Voice guidance device, voice guidance method, program, and voice guidance system

Abstract

We provide technology that can enhance the realism of voice guidance. [Solution] An exemplary voice guidance device generates a first prompt to identify the range in which a visible object can be seen when the user is walking along the guidance route, generates a second prompt to obtain guidance text about the object of a length corresponding to the visible range, transmits the generated first and second prompts to a response generation server including an LLM, and uses the response from the response generation server to provide voice guidance about the object within the visible range.

Description

Technical Field

[0001] The present invention relates to a voice guidance device, a voice guidance method, a program, and a voice guidance system.

Background Art

[0002] Conventionally, there is a technology of reading out a guidance text regarding a landmark within the arrival time to a predetermined landmark (see, for example, Patent Document 1 and Patent Document 2).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the conventional technology, when the user cannot see the landmark while the guidance text regarding the landmark is being read out, there is a problem that the sense of presence of the guidance is low and it is difficult to increase the user's satisfaction.

[0005] In view of the above circumstances, an object of the present invention is to provide a technology capable of enhancing the sense of presence of voice guidance.

Means for Solving the Problems

[0006] An exemplary voice guidance device of the present invention generates a first prompt to identify the range in which a visible object can be seen when the user is traveling along a guidance route, generates a second prompt to obtain guidance text about the object of a length corresponding to the visible range, transmits the generated first and second prompts to a response generation server including LLMs (Large Language Models), and uses the response from the response generation server to provide voice guidance about the object within the visible range. [Effects of the Invention]

[0007] According to an exemplary version of the present invention, by providing guidance text about the object that corresponds to the visible range of the object, the object can be guided by voice within that visible range, thereby enhancing the sense of realism in the voice guidance. [Brief explanation of the drawing]

[0008] [Figure 1] A diagram showing an example configuration of the voice guidance system according to the embodiment. [Figure 2] Diagram showing an example of the configuration of a navigation system. [Figure 3] A diagram showing an example of the information that can be obtained. [Figure 4] A diagram showing an example of visible range information. [Figure 5] A diagram showing an example of landscape information. [Figure 6] Diagram showing an example configuration of the response generation server. [Figure 7] Flowchart for voice guidance processing [Modes for carrying out the invention]

[0009] Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings.

[0010] [Voice guidance system] Figure 1 shows an example of the configuration of the voice guidance system SYS1 according to an embodiment. As shown in Figure 1, the voice guidance system SYS1 includes a navigation device 1 and an answer generation server SV1.

[0011] In this embodiment, the navigation device 1 is used inside the vehicle V1. The navigation device 1 may be a device that is permanently installed in the vehicle V1, or it may be a portable device that can be moved in and out of the vehicle V1.

[0012] Navigation device 1 provides route guidance and, when the landscape guidance function is enabled, uses the responses generated by the response generation server SV1 to provide voice guidance for target objects (landmarks). In other words, navigation device 1 is an example of a voice guidance device that provides voice guidance for target objects (landmarks). The landscape guidance function can be switched on or off by the user's operation on the control device 6, which will be described later.

[0013] The answer generation server SV1 includes an LLM. The answer generation server SV1 inputs the question text sent from the navigation device 1 via the network (Internet) NT1 into the LLM and generates an answer text corresponding to the question text.

[0014] The answer generation server SV1 can access numerous Web (World Wide Web) site provider servers SV2 via the network NT1. Each Web provider server SV2 provides its own unique website.

[0015] [Navigation system] Figure 2 shows the configuration of the navigation device 1. The navigation device 1 comprises a controller 2, a memory 3, and a communication unit 4.

[0016] The controller 2 includes a processor that performs arithmetic processing and the like. The processor may be configured to include, for example, a CPU (Central Processing Unit). The controller 2 may be constituted by one processor, or may be constituted by a plurality of processors. When constituted by a plurality of processors, those processors may be connected to be communicable with each other. Further, the controller 2 includes computer components such as memory (such as RAM (Random Access Memory), ROM (Read Only Memory), etc.) necessary for program execution.

