the function of inputting adjustment instructions 610, 710, etc. using the operation input unit of the information terminal TM so that it is enabled only when the vehicle V1 is stopped. The controller 11 may also lock the function of inputting adjustment instructions 610, 710, etc. using the microphone of the information terminal TM (hereinafter referred to as the terminal hands-free function) so that it is enabled only when the vehicle V1 is stopped. However, the terminal hands-free function may also be enabled while the vehicle V1 is moving. Whether the terminal hands-free function is enabled may be set in advance by the user U1 in the information notification device 10. The same applies to the function of inputting adjustment instructions 610, 710, etc. using a microphone installed in the vehicle V1 other than the microphone of the information terminal TM.

[0144] The data approved by the user U1 in the notification pattern adjustment process (step S6 in FIG. 7) is stored in the recording medium 14 by the controller 11 and in the database 6 by the management device 3 (see FIG. 1). The data approved by the user U1 is the data of the test pattern 671 or 771 stored in the custom table TBL2 in the registration process 690 or 790 (see FIG. 9).

[0145] In registration processing 690, the controller 11 transmits a first save instruction signal to the server device 2 instructing that the data of the test pattern 671 be saved in the database 6. Based on reception of the first save instruction signal, the management device 3 executes a first save processing in which the data of the test pattern 671 is saved in the database 6. In registration processing 790, the controller 11 transmits a second save instruction signal to the server device 2 instructing that the data of the test pattern 771 be saved in the database 6. Based on reception of the second save instruction signal, the management device 3 executes a second save processing in which the data of the test pattern 771 is saved in the database 6.

[0146] The notification pattern adjustment process may be performed multiple times for various types of notifications. By performing the notification pattern adjustment process multiple times, the custom pattern data CST[i] for the i-th notification may be updated multiple times. The controller 11 may continue to store each piece of data approved by user U1 in the recording medium 14 for a certain period of time. The management device 3 may continue to store each piece of data approved by user U1 in the database 6 for a certain period of time. In response to a request from user U1, the controller 11 may try out or reuse the data stored in the recording medium 14 as test pattern or custom pattern data for any notification. In response to a request from user U1, the controller 11 may also read data stored in the database 6 through the management device 3. The notification pattern indicated by the data read from the database 6 is referred to as a downloaded pattern. The controller 11 may try out or reuse the downloaded pattern as a test pattern or custom pattern for any notification.

[0147] In the first storage process, the management device 3 stores data of the test pattern 671 in the database 6 in association with the identification information of the notification corresponding to the test pattern 671 and the identification information of the user U1. In the second storage process, the management device 3 stores data of the test pattern 771 in association with the identification information of the notification corresponding to the test pattern 771 and the identification information of the user U1 in the database 6. The identification information of the notification corresponding to the test pattern 671 or 771 is identification information that specifies whether the notification to be adjusted is one of the first to nth notifications. The identification information of the user U1 is identification information unique to the user U1, and is assumed to have been set in the information notification device 10 and stored in the recording medium 14 before the operation shown in FIG. 7 is performed. The identification information of the notification corresponding to the test pattern 671 or 771 and the identification information of the user U1 are transmitted from the controller 11 to the server device 2.

[0148] Using the user U1's identification information, the user U1 can access the data in the database 6 in another vehicle. The other vehicle is a vehicle different from the vehicle V1. However, it is assumed that an in-vehicle system equivalent to the in-vehicle system 1 in FIG. 1 is also installed in the other vehicle. When the user U1 gets into the other vehicle, the user U1 inputs the user U1's identification information into the information notification device 10 installed in the other vehicle and then inputs a predetermined read request operation. Based on the input identification information of the user U1 and the read request operation, the information notification device 10 installed in the other vehicle reads data stored in the database 6 and associated with the user U1's identification information via the server device 2. This enables the information notification device 10 installed in the other vehicle to provide various notifications based on the data read from the database 6 (download pattern data). The same applies if the information notification device 10 in the vehicle V1 is replaced from the first information notification device 10 to a second information notification device 10 due to a malfunction or the like, and the data in the database 6 can also be accessed by the second information notification device 10.

