Information processing device, information processing method, and program

The information processing device uses history information to generate prompts and acquire answers through an interactive model, addressing the limitation of existing voice dialogue systems by providing contextually relevant responses.

JP2025151554APending Publication Date: 2025-10-09PIONEER IP
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2024053052
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-03-28
Publication Date
2025-10-09

AI Technical Summary

Technical Problem

Existing voice dialogue systems struggle to provide answers that utilize past dialogues with users, limiting their effectiveness in providing relevant responses.

Method used

An information processing device that generates prompts using history information associating past questions with locations where they were asked, and utilizes an interactive generative model to acquire answers, allowing for the creation of prompts and retrieval of relevant responses.

Benefits of technology

Enables the system to provide appropriate answers by leveraging past interactions, reducing processing load, and ensuring responses are contextually relevant.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2025151554000001_ABST
    Figure 2025151554000001_ABST
Patent Text Reader

Abstract

To cause a system to provide an appropriate response utilizing past conversations.SOLUTION: An information processing device 100 comprises a creation unit 116 and a response acquisition unit 117. The generation unit 116 generates a prompt using at least a part of history information which associates the past questions with the locations where the past questions are asked and new questions. The response acquisition unit 117 acquires a response to the generated prompt using a dialogue-based generation model, which generates responses to prompts.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] For example, Patent Document 1 discloses a voice dialogue system for having a voice dialogue with a user. The voice dialogue system described in Patent Document 1 includes a dialogue scenario storage means for storing a plurality of dialogue scenarios, and a dialogue generation means for generating a dialogue in response to a user utterance based on the results of speech recognition. A dialogue scenario is a set of three items: the content of a first system utterance, the content of a user utterance expected as a response to the first system utterance, and the content of a second system utterance that is a response to the expected user utterance. The dialogue generation means determines whether the user utterance is expected as a response to the previous system utterance, and if so, generates a second system utterance defined in the dialogue scenario as a response to the user utterance, as a dialogue in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-54790 Summary of the Invention [Problem to be solved by the invention]

[0004] The voice dialogue system described in Patent Document 1 uses dialogue scenarios stored in advance, which makes it difficult to make the system provide answers that utilize past dialogues with the user.

[0005] In view of the above-mentioned problems, one example of the objective of the present disclosure is to enable a system to provide an appropriate response by utilizing past dialogues. [Means for solving the problem]

[0006] The invention described in claim 1 is a generating unit that generates a prompt using at least a portion of history information that associates past questions with locations where the past questions were asked, and a new question; an answer acquisition unit that acquires an answer to the generated prompt using an interactive generation model that generates an answer to the prompt; It is an information processing device.

[0007] The invention described in claim 13 is One or more computers storing information relating a past question to a location where the past question was asked as history information in a storage unit; creating a prompt using at least a portion of the historical information and a new question; Obtaining answers to the generated prompts using an interactive generative model that generates answers to the prompts. It is an information processing method.

[0008] The invention described in claim 14 is On one or more computers, storing information relating a past question to a location where the past question was asked as history information in a storage unit; creating a prompt using at least a portion of the historical information and a new question; A program for executing a program for obtaining an answer to a prompt created using an interactive generative model that generates an answer to the prompt. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of an information processing system according to an embodiment. [Figure 2] FIG. 1 is a diagram illustrating an example of a functional configuration of an information processing device according to an embodiment. [Figure 3] 10 is a flowchart illustrating an example of information processing according to an embodiment. [Figure 4] 10 is a flowchart illustrating an example of information processing according to an embodiment. [Figure 5] FIG. 1 is a diagram illustrating an example of the physical configuration of an information processing apparatus according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In all drawings, like components are denoted by like reference numerals, and descriptions thereof will be omitted where appropriate.

[0011] 1 is a diagram showing an example of the configuration of an information processing system S1 according to an embodiment. The information processing system S1 is a system for interacting with a user. The information processing system S1 includes a dialogue-based generation device 20, a route search device 30, and an information processing device 100.

[0012] The interaction-based generation device 20, the route search device 30, and the information processing device 100 are connected to one another via a communication network NT. As a result, the interaction-based generation device 20, the route search device 30, and the information processing device 100 transmit and receive information to and from one another via the communication network NT. The communication network NT may be configured as a wired network, a wireless network, or a combination of these, for example.