[0017] As its functions, the controller 2 includes an information acquisition unit 21, a destination setting unit 22, a route search unit 23, a timing unit 24, a route guidance unit 25, a prompt generation unit 26, a landscape guidance unit 27, and an output control unit 28.

[0018] Also, each of the functional units 21 to 28 may be realized by the processor executing arithmetic processing according to one program. However, for example, it may be a configuration realized by the processor executing arithmetic processing according to separate programs for each functional unit. Further, as described above, each of the functional units 21 to 28 may be realized by causing the processor to execute a program, that is, by software, but may also be realized by other methods. At least a part of each of the functional units 21 to 28 may be realized using, for example, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or the like. That is, each of the functional units 21 to 28 may be realized by hardware using a dedicated IC or the like. Also, each of the functional units 21 to 28 may be realized by using software and hardware in combination. Also, each of the functional units 21 to 28 is a conceptual component. The functions executed by one component may be distributed among a plurality of components. Also, the functions possessed by a plurality of components may be integrated into one component.

[0019] The information acquisition unit 21 acquires various types of information. For example, the information acquisition unit 21 acquires map information 32 from the memory 3. Also for example, the information acquisition unit 21 acquires the information input and selected by the user from the operation device 6. Also for example, the information acquisition unit 21 acquires position information indicating the current position of the navigation device 1 from the position information acquisition device 5. Also for example, the information acquisition unit 21 acquires the response of the response generation server SV1 via the communication unit 4.

[0020] The destination setting unit 22 sets the destination of the guidance route. Specifically, for example, the destination setting unit 22 extracts candidates for the destination that match the search conditions of the destination input by the user via the operation device 6. Further, the destination setting unit 22 sets the destination selected by the user via the operation device 6.

[0021] The route search unit 23 searches for a route (guidance route) from the departure point (the current position of the navigation device 1 at the time of route search) to the destination. Specifically, for example, the route search unit 23 uses the destination set by the destination setting unit 22, the position information indicating the current position of the navigation device 1 acquired by the position information acquisition device 5, and the map information 32 stored in the memory 3 to search for a route (guidance route) from the departure point to the destination.

[0022] The time measurement unit 24 measures time information indicating the current time. For example, the time measurement unit 24 measures time using the time data included in the GPS information received by a GPS (Global Positioning System) receiver described later, or measures time using a timer that counts the clock signal used in a computer. Further, the time measurement unit 24 calculates the required time from the departure point to the destination. Specifically, for example, the time measurement unit 24 calculates the required time from the departure point to the destination based on the distance of the route from the departure point to the destination based on the map data, information such as the legal speed limit of each road link on the guidance route, etc.

[0023] The route guidance unit 25 provides directions (guided routes) from the starting point to the destination. Specifically, for example, the route guidance unit 25, via the output control unit 28, displays an image on the video output device 7 in which the guided route is superimposed on a map image of the vicinity of the navigation device 1's current location. It also outputs voice guidance related to route guidance, such as right turn instructions and left turn instructions, to the voice output device 8 when turning right or left, thereby guiding the user on the route (guided route) from the starting point to the destination.

[0024] The prompt generation unit 26 uses the prompt template 33 to generate a prompt appropriate for performing the landscape guidance function, and sends the generated prompt to the response generation server SV1 via the communication unit 4.

[0025] The landscape guidance unit 27 uses the response from the response generation server SV1 to provide voice guidance for the target object (landmark). Specifically, for example, the landscape guidance unit 27 converts the landscape information 36, which is text data, into voice data, and provides the converted voice data to the output control unit 28 to realize voice guidance for the target object (landmark).

[0026] The output control unit 28 controls the output from the video output device 7 (video output) and the output from the audio output device 8 (audio output).

[0027] Memory 3 is composed of a non-volatile storage medium such as flash memory or a hard disk drive. Memory 3 stores various programs 31, map information 32, prompt templates 33, acquired information 34, visible range information 35, and scenery information 36. Part or all of Memory 3 may be composed of a portable recording medium and be detachable from the main body of the navigation device 1.