[0149] <<Sixth Example>> A sixth embodiment will be described. Any user other than user U1 will be referred to as the "other user." In the sixth embodiment, user U1 is a person riding in vehicle V1, while the other user is a person riding in a vehicle other than vehicle V1. It is assumed that an in-vehicle system equivalent to the in-vehicle system 1 of FIG. 1 is also installed in the other vehicle. In the sixth embodiment, for the sake of concreteness of explanation, the other user will be referred to as user U2, the other vehicle will be referred to as vehicle V2, the in-vehicle system 1 installed in vehicle V2 will be referred to as in-vehicle system 1', and the information notification device 10 provided in the in-vehicle system 1' will be referred to as information notification device 10'.

[0150] Users U1 and U2 can share the notification data α created by the information notification device 10 and the notification data β created by the information notification device 10' via SNS (social networking service) or the like. Notification data α refers to data of a notification pattern (test pattern or custom pattern) created by the information notification device 10 in collaboration with the generation AI 5. Notification data β refers to data of a notification pattern (test pattern or custom pattern) created by the information notification device 10' in collaboration with the generation AI 5. Notification data α or β may be notification data that has received high praise on SNS or the like (notification data that many users have a favorable impression of).

[0151] The use of notification data β by user U1 will now be described. User U1 can download notification data β published on the communication network NET to the information notification device 10. Alternatively, user U1 can download notification data β published on the communication network NET to any computer device and load the downloaded notification data β into the information notification device 10. The notification data β published on the communication network NET may be data stored in the database 6. The notification pattern indicated by the notification data β belongs to the download pattern. After the notification data β is acquired by the information notification device 10, the controller 11 in the information notification device 10 can execute a test notification by setting the notification data β as a test pattern, or can set the notification data β as a custom test pattern for any notification.

[0152] After the test notification in which the information notification device 10 sets the notification data β as a test pattern, the user U1 can request modification of the test pattern, similar to the operational flow shown in FIG. 9 (the controller 11 can transmit a modification instruction prompt 730 based on the user U1's request to the server device 2). The notification data β may preferably include, as additional information, a request statement (REQ1 or REQ2) that the user U2 input to the information notification device 10′ when the notification data β was generated by the generation AI5. The information notification device 10 or any computer device can visually or audibly convey the additional information to the user U1 (for example, by displaying it on the display device 81 or outputting it from the speaker 82). By referring to the additional information, the user U1 can know what request statement was used to create the notification data β. This is expected to be useful for modifying the notification data β to the form desired by the user U1.

[0153] The use of the notification data α by user U2 is similar to the use of the notification data β by user U1. That is, user U2 can download the notification data α published on the communication network NET to the information notification device 10′. Alternatively, user U2 can download the notification data α published on the communication network NET to any computer device and load the downloaded notification data α into the information notification device 10′. The notification data α published on the communication network NET may be data stored in the database 6. The notification pattern indicated by the notification data α belongs to the download pattern. After the notification data α is acquired by the information notification device 10′, the controller 11 in the information notification device 10′ can execute a test notification by setting the notification data α as a test pattern, or can set the notification data α as a custom test pattern for any notification.

[0154] After the test notification in which the information notification device 10′ sets the notification data α as a test pattern, the user U2 can request modification of the test pattern, similar to the operational flow shown in FIG. 9 (the controller 11 can transmit a modification instruction prompt 730 based on the request of the user U2 to the server device 2). The notification data α may preferably include, as additional information, a request statement (REQ1 or REQ2) that the user U1 input to the information notification device 10 when the notification data α was generated by the generation AI 5. The information notification device 10′ or any computer device can visually or audibly convey the additional information to the user U2 (for example, by displaying it on the display device 81 or outputting it from the speaker 82). By referring to the additional information, the user U2 can know what request statement was used to create the notification data α. This is expected to be useful for modifying the notification data α to the form desired by the user U2.