[0013] (Regarding the interactive generation device 20) The dialogue generation device 20 is a device that generates a response to a prompt (a command statement). The dialogue generation device 20 includes, for example, a dialogue generation model. The dialogue generation model is a model that generates a response to a prompt, and is, for example, a machine learning model that generates a response to a prompt when the prompt is input.

[0014] For example, when the dialogue generation device 20 receives a prompt from the information processing device 100, the dialogue generation device 20 generates an answer to the prompt. Then, the dialogue generation device 20 transmits the generated answer to the information processing device 100.

[0015] The information processing device 100 may also have the functions of the dialogue-based generation device 20.

[0016] (Regarding the route search device 30) The route search device 30 is a device that searches for a route from a certain position to a destination. For example, when the route search device 30 receives the current position and the destination from the information processing device 100, it searches for a route from the current position to the destination. Then, the route search device 30 transmits the route to the destination, which is the search result, to the information processing device 100.

[0017] The information processing device 100 may also have the functions of the route search device 30.

[0018] (Regarding information processing device 100) The information processing device 100 is a device used by a user (questioner), which creates a prompt in response to the user's question and obtains an answer to the prompt using an interactive generation model. The information processing device 100 is, for example, a smartphone or a tablet terminal, but is not limited to these.

[0019] For example, when the information processing device 100 acquires a voice containing a user's question, it creates a prompt corresponding to the question and transmits it to the dialogue generation device 20. Then, the information processing device 100 acquires an answer to the prompt from the dialogue generation device 20, thereby acquiring the answer to the prompt using the dialogue generation model.

[0020] 2 is a diagram illustrating an example of the functional configuration of an information processing device 100 according to an embodiment. The information processing device 100 includes, for example, a destination reception unit 111, a question acquisition unit 112, a position acquisition unit 113, a history storage unit 114, a storage control unit 115, a creation unit 116, an answer acquisition unit 117, a display unit 118, an answer output unit 119, and a route guidance unit 120.

[0021] The destination receiving unit 111 receives a destination from a user. The destination receiving unit 111 may receive the destination by inputting characters or the like, or may receive the destination by voice, for example.

[0022] The question acquisition unit 112 acquires a question from a user.

[0023] The position acquisition unit 113 acquires the current position. To detect the current position, a general technology such as a GPS (Global Positioning System) function or a wireless LAN (Local Area Network) function may be used.

[0024] For example, when the information processing device 100 is carried or mounted on a vehicle, the location acquired by the location acquisition unit 113 is, for example, the current location of the user (questioner).

[0025] The history storage unit 114 is a storage unit for storing history information. The history information is information that associates a question with the location where the question was asked. The location where the question was asked is, for example, the current location when the question was asked. The history information may also be information that further associates the date and time when a past question was asked.

[0026] The storage control unit 115 stores the history information in the history storage unit 114. For example, the storage control unit 115 stores information associating a new question with the location where the new question was asked in the history storage unit 114 as history information.

[0027] In detail, for example, the memory control unit 115 stores, as history information, information associating a new question with a current location at the time the new question was asked, in the history storage unit 114. For example, when the question acquisition unit 112 acquires a new question from the user, the memory control unit 115 may associate the current location acquired by the location acquisition unit 113 with the acquired new question and store the associated information in the history storage unit 114.

[0028] The memory control unit 115 may store in the history memory unit 114 the historical information for all of the questions acquired by the question acquisition unit 112, or may store in the history memory unit 114 the historical information for some of the questions acquired by the question acquisition unit 112.

[0029] For example, when predetermined storage conditions are met, the storage control unit 115 may store history information in the history storage unit 114. The storage conditions may include at least one of the following storage condition examples 1 and 2. Note that the storage conditions are not limited to the following examples.

[0030] (Memory condition example 1) The storage condition may be, for example, that the dialogue-based generation model generates an answer related to a specific location in response to a new question. In this case, when the dialogue-based generation model generates an answer related to a specific location in response to a new question, the storage control unit 115 may store information associating the new question with the location where the new question was asked as history information in the history storage unit 114.