[0028] Program 31 includes a general navigation program as well as a program that provides voice guidance on scenic information. Generally, the program is written to memory 3 at the factory or elsewhere during the manufacturing of the navigation device 1. However, it is also possible to download the program from a server and write it to memory 3, or to have the navigation device 1 read a storage medium such as an optical disc or memory card on which the program is written and have the navigation device 1 write it to memory 3, or to use a data writer that reads the program from these storage media and writes it to the memory 3 of the navigation device 1.

[0029] Map information 32 is map-related data used for route guidance, etc., and includes road information, facility information, etc. Road information includes various information about each road (each link: unit road), such as distance of the road link, legal speed limit, road type, etc. Facility information includes event venue information. Map data is divided into predetermined areas called meshes (for example, 100m square), and each mesh is filed separately. Map data for the area used for processing, etc., is selected from map information 32 and used.

[0030] The prompt template 33 is a template (standard phrase) for multiple prompts appropriate for performing the landscape guidance function. Some of the prompt templates (standard phrases) have insertion points where a portion of the input information 34 is inserted. The prompt template 33 also includes information indicating which prompts are used and in what order. The prompt template 33 can be set in advance by the designer or the user. If the user sets it in advance, the setting is input to the navigation device 1 using the operating device 6.

[0031] Furthermore, prompt template 33 may include a query prompt that queries the answer generation server SV1 for a suitable prompt to obtain an answer regarding specific information, and prompts obtained as a result of sending the query prompt to the answer generation server SV1 may also be included in prompt template 33. An example of a query prompt might be: "When driving along a certain road, you may come across a scenic landmark. I would like you to find such a landmark; please tell me a suitable prompt."

[0032] Furthermore, the prompt template 33 may include a confirmation prompt that prompts the response generation server SV1 to verify the accuracy of the response. By using the response from the response generation server SV1 to the confirmation prompt, it becomes possible to prevent, for example, the use of inaccurate responses for objects (landmarks) in voice guidance.

[0033] The input information 34 is the information obtained by the navigation device 1 that is necessary for performing the landscape guidance function. Figure 3 shows an example of the obtained information 34.

[0034] The visible range information 35 represents the range in which the occupants of vehicle V1 can see the target object (landmark). Figure 4 shows an example of the visible range information 35.

[0035] Landscape information 36 is text data, which is a description (guide) of the target object (landmark) from the response of the response generation server SV1. Landscape information 36 is divided by target object (landmark), for example, as shown in Figure 5.

[0036] The communication unit 4 transmits and receives arbitrary signals with the answer generation server SV1. Although the controller 2 can transmit and receive arbitrary information with the answer generation server SV1 using the communication unit 4, the description of the communication unit 4 may be omitted below.

[0037] The location information acquisition device 5, the operation device 6, the video output device 7, the audio output device 8, and the vehicle speed sensor 9 are connected to the navigation device 1 via an in-vehicle network or the like.

[0038] The location information acquisition device 5 acquires location information indicating the current location of the navigation device 1. For example, a GPS receiver can be used as the location information acquisition device 5. The location information includes, for example, latitude and longitude information.

[0039] The operating device 6 is a device that allows the user to input commands, information, etc., to operate the navigation device 1. For example, the operating device 6 can be a touch panel, mechanical buttons, etc. The operating device 6 may be configured by placing a touch panel on the screen of the video output device 7. Alternatively, the operating device 6 may be configured by arranging mechanical buttons around the screen of the video output device 7.

[0040] The video output device 7 is a display device that outputs video based on video signals sent from the output control unit 28, and its screen faces the interior of the vehicle. Examples of video output devices 7 include liquid crystal displays, organic EL (Electro-Luminescence) displays, and head-up displays that project images onto the vehicle's windshield. The navigation device 1 can display, for example, maps, road information, and the vehicle's location on the screen of the video output device 7. Note that the video output device 7 is not limited to a device installed in the vehicle. For example, the navigation device 1 may have wireless communication functions such as Wi-Fi® communication or Bluetooth® communication, and may communicate wirelessly with a portable device brought into the vehicle, with the display unit of the portable device being used as the video output device 7.