[0155] <<Seventh Example>> A seventh embodiment will now be described. Fig. 14 shows a functional block diagram of the controller 11. The controller 11 is provided with functional blocks F1 to F3. The functions of the functional blocks F1 to F3 may be realized by the controller 11 executing a program recorded in the memory 12 or any other recording medium. The relationship between each functional block and the flowcharts of Figs. 7 and 12 will now be described (see also Fig. 9 as appropriate).

[0156] The functional block F1 is a notification determination unit. The notification determination unit F1 performs the processing of step S2 in FIG. 7. That is, the notification determination unit F1 determines which of the first to n-th notifications to execute. The functional block F2 is a notification execution unit. The notification execution unit F2 performs the information notification processing of steps S4 and S5 in FIG. 7. The notification execution unit F2 is also responsible for the judgment processing of step S3 in FIG. 7. That is, the notification execution unit F2 selectively transitions from step S3 to step S4 or S5 by determining whether custom pattern data is set for the notification to be output. The functional block F3 is a pattern adjustment unit. The pattern adjustment unit F3 performs the notification pattern adjustment processing of step S6 in FIG. 7. The pattern adjustment unit F3 is also responsible for the judgment processing of step S1 in FIG. 7. That is, the pattern adjustment unit F3 determines in step S1 whether an adjustment instruction from the user U1 has been input to the information notification device 10.

[0157] <<Eighth Example>> An eighth embodiment will now be described. As shown in Fig. 15, the generation AI 5 may be provided within the controller 11. When the generation AI 5 is provided within the controller 11, the transmission and reception of the various prompts and creative works described above performed by the information notification device 10 and the server device 2 becomes transmission and reception of signals completed within the controller 11. Although not specifically shown, the generation AI 5 may be built into a computer device provided outside the information notification device 10 and within the in-vehicle system 1.

[0158] <<Ninth Example>> A ninth embodiment will now be described.

[0159] Although notifications in audio output format have been mainly cited as examples of notifications to be adjusted, the notifications to be adjusted may also be notifications in image output format or vibration output format. The notification pattern adjustment process when the notification to be adjusted is a notification in image output format or vibration output format is the same as when the notification to be adjusted is a notification in audio output format. If the notification to be adjusted is a notification in image output format, the user U1 may request, in the adjustment instruction 610, how they would like the display content (display pattern) of the display device 81 for the notification to be adjusted to be changed. A generation instruction prompt 630 is created in accordance with this request from the controller 11. The same applies to the readjustment instruction 710. If the notification to be adjusted is a notification in vibration output format, the user U1 may request, in the adjustment instruction 610, how they would like the vibration content (vibration frequency, intensity, etc.) of the vibration device 85 for the notification to be adjusted to be changed. A generation instruction prompt 630 is created in accordance with this request from the controller 11. The same applies to the readjustment instruction 710.

[0160] The vehicle V1 is not limited to a car or the like that travels on a road, but may also be a train or the like. Furthermore, the application scope of the information notification device 10 is not limited to vehicles. The present invention is applicable to any information notification device that provides any notification to a user. For example, the information notification device 10 may be applied to a factory, in which case the user U1 may be a factory manager or the like, and examples of notifications include a notification that materials should be input into a machine or a notification that the machine's temperature is abnormal.

[0161] A program that causes a computer device to execute any of the methods described in the embodiments of the present invention, and a non-volatile recording medium on which the program is recorded, are included within the scope of the embodiments of the present invention. The program that causes a computer device to execute any of the methods described in the embodiments of the present invention may be a subprogram incorporated into any main program or called by any main program. The information notification device 10 is a type of computer device. The method executed by the information notification device 10 can be referred to as a notification pattern setting method. Any of the processes in the embodiments of the present invention may be realized by hardware such as a semiconductor integrated circuit, software equivalent to the program, or a combination of hardware and software.