[0031] The specific location includes, for example, at least one of a point, a facility, and an area. A point may be represented, for example, by an address, by latitude and longitude, or by the name of a point such as an intersection. A facility may be represented, for example, by the name of a building, a station name, a parking lot, or by the location of the facility such as an address. An area may be represented, for example, by the name of a city, ward, town, or village.

[0032] The answers for a particular location may include at least one of suggestions and search results.

[0033] (Memory condition example 2) The storage condition may be, for example, that the location where the new question was asked is within a predetermined range from the destination of the user (questioner). The location where the new question was asked is, for example, the current location of the user (questioner) when the new question was asked. In this case, if the location where the new question was asked is within a predetermined range from the destination of the user (questioner), the storage control unit 115 may store information associating the new question with the location where the new question was asked in the history storage unit 114 as history information.

[0034] The creation unit 116 creates a prompt using at least a part of the history information and the new question. For example, the creation unit 116 creates a prompt that includes at least a part of the history information and the new question.

[0035] In detail, for example, the creation unit 116 includes a question selection unit 116a and a prompt creation unit 116b.

[0036] The question selection unit 116a selects past questions that satisfy predetermined selection conditions from among past questions. The selection conditions may include at least one of the following selection condition examples 1 to 3. Note that the selection conditions are not limited to the following examples.

[0037] (Example 1 of selection criteria) The selection condition may be, for example, that a location within a predetermined distance range from the location where the new question was asked is associated with a past question in the history information. As described above, the location where the new question was asked is, for example, the current location of the user (questioner) when the new question was asked. The predetermined distance range may be determined in advance or may be set by the user.

[0038] In this case, the question selection unit 116a may select, for example, from past questions included in the history information, a past question that is associated in the history information with a location within a predetermined distance range from the location where the new question was asked.

[0039] (Example 2 of selection criteria) The selection condition may be, for example, that the previous question contains a keyword common to the new question. The keyword may be extracted according to a predetermined rule, such as a noun or proper noun included in the question. However, the method for extracting keywords is not limited to these.

[0040] In this case, the question selection unit 116a may extract keywords from, for example, past questions included in the history information and the new question.The question selection unit 116a may then select, from the past questions included in the history information, past questions from which keywords that are all or partly common to the keywords extracted from the new question have been extracted.

[0041] It should be noted that the keywords of past questions may be extracted, for example, by the storage control unit 115 when storing a new question as history information in the history storage unit 114, and may be further associated in the history information.

[0042] (Example 3 of selection criteria) The selection condition may be, for example, that a date and time within a predetermined time range from the date and time when the new question is asked is associated with a past question.

[0043] In this case, the question selection unit 116a may select, for example, from past questions included in the history information, a past question associated with a date and time within a predetermined time range from the date and time when the new question was asked.

[0044] The prompt generator 116b generates a prompt including the selected past question, the location where the past question was asked, and the new question.

[0045] The prompt generator 116b may use keywords contained in or related to the past questions instead of the past questions. As described above, the keywords may be extracted according to predetermined rules, such as nouns, proper nouns, etc. contained in the questions.

[0046] The answer acquisition unit 117 acquires an answer to the generated prompt using an interactive generation model that generates an answer to the prompt.

[0047] For example, the answer acquisition unit 117 uses the dialogue generation model by sending the created prompt to the dialogue generation device 20. When the dialogue generation device 20 receives the sent prompt, it inputs the prompt into the dialogue generation model to obtain an answer to the prompt and transmits it to the information processing device 100. In this way, the answer acquisition unit 117 obtains the answer to the prompt.

[0048] The display unit 118 displays various types of information.

[0049] The answer output unit 119 outputs an answer to the prompt. For example, the answer output unit 119 outputs the answer by voice. Note that the method by which the answer output unit 119 outputs the answer is not limited to voice, and the answer may also be displayed on the display unit 118, for example.

[0050] When the route guidance unit 120 acquires the destination of the user (questioner), it provides guidance information for guiding the route from the current position to the destination.

[0051] For example, when the destination receiving unit 111 receives a destination from a user, the route guidance unit 120 obtains a route from the current position obtained by the position obtaining unit 113 to the destination from the route search device 30. The route guidance unit 120 causes the display unit 118 to display the route to the destination, for example.