[0041] The audio output device 8 is a device that outputs audio based on an audio signal sent from the output control unit 28, and faces the vehicle interior. For example, a speaker can be used as the audio output device 8. The navigation device 1 can provide audio information such as maps, road information, and the vehicle's location from the audio output device 8. Note that the audio output device 8 is not limited to a device installed in the vehicle. For example, the navigation device 1 may have wireless communication capabilities such as Wi-Fi communication or Bluetooth communication, and the navigation device 1 may communicate wirelessly with a portable device brought into the vehicle interior, and the audio output unit of the portable device may be used as the audio output device 8.

[0042] The vehicle speed sensor 9 detects the speed of the vehicle V1.

[0043] [Answer generation server] Figure 6 shows an example configuration of the response generation server SV1. The response generation server SV1 comprises a communication unit 110, a memory 120, and a controller 130.

[0044] The communication unit 110 transmits and receives arbitrary signals to and from the navigation device 1. The controller 130 can also transmit and receive arbitrary information to and from the navigation device 1 using the communication unit 110; however, the description of the communication unit 110 may be omitted below.

[0045] Memory 120 is comprised of non-volatile memory such as ROM or flash memory, and volatile memory such as RAM. Memory 120 stores program 121 and LLM 122.

[0046] Program 121 is a program that should be executed by controller 130.

[0047] The controller 130 comprehensively controls the operation of each component in the response generation server SV1. The controller 130 is equipped with a processing unit including a CPU as a hardware resource. The controller 130 has an LLM execution unit 131.

[0048] When the LLM execution unit 131 receives a prompt from the navigation device 1, it uses the prompt as input data to execute the LLM 122 and causes the LLM 122 to generate a response. Once the LLM 122 generates a response, the controller 130 transmits the response generated by the LLM 122 to the navigation device 1.

[0049] [Voice guidance processing] Figure 7 is a flowchart of the voice guidance processing performed by controller 2. This voice guidance processing is achieved when controller 2 executes the program 31 described above. The voice guidance processing shown in Figure 7 starts when route guidance is initiated while the scenic guidance function of navigation device 1 is turned on.

[0050] First, in step S10, the controller 2 (prompt generation unit 26) generates a portion of multiple prompts and sends them to the response generation server SV1. The multiple prompts generated by the prompt generation unit 26 include a first prompt for identifying the range in which objects (landmarks) visible to the user (the user riding in the vehicle V1) as they travel along the guidance route, and a second prompt for obtaining guidance text about objects (landmarks) of a length corresponding to the visible range. The first prompt may be one prompt or multiple prompts. The second prompt may also be one prompt or multiple prompts, similar to the first prompt. Once the processing in step S10 is complete, the process proceeds to step S20.

[0051] In step S20, the controller 2 (landscape guidance unit 27) determines whether it has received a response from the response generation server SV1. If a response is received, the process proceeds to step S30.

[0052] In step S30, the controller 2 (landscape guidance unit 27) stores the response from the response generation server SV1 in memory 3 as an element of acquired information 34, visible range information 35, or landscape information 36. Once the processing in step S30 is complete, the process proceeds to step S40.

[0053] In step S40, the controller 2 (landscape guidance unit 27) determines whether all prompts have been sent. If the prompts have not been sent, the process returns to step S10; if the prompts have been sent, the process proceeds to step S50.

[0054] The prompts generated by the controller 2 (prompt generation unit 26) in the processing loop from steps S10 to S40 may include, for example, the following:

[0055] The first prompt is: "I want to know exactly from where on the road a landmark is visible. Please tell me one of the best websites." Such a prompt would include a request for a website that helps identify the range from which a landmark is visible when the user (a user riding in vehicle V1) travels along the guided route, and is expected to improve the accuracy of identifying that visible range.