[0162] The embodiments of the present invention can be modified in various ways as appropriate within the scope of the technical ideas set forth in the claims. The above-described embodiments are merely examples of the present invention, and the meanings of the terms of the present invention and each constituent element are not limited to those described in the above-described embodiments. The specific numerical values ​​shown in the above description are merely examples, and as a matter of course, they can be changed to various numerical values. [Explanation of symbols]

[0163] 1. In-vehicle systems 2. Server device 3 Management device 4. Artificial Intelligence Device 5 Generation AI 6 Database V1 vehicle U1 User TM Information Terminal NET communication network ST1 seat 10 Information notification device 11 Controller 12 Memory 13 Communications Department 14 Recording media 20 Driving control device 30 Actuator section 40 Vehicle sensor unit 60 Biometric Sensor 71 Exterior camera 72 In-car camera 80 HMI 81 Display device 82 Speaker 83 Microphone 84 Operation input section 85 Vibration device F1 Notification judgment part F2 Notification Execution F3 Pattern adjustment section

Claims

1. An information notification device having a controller and notifying a user of information, the controller creates a generation instruction prompt by combining first text data instructing generation of the information notification pattern with second text data according to the user's preference; the controller transmits the generation instruction prompt to an external device that is an external device provided outside the information notification device and includes a generation AI, and then receives AI output information generated by the generation AI based on the generation instruction prompt; The controller sets a pattern of the information notification based on the AI ​​output information. , information notification device.

2. After receiving the AI ​​output information, the controller issues a test notification to the user using a test pattern based on the AI ​​output information, and sets the test pattern as the information notification pattern depending on the user's reaction to the test notification. The information notification device according to claim 1 .

3. If the user responds positively to the test notification, the controller sets the test pattern as the information notification pattern, and if the user responds negatively to the test notification, the controller cooperates with the generation AI to set a new test pattern as a candidate for the information notification pattern. The information notification device according to claim 2 .

4. If the user responds negatively to the test notification, the controller creates a correction instruction prompt including text data based on the user's response to the test notification, transmits the correction instruction prompt to the external device, and then receives new AI output information generated by the generation AI based on the correction instruction prompt, and sets the new test pattern based on the new AI output information. The information notification device according to claim 3 .

5. After receiving the new AI output information, the controller issues a new test notification to the user using the new test pattern, and sets the new test pattern as the information notification pattern depending on the user's reaction to the new test notification.

5. The information notification device according to claim 4.

6. When creating the correction instruction prompt, the controller includes text data corresponding to the user's emotion upon receiving the test notification in the correction instruction prompt.

5. The information notification device according to claim 4.

7. The controller creates the second text data based on the user's preference information, and creates the first text data using fixed text data that does not depend on the preference information.

7. The information notification device according to claim 1.

8. When the controller receives an instruction from the user requesting to change the information notification pattern from the specific pattern after notifying the user of the information in the specific pattern, the controller includes characteristic text data indicating the characteristics of the specific pattern in the generation instruction prompt.

7. The information notification device according to claim 1.

9. An information notification device having a controller and a communication unit, which notifies a user of information, the controller creates a generation instruction prompt by combining first text data instructing generation of the information notification pattern with second text data according to the user's preference; the controller uses the communication unit to send the generation instruction prompt to an external device that is an external device provided outside the information notification device and includes a generation AI, and then receives AI output information generated by the generation AI based on the generation instruction prompt; The controller sets a pattern of the information notification based on the AI ​​output information. , information notification device.

10. A notification pattern setting method executed in an information notification device that notifies a user of information, creating a generation instruction prompt by combining first text data instructing the generation of the information notification pattern with second text data according to the user's preference; transmitting the generation instruction prompt to an external device that is an external device provided outside the information notification device and includes a generation AI, and then receiving AI output information generated by the generation AI based on the generation instruction prompt; Setting the information notification pattern based on the AI ​​output information ,How to set notification patterns.

11. A program that causes a computer device to perform the notification pattern setting method according to claim 10.

12. An information notification device according to any one of claims 1 to 6 and 9; The external device ,Notification pattern setting system.

13. An information notification device having a controller including a generation AI and notifying a user of information, the controller creates a generation instruction prompt by combining first text data instructing generation of the information notification pattern with second text data according to the user's preference; The controller sets the pattern of the information notification based on AI output information generated by the generation AI by inputting the generation instruction prompt to the generation AI. , information notification device.

Images