[0052] For example, when the answer acquired by the answer acquisition unit 117 includes information about a specific location, the route guiding unit 120 may provide guidance information including the specific location and the route.

[0053] The guidance information may include a route from the current location to the specific point acquired by the location acquisition unit 113. In this case, for example, the route guidance unit 120 may acquire the route from the current location to the specific point from the route search device 30. The route guidance unit 120 may, for example, cause the display unit 118 to display the route to the specific point.

[0054] (Example of processing operation of information processing system S1)

[0055] The information processing system S1 executes information processing for interacting with a user.

[0056] 3 and 4 are flowcharts showing an example of information processing according to an embodiment. The information processing is started, for example, when the information processing device 100 executes a predetermined application program. The information processing is repeatedly executed in response to, for example, a user instruction, until the predetermined application program ends. Note that triggers for starting and ending the information processing are not limited to these.

[0057] The destination receiving unit 111 receives a destination from a user (step S101).

[0058] The position acquisition unit 113 acquires, for example, the current position at the time when the destination is accepted in step S101 (step S102).

[0059] The route guiding unit 120 provides guidance information including a route from the current position to the destination, for the destination and current position acquired in steps S101 and S102 (step S103).

[0060] For example, the route guidance unit 120 transmits the destination and current position acquired in steps S101 and S102 to the route search device 30. The route guidance unit 120 acquires the route from the current position to the destination searched by the route search device 30. The route guidance unit 120 displays the route from the current position to the destination on the display unit 118. The route guidance unit 120 may also display a map showing the route from the current position to the destination on the display unit 118.

[0061] The question acquisition unit 112 acquires a new question from the user (step S104).

[0062] For example, the question acquisition unit 112 acquires a new question by voice, for example. Note that the new question is not limited to being acquired by voice, and may be acquired by inputting, for example, characters or the like.

[0063] The location acquisition unit 113 acquires, for example, the current location at the time when the new question is acquired in step S104 (step S105).

[0064] The current location acquired in step S105 is the location where the new question was asked.

[0065] The creation unit 116 creates a prompt using at least a part of the history information and the new question acquired in step S104 (step S106).

[0066] For example, suppose that the past dialogue between the user and the system includes the following Past Dialogue Example 1 and a new question 1 that follows.

[0067] (Past dialogue example 1) User's previous question 1A: "I want to eat ramen. Is there anywhere nearby?" System answer 1A: "There are three ramen shops nearby. Ramen shop X is..."

[0068] Assume that this past question 1A was asked while moving near area A. If the information processing device 100 is a smartphone or the like moving in a vehicle, it may be asked while the vehicle is moving. In this case, the history information includes, for example, information associating past question 1A "I want to eat ramen. Is there anywhere nearby?" with the location "Area A" where past question 1A was asked.

[0069] After past dialogue example 1, suppose that new question 1B acquired in step S104 is "Where was that ramen shop we just went to?" The creation unit 116 creates prompt 1, for example, including past question 1A "I want to eat ramen. Is there one nearby?", the location "Area A" where past question 1A was asked, and new question 1B "Where was that ramen shop we just went to?"

[0070] For example, suppose that the past dialogue between the user and the system includes the following past dialogue example 2, followed by a new question 2B.

[0071] (Past dialogue example 2) The system suggests: "It's almost lunchtime. Would you like something to eat?" User Question 2A: "Is there anything?" System response 2A: "There are three ramen shops nearby. Would you like to stop by X ramen shop?" User response: "Not now"

[0072] For example, this past question 2A was asked while moving near area A. In this case, the history information includes, for example, information associating past question 2A "Is there anything?" with the location "Area A" where past question 2A was asked.

[0073] After past dialogue example 2, suppose that new question 2B acquired in step S104 is "Where was that ramen shop we just went to?" The creation unit 116 creates prompt 2, for example, including past question 2A "Is there anything?", the location "Area A" where past question 2A was asked, and new question 2B "Where was that ramen shop we just went to?"

[0074] For example, suppose that the past dialogue between the user and the system includes the following past dialogue example 3, followed by a new question 3B.