[0056] The second prompt is: "I believe that when driving on a certain road, there are points where a certain landmark becomes visible and points where it disappears. Please tell me the steps to accurately confirm these points using the website referenced in Information 34."

[0057] The third prompt is "Please use the following instructions (referenced website in acquired information 34) to investigate." With such a prompt, the first prompt will include instructions to use a website obtained from the response of the response generation server SV1 to a query about a website that helps identify the range in which visible objects (landmarks) can be seen when the user (the user riding in vehicle V1) travels along the guided route, thereby improving the accuracy of identifying the visible range.

[0058] The fourth prompt is: "Travel from (current location in Information 34) to (destination in Information 34) using (means of transport in Information 34) along (road name in Information 34). The destination is (destination in Information 34). Your current location is (current location in Information 34). Please list any landmarks that are visible along the route from your current location to your destination."

[0059] The fifth prompt states: "While traveling along (Road name from acquired information 34) from (Current location from acquired information 34) to (Destination from acquired information 34), the point where a landmark becomes visible will be defined as the landscape confirmation point. The point where it is no longer visible will be defined as the landscape end point. For each landmark, please confirm the latitude and longitude of the landscape start point and landscape end point. The location information for the landscape start point and landscape end point can be confirmed using (Confirmation procedure for acquired information 34)."

[0060] The prompt for the sixth round reads: "The current vehicle speed is (vehicle speed from acquired information 34). Please check the current road conditions and, taking the current vehicle speed into consideration, confirm the estimated arrival times at the start and end points of each landmark's scenery. Also, for each landmark, please confirm the time difference between the estimated arrival time at the start point and the estimated arrival time at the end point. This time difference will be defined as the scenery confirmation time. The unit of the scenery confirmation time should be seconds. You can find the estimated times on (the website referenced in acquired information 34)." With prompts like this, the vehicle V1's speed information will be included in the second prompt, making it possible to obtain guidance text of a length corresponding to the vehicle V1's speed.

[0061] The seventh prompt reads: "Please provide a description of each landmark. The length of each description should be approximately the number of characters obtained by multiplying the landscape confirmation time by 5. For example, if you reach the starting point of another landmark B while in the section where landmark A is visible, please adjust the landscape confirmation time for landmark A to the difference between the estimated arrival time at the starting point of landmark A and the estimated arrival time at the starting point of landmark B." This character limit for prompts allows the navigation device 1 to provide voice guidance for the objects (landmarks) within the visible range described above using the answers from the answer generation server SV1, thereby enhancing the realism of the voice guidance regarding the objects (landmarks). Furthermore, according to the note in such prompts, if there is an overlap in the visible range described above for two objects (landmarks), the second prompt will include a prompt to obtain a guidance text for one of the two objects, with a length corresponding to the range obtained by excluding the overlapping portion from the visible range described above. This eliminates the overlapping of two separate audio instructions for each of the two landmarks, so users no longer need to distinguish between the two separate audio instructions for each landmark.

[0062] In the seventh prompt, the note may be omitted. In this case, if there is an overlap in the visible range for the two objects (landmarks) as described above, it is desirable for the output control unit 28 to provide voice guidance for each of the two objects (landmarks) in a different manner. This makes it easier for the user to distinguish between the two voice guidance statements for each of the two objects (landmarks) when the two guidance statements for each object (landmark) overlap and are voiced together. Examples of different manners include providing voice guidance for one of the two objects (landmarks) in a female voice and voice guidance for the other object (landmark) in a male voice, i.e., using different sound quality; providing voice guidance for one of the two objects (landmarks) from a speaker (part of the voice output device) installed on the right side of the vehicle V1 and voice guidance for the other object (landmark) from a speaker (part of the voice output device) installed on the left side of the vehicle V1, i.e., using different sound generation locations; or using different sound quality and sound generation locations.