[0075] (Past dialogue example 3) System announcement: "You're almost at your destination." User Question 3A: "Is there anywhere to eat?" System answer 3B: "XX, YY, and ZZ are popular."

[0076] Assume that this past question 3A was, for example, parking lot Q at destination P. In this case, the history information includes, for example, information associating past question 2A "Is there anywhere to eat?" with the location "P" (destination) and "parking lot Q" where past question 2A was asked.

[0077] After past dialogue example 3, suppose that new question 3B acquired in step S104 is "What's around here?" The creation unit 116 creates prompt 3, which includes, for example, past question 3A "Is there anywhere to eat?", the location "P" (destination) and "parking lot Q" where past question 2A was asked, and new question 3B "What's around here?"

[0078] In step S106, the question selection unit 116a may select a past question that satisfies a predetermined selection condition from among past questions, and the prompt creation unit 116b may create a prompt that includes the selected past question, the location where the past question was asked, and the new question.

[0079] This allows a user to select a past question related to a new question and a location where the past question was asked, and to create a prompt including the selected past question and the location where the past question was asked. This allows a user to associate a past question related to a new question and a location where the past question was asked with the new question and obtain an answer to the new question.

[0080] The answer acquisition unit 117 acquires, for example, an answer to the prompt created in step S106 (step S107).

[0081] For example, when the above-mentioned prompt 1 is created in step S106, the answer acquisition unit 117 transmits prompt 1 to the dialogue generation device 20. The dialogue generation device 20 inputs prompt 1 into the dialogue generation model, thereby acquiring, for example, answer 1B below as an answer to the prompt.

[0082] System response 1B: "I just searched for ramen restaurants near area A."

[0083] The dialogue-based generation device 20 transmits the answer 1B to the information processing device 100. The answer acquisition unit 117 acquires the answer 1B from the dialogue-based generation device 20.

[0084] The same applies when prompt 2 or prompt 3 is created instead of prompt 1. The answer acquisition unit 117 sends prompt 2 or prompt 3 to the dialogue generation device 20. By inputting prompt 2 or prompt 3 into the dialogue generation model, the dialogue generation device 20 acquires, for example, answer 2B or answer 3B below as an answer to the prompt.

[0085] System response 2B: "I just searched for ramen restaurants in Area A."

[0086] System answer 3B: "The townscape of △△△ is famous."

[0087] The dialogue-based generation device 20 transmits the answer 2B or the answer 3B to the information processing device 100. The answer acquisition unit 117 acquires the answer 2B or the answer 3B from the dialogue-based generation device 20.

[0088] In this way, by using history information that associates the location where a past question was asked with the past question, a prompt including the location where the past question was asked can be created. Answers to new questions can be obtained by associating past questions with the locations where the past questions were asked. Therefore, it is possible for the system to provide appropriate answers that utilize past interactions.

[0089] The answer output unit 119 outputs the answer acquired in step S107 (step S108).

[0090] The answer output unit 119 outputs the answer by voice, for example. Note that the method by which the answer output unit 119 outputs the answer is not limited to voice, and may be, for example, display of characters, figures, or the like.

[0091] The storage control unit 115 stores information that associates the new question acquired in steps S104 and S105 with the current location at the time the new question was asked as history information in the history storage unit 114 (step S109).

[0092] Here, the storage control unit 115 may store the history information in the history storage unit 114 when the above-mentioned storage conditions are satisfied. Assume that the storage condition is the above-mentioned storage condition example 1 (the interactive generation model generates an answer related to a specific location in response to a new question). In this case, both Answer 1B and Answer 2B are answers related to the specific location "Area A."

[0093] Therefore, when there is a past dialogue example 1, a new question 1B, and an answer 1B, the memory control unit 115 stores the history information in the history memory unit 114. This history information is information that associates the new question 1B, "Where was that ramen shop just now?" with the location, "Area A," where the new question was asked.

[0094] Furthermore, when there is a past dialogue example 2, a new question 2B, and an answer 2B, the memory control unit 115 stores the history information in the history memory unit 114. This history information is information that associates the new question 2B, "Where was that ramen shop just now?" with the location, "Area A," where the new question was asked.

[0095] The above-mentioned answer 3B is an answer regarding a specific location "△△△". When the storage condition is the above-mentioned storage condition example 1 and there is a past dialogue example 3, a new question 3B, and an answer 3B, the storage control unit 115 stores the history information in the history storage unit 114. This history information is information that associates the new question 3B "What's around here?" with the location "P" (destination) and "parking lot Q" where the new question was asked.

[0096] In this way, by storing historical information including questions that have been answered about specific locations, answers to new questions can be obtained by associating them with specific locations that have been asked in the past, allowing the system to provide appropriate answers that utilize past conversations.

[0097] Here, the history information given in Past Dialogue Example 3 is an example of history information that is stored with the storage condition that the location where the new question was asked is within a predetermined range from the user's (questioner's) destination (storage condition example 2). After eating, shopping, sightseeing, etc. at the destination, the user may return to the parking lot and want to get further answers related to the destination.

[0098] In such cases, storing history information including destination-related questions can provide further answers related to the destination. Answers that overlap with previous questions can also be avoided. Therefore, the system can provide appropriate answers by utilizing past conversations.

[0099] For example, if the answer acquired in step S107 includes information about a specific location, the route guiding unit 120 provides guidance information including the specific location and the route (step S110).

[0100] For example, the route guidance unit 120 determines whether the answer acquired in step S107 includes information about a specific location. If the answer includes information about the specific location, the route guidance unit 120 provides the information about the specific location. For example, the route guidance unit 120 may superimpose the specific location on the map displayed in step S103 and display it on the display unit 118.

[0101] Furthermore, for example, if the response includes information about a specific point, the route guidance unit 120 may transmit the specific point and the route to the route search device 30, and acquire the route from the current position to the specific point searched for by the route search device 30. The route guidance unit 120 may display the route from the current position to the specific point on the display unit 118. For example, the route guidance unit 120 may superimpose the route from the current position to the specific point on the map displayed in step S103 and display it on the display unit 118.

[0102] Note that steps S101 to S103 do not have to be executed. Also, step S110 does not have to be executed. When the information processing is repeatedly executed, in the second and subsequent information processing, steps S101 to S103 may be executed when the destination is changed, and steps S104 to S109 or steps S104 to S110 may be executed repeatedly in other cases.

[0103] (Example of physical configuration of information processing device 100) The information processing device 100 is physically, for example, a general-purpose computer, a tablet PC (Personal Computer), a smartphone, or the like.

[0104] 5 is a diagram showing an example of the physical configuration of an information processing device 100 according to an embodiment. The information processing device 100 includes, for example, a bus 1010, a processor 1020, a memory 1030, a storage device 1040, a network interface 1050, an input interface 1060, and an output interface 1070.

[0105] The bus 1010 is a data transmission path for transmitting and receiving data among the processor 1020, memory 1030, storage device 1040, network interface 1050, input interface 1060, and output interface 1070. However, the method of connecting the processor 1020 and other components to each other is not limited to bus connection.

[0106] The processor 1020 is implemented by a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like.

[0107] The memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.

[0108] The storage device 1040 is an auxiliary storage device realized by a hard disk drive (HDD), a solid state drive (SSD), a memory card, a read only memory (ROM), or the like. The storage device 1040 stores program modules for realizing the functions of the device that includes it. The processor 1020 reads each of these program modules into the memory 1030 and executes them, thereby realizing the function corresponding to that program module.

[0109] The network interface 1050 is an interface for connecting a device equipped with it to a communication network NT.

[0110] The input interface 1060 is an interface for inputting information by the operator of the information processing device 100. The input interface 1060 is configured by, for example, a touch panel, a keyboard, a mouse, a microphone, and the like.

[0111] The output interface 1070 is an interface for presenting information to an operator of the information processing device 100. The output interface 1070 is configured by, for example, a liquid crystal panel, an organic EL (Electro-Luminescence) panel, a speaker, and the like.

[0112] Note that each of the dialogue-based generation device 20 and the route search device 30 may be physically configured in the same way as the information processing device 100. Furthermore, the information processing system S1 may include, in addition to the dialogue-based generation device 20 and the route search device 30, a terminal device and a server that are capable of transmitting and receiving information to and from each other via a communication network NT or the like. The terminal device is, for example, a smartphone, a tablet terminal, or the like. In this case, for example, the terminal device and the server may be provided together with the functions of the information processing device 100 described in the embodiment. Furthermore, for example, the server may be provided with all of the functions of the information processing device 100 described in the embodiment.