[0063] The eighth prompt is "Please update the estimated arrival time at the start point of the scenery and the estimated arrival time at the end point of the scenery according to the (checking cycle for the estimated arrival time at the start point and end point of the scenery in the acquired information 34)." While the flowchart shown in Figure 7 is being processed, the controller 2 (scenery guidance unit 27) constantly checks whether it has received a response from the response generation server SV1 corresponding to this eighth prompt. Once it receives a response from the response generation server SV1 corresponding to the eighth prompt, it updates the estimated arrival time at the start of the scenery and the scenery confirmation time in the visible range information 35 via interrupt processing.

[0064] Now, let's return to Figure 7 and explain the process from step S50 onwards. In step S50, the controller 2 (landscape guidance unit 27) determines whether vehicle V1 has reached the landscape starting point based on the output of the location information acquisition device 5. Instead of determining whether vehicle V1 has reached the landscape starting point, it may be possible to determine whether the estimated arrival time at the landscape starting point has arrived, but considering the prediction error, it is preferable to determine whether vehicle V1 has reached the landscape starting point.

[0065] When vehicle V1 reaches the scenic starting point, the process proceeds to step S60, where the controller 2 (scenic guidance unit 27) uses the scenic information 36 to provide an audio description (guide text) of the landmark corresponding to the scenic starting point reached by vehicle V1. Once the processing in step S60 is complete, the process proceeds to step S70.

[0066] In step S70, the controller 2 (landscape guidance unit 27) refers to the visible range information 35 to determine if there are any objects (landmarks) for which voice guidance is not provided. If there are objects (landmarks) for which voice guidance is not provided, the process returns to step S50. If there are no objects (landmarks) for which voice guidance is not provided, the voice guidance process for the objects (landmarks) is terminated.

[0067] <Notes, etc.> The various technical features disclosed in the embodiments for carrying out the invention as specified herein can be modified in various ways without departing from the spirit of the technical creation. Furthermore, the multiple embodiments and modifications disclosed in the embodiments for carrying out the invention as specified herein may be combined to the extent possible.

[0068] For example, LLM122 may be configured as a multimodal LLM capable of image recognition, and the prompts generated by the controller 2 (prompt generation unit 26) in the processing loop of steps S10 to S40 may include the following prompts 1 to 4, and the method for confirming the latitude and longitude of the location information of the landscape start point and landscape end point may be changed for each landmark.

[0069] Prompt 1: "Please obtain images from a website of landmarks that are visible along the route you will take from your current location to your destination."

[0070] Prompt 2: "Please convert the acquired landmark images into viewpoint-transformed images viewed from a virtual viewpoint, which is set based on the positional relationship (distance, angle) between each sampling point (sampled every 100m along the guidance route) and the landmark."

[0071] Prompt 3: "Obtain Street View images from the website at each sampling point, sampled every 100m along the guidance route, and check whether objects similar to the corresponding viewpoint transformation image are included in the Street View images at each sampling point." The response from the response generation server SV1 to this prompt 3 allows us to check whether the landmark is not obscured by an obstacle (obstruction) located between the landmark and the vehicle V1.

[0072] Prompt 4: "Select the sampling point on the side of the guidance route closest to your current location where an object similar to the viewpoint-transformed image can be found as the landscape start point, and the sampling point on the side of the guidance route closest to your destination where an object similar to the viewpoint-transformed image can be found as the landscape end point. For each landmark, please confirm the latitude and longitude of the landscape start point and landscape end point."

[0073] According to the above modification, the first prompt includes instructions to obtain an image of the target object (landmark) from the website and use the image of the target object (landmark) to identify the range in which the target object (landmark) is visible when the user (user riding in vehicle V1) travels along the guided route. As a result, obstacles (obstructions) located between the landmark and vehicle V1 are taken into consideration when identifying the visible range, and an improvement in the accuracy of identifying the visible range can be expected.

[0074] In the embodiment described above, the navigation device 1, which is an example of a voice guidance device, is used in a vehicle V1, but the location of use of the voice guidance device is not limited to a vehicle V1. For example, it may be used on a bicycle, train, or ship, and the voice guidance device may be a mobile device carried by a pedestrian.