[0113] (Actions and Effects) As described above, according to this embodiment, the information processing device 100 includes the creation unit 116 and the answer acquisition unit 117. The creation unit 116 creates a prompt using a new question and at least a portion of history information that associates past questions with the locations at which the past questions were asked. The answer acquisition unit 117 acquires an answer to the created prompt using an interactive generation model that generates an answer to the prompt.

[0114] This allows the system to associate past questions with the locations where the past questions were asked and obtain answers to new questions, thereby enabling the system to provide appropriate answers that utilize past conversations.

[0115] According to this embodiment, the information processing device 100 includes a location acquisition unit 113 and a memory control unit 115. The location acquisition unit 113 acquires a current location. The memory control unit 115 stores information associating a new question with the current location at the time the new question was asked in the history memory unit 114 as history information.

[0116] This allows history information that associates a question with the current location at the time the question was asked to be stored in the history storage unit 114, making it possible to obtain an answer to the next question. This makes it possible for the system to provide an appropriate answer that utilizes past conversations.

[0117] According to this embodiment, the storage control unit 115 stores history information in the history storage unit 114 when a predetermined storage condition is met.

[0118] This makes it possible to prevent the processing load from increasing due to an increase in the amount of history information, and therefore reduces the processing load required for the system to provide an appropriate response that utilizes past dialogues.

[0119] According to this embodiment, the storage condition includes that the interactive generative model generates an answer for a specific location in response to a new question.

[0120] This allows the system to store historical information including questions that have been answered about specific locations, and to obtain answers to new questions by associating them with specific locations that have been asked in the past. This makes it possible for the system to provide appropriate answers that utilize past conversations.

[0121] According to this embodiment, the specific location includes at least one of a point, a facility, and an area.

[0122] This allows the system to store historical information including questions that have been answered regarding at least one of locations, facilities, and areas, and to obtain answers to new questions by associating them with at least one of locations, facilities, and areas that have been asked in the past. This makes it possible for the system to provide appropriate answers that utilize past conversations.

[0123] According to this embodiment, the answer includes at least one of a suggestion and a search result.

[0124] When an answer includes suggestions and search results, there is a high possibility that the user will ask a new question related to the answer, such as a request for more detailed information. Therefore, by storing historical information including questions that have resulted in at least one suggestion or search result for a specific location, it is possible to obtain an appropriate answer to the new question. Therefore, it is possible for the system to provide an appropriate answer by utilizing past conversations.

[0125] According to this embodiment, the storage conditions include that the location where the new question is asked is within a predetermined range from the questioner's destination.

[0126] This allows the system to store history information including destination-related questions, allowing further answers related to the destination to be obtained. It also avoids duplicate answers from previous questions. This allows the system to provide appropriate answers by utilizing past conversations.

[0127] According to this embodiment, the creation unit 116 creates a prompt that includes at least a part of the history information and a new question.

[0128] This allows the system to associate past questions with the locations where the past questions were asked and obtain answers to new questions, thereby enabling the system to provide appropriate answers that utilize past conversations.

[0129] According to this embodiment, the creation unit 116 includes a question selection unit 116a and a prompt creation unit 116b. The question selection unit 116a selects a past question that satisfies predetermined selection conditions from past questions. The prompt creation unit 116b creates a prompt that includes the selected past question, the location where the past question was asked, and a new question.

[0130] This allows the system to associate a new question with previous questions related to the new question and the location where the previous questions were asked, thereby obtaining an answer to the new question. This makes it possible for the system to provide an appropriate answer that utilizes past conversations.

[0131] According to this embodiment, the selection conditions include at least one of the following: a location within a predetermined distance range from the location where the new question is asked is associated with a past question; and the past question contains a keyword that is common to the new question.

[0132] This allows a past question to be selected that is asked in a location close to the new question or that contains common keywords. The selected past question and the location where the question was asked can then be associated to obtain an answer to the new question. This allows the system to provide an appropriate answer that utilizes past conversations.

[0133] According to this embodiment, the history information is information that further associates the date and time of a past question with the new question. The selection condition further includes that a date and time within a predetermined time range from the date and time of the new question is associated with the past question.

[0134] This allows a past question that was asked close to the new question to be selected. The selected past question and the location where the question was asked can then be associated to obtain an answer to the new question. This allows the system to provide an appropriate answer that utilizes past conversations.

[0135] According to this embodiment, the information processing device 100 includes a route guidance unit 120 that, when the destination of the questioner is acquired, provides guidance information for guiding the route from the current location to the destination. When the acquired answer includes information about a specific location, the route guidance unit 120 provides guidance information including the specific location and the route.

[0136] This allows the user to be provided with route guidance to the destination as well as guidance information including specific locations related to the answer to the user's question. The user can use the guidance information to easily recognize the association between the destination and the specific location.

[0137] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various other configurations can also be adopted.

[0138] In addition, in the flowcharts used in the above description, multiple steps (processes) are described in order, but the order of execution of the steps performed in each embodiment is not limited to the order described. In each embodiment, the order of the steps shown in the drawings can be changed to the extent that the content is not affected. Furthermore, the above-mentioned embodiments can be combined to the extent that the content is not contradictory. [Explanation of symbols]

[0139] S1 Information Processing System 55 Destination Reception 100 Information processing device 100 Information processing device 111 Destination Reception 112 Question acquisition part 113 Position acquisition part 114 History memory unit 115 Memory control unit 116 Creation Department 116a Question selection section 116b Prompt Creation Section 117 Answer acquisition part 118 Display section 119 Answer output section 120 Route Guidance Section

Claims

1. a generating unit that generates a prompt using at least a portion of history information that associates past questions with locations where the past questions were asked, and a new question; an answer acquisition unit that acquires an answer to the generated prompt using an interactive generation model that generates an answer to the prompt; Information processing device.

2. a location acquisition unit that acquires a current location; The device further includes a storage control unit that stores information that associates the new question with the current location at the time the new question was asked as the history information in the history storage unit. The information processing device according to claim 1 .

3. The storage control unit stores the history information in the history storage unit when a predetermined storage condition is satisfied. The information processing device according to claim 2 .

4. The storage condition includes that the interactive generative model generates an answer for the new question about a specific location. The information processing device according to claim 3 .

5. The specific location includes at least one of a point, a facility, and an area. The information processing device according to claim 4 .

6. The answer includes at least one of a suggestion and a search result.

6. The information processing device according to claim 4 or 5.

7. The storage condition includes that the location where the new question was asked is within a predetermined range from the destination of the questioner. The information processing device according to claim 3 .

8. The creation unit creates a prompt including at least a portion of the history information and the new question. The information processing device according to claim 1 .

9. The creation unit a question selection unit that selects, from the past questions, past questions that satisfy predetermined selection conditions; a prompt creation unit that creates the prompt including the selected past question, the location where the past question was asked, and the new question. The information processing device according to claim 8 .

10. The selection conditions are: a location within a predetermined distance from the location where the new question was asked is associated with the previous question; and the previous question contains a keyword common to the new question. The information processing device according to claim 9 .

11. The history information is information further associated with the date and time of the past question, The selection conditions further include that a date and time within a predetermined time range from the date and time of the new question being asked is associated with the past question. The information processing device according to claim 9 .

12. a route guidance unit that, when the destination of the requester is acquired, provides guidance information for guiding the route from the current position to the destination; When the acquired answer includes information about a specific location, the route guidance unit provides guidance information including the specific location and the route. The information processing device according to claim 1 .

13. One or more computers storing information relating a past question to a location where the past question was asked as history information in a storage unit; creating a prompt using at least a portion of the historical information and a new question; Obtaining answers to the generated prompts using an interactive generative model that generates answers to the prompts. Information processing methods.

14. On one or more computers, storing information relating a past question to a location where the past question was asked as history information in a storage unit; creating a prompt using at least a portion of the historical information and a new question; A program for causing a user to obtain an answer to a prompt using an interactive generative model that generates an answer to the prompt.

Citation Information

Patent Citations

  • Voice interaction system and voice interaction method

    JP2018054790A