[0075] Since there may be cases where the number of objects (landmarks) is too large to provide all the voice guidance, you may include a prompt in the first prompt that says, "Please narrow down the landmarks to three, prioritizing well-known ones." Alternatively, you may divide the guidance route into multiple sections and then narrow down the number of objects (landmarks). For example, you may include a prompt in the first prompt that says, "Divide the guidance route in half by distance, and in the first half of the guidance route, narrow down the landmarks to three, prioritizing well-known ones, and in the second half of the guidance route, narrow down the landmarks to three, prioritizing well-known ones."

[0076] Since there is likely to be some variation in the number of characters in the descriptions of each landmark, the landscape information unit 27 may adjust the reading speed according to the number of characters in the description of each landmark when converting it into audio data. For example, the more characters there are, the faster the reading speed should be. Alternatively, the reading speed may be adjusted according to the speed of the vehicle V1 immediately before the description of each landmark is converted into audio data. For example, the faster the speed of the vehicle V1, the faster the reading speed should be.

[0077] In the embodiment described above, the navigation device 1 determined whether it had reached the scenic starting point and started voice guidance at the time of arrival at the scenic starting point. However, a prompt may be generated instructing the response generation server SV1 to send a description of the landmark at the time of arrival at the scenic starting point, and the timing of starting voice guidance may be left to the response generation server SV1. [Explanation of symbols]

[0078] 1. Navigation system SV1 ···Answer generation server SV2...Website server SYS1... Guidance System

Claims

1. A first prompt is generated to identify the range of visible objects as the user travels along the guided path. A second prompt is generated to obtain a guide text regarding the object of a length corresponding to the visible range. The generated first prompt and second prompt are sent to the answer generation server including the LLM. A voice guidance device that uses the response from the response generation server to provide voice guidance to the object within the visible range.

2. The voice guidance device according to claim 1, wherein the first prompt includes an inquiry to a website that helps identify the visible area.

3. The voice guidance device according to claim 2, wherein the first prompt includes instructions for using a website obtained from the response of the response generation server to the inquiry.

4. The aforementioned LLM is a multimodal LLM capable of image recognition. The voice guidance device according to claim 1, wherein the first prompt includes an instruction to obtain an image of the object from a website and to use the image of the object to specify the visible range.

5. The aforementioned voice guidance device is installed in the vehicle, The voice guidance device according to claim 1, wherein the second prompt includes vehicle speed information.

6. Using the overlapping portion of the visible range with respect to the two aforementioned objects, The voice guidance device according to claim 1, wherein the second prompt includes a prompt for obtaining a guidance statement relating to one of the two objects, the guidance statement having a length corresponding to the range obtained by excluding the overlapping portion from the visible range.

7. Using the overlapping portion of the visible range with respect to the two aforementioned objects, The voice guidance device according to claim 1, which provides voice guidance to each of the two aforementioned objects in different manners.

8. A first prompt is generated to identify the range of visible objects as the user travels along the guided path. A second prompt is generated to obtain a guide text regarding the object of a length corresponding to the visible range. The generated first prompt and second prompt are sent to the answer generation server including the LLM. A voice guidance method that uses the response from the response generation server to provide voice guidance to the object within the visible range.

9. To generate a first prompt to identify the range of visible objects as the user follows the guided path, To generate a second prompt for obtaining a guide text about the object of a length corresponding to the visible range, The generated first prompt and second prompt are sent to the answer generation server including LLM, Using the response from the aforementioned response generation server, the object is given voice guidance within the visible range. A program that causes a computer to execute something.

10. A voice guidance system including a voice guidance device and a response generation server including an LLM, The aforementioned voice guidance device, A first prompt is generated to identify the range of visible objects as the user travels along the guided path. A second prompt is generated to obtain a guide text regarding the object of a length corresponding to the visible range. The generated first prompt and second prompt are sent to the response generation server. A voice guidance system that uses the response from the response generation server to provide voice guidance to the object within the visible range.

Images