Parameter acquisition system

Abstract

This parameter acquisition system acquires a parameter to set for a character that is made to perform actions in a virtual space, said parameter acquisition system comprising: a topic acquisition unit that acquires at least one topic having a distance which satisfies a prescribed condition, said distance being that from a user embedded expression representing a user with a real number vector and from a topic embedded expression representing a topic with a real number vector; an interest acquisition unit that, on the basis of correspondence information representing the correspondence between topics and interests, acquires an interest corresponding to the topic acquired by the topic acquisition unit; and a setting information output unit that outputs the interest acquired by the interest acquisition unit as interest information for setting a parameter of a character corresponding to the user.

Description

Technical Field

[0001] The present invention relates to a parameter acquisition system.

Background Art

[0002] For example, in a virtual space called the metaverse, communication between characters is carried out by allowing characters to perform activities such as wandering and interacting. Also, regarding the behavior of an avatar in a virtual space, a technique is known in which the operation of the avatar by a user is learned, and based on the learning result, the avatar is made to act autonomously to some extent (for example, see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the behavior of an avatar based solely on the learning results of user actions does not necessarily accurately reflect the user's personality. In the so-called online state, where a user controls a character in a virtual space in real time, the character naturally behaves in a way that reflects the user's personality. Characters controlled by users in this way are called, for example, player characters. In contrast, characters not controlled by users (players) are called non-player characters (NPCs). NPCs include not only offline characters that correspond to a specific user, but also characters that act autonomously within the virtual space. NPCs engage in activities such as wandering and interacting in the virtual space based on information (parameters) that constitute their personality. For example, by applying the parameters that constitute a character's personality to a dialogue model for automatically generating conversations, it is possible to generate speech from that character. Therefore, in order to sustain and activate dialogue between characters, it is necessary to set appropriate parameters for NPCs.

[0005] Therefore, the present invention has been made in view of the above problems, and aims to obtain parameters that suitably reflect the user's personality in a character that operates in a virtual space. [Means for solving the problem]

[0006] To solve the above problems, a parameter acquisition system relating to one aspect of this disclosure is a parameter acquisition system for acquiring parameters to be set for a character to operate in a virtual space, comprising: a topic acquisition unit that acquires at least one topic whose proximity to a user embedding representation, which is an embedding representation of a user represented by a real number vector, and a topic embedding representation, which is an embedding representation of a topic represented by a real number vector, meets a predetermined condition; a hobby acquisition unit that acquires hobbies corresponding to the topics acquired by the topic acquisition unit based on correspondence information representing the correspondence relationship between topics and hobbies; and a setting information output unit that outputs the hobbies acquired by the hobby acquisition unit as hobby information for setting the parameters of a character corresponding to a user.

[0007] Based on the above aspects, the distance between the user and the topic can be calculated based on user embeddings and topic embeddings that represent the characteristics of the user and the topic, respectively. Therefore, topics that meet the predetermined conditions for proximity to the user can be acquired. Then, based on the corresponding information, hobbies corresponding to the topic are acquired. Consequently, the acquired hobbies have a certain degree of closeness to the user. Since the acquired hobbies are output as hobby information, it becomes possible to apply the hobby information to the parameters set for the user's character. [Effects of the Invention]

[0008] This makes it possible to obtain parameters that appropriately reflect the user's personality in the character that operates in the virtual space. [Brief explanation of the drawing]

[0009] [Figure 1] This is a block diagram showing the functional configuration of the parameter acquisition device of this embodiment. [Figure 2] This is a hard block diagram of the parameter acquisition device and the embedded representation generation device. [Figure 3] This diagram provides a schematic explanation of the process for obtaining embedded representations. [Figure 4] This figure shows an example of a topic obtained based on the distance between the user-embedded representation and the topic-embedded representation. [Figure 5] This figure shows an example of a given list of hobbies, including hobby-related terms. [Figure 6] This figure shows an example of correspondence information that defines the relationship between hobbies and topics. [Figure 7] As an example of correspondence information, this figure shows an example of a thesaurus that is hierarchically structured, including topical terms and hobby-related terms. [Figure 8] As an example of correspondence information, the figure shows an example of calculating the similarity between topic words and hobby-related words. [Figure 9] This diagram schematically illustrates the output of hobby information for setting as character parameters. [Figure 10] It is a diagram showing an example of an attribute list associating interests with attribute information. [Figure 11] It is a diagram schematically explaining the output of interest information and attribute information to be set as parameters of a character. [Figure 12] It is a flowchart showing the processing contents of a parameter acquisition method in a parameter acquisition device. [Figure 13] It is a diagram showing the configuration of a parameter acquisition program. [Figure 14] It is a block diagram showing the functional configuration of the embedded expression generation device of the present embodiment. [Figure 15] It is a diagram schematically explaining the process of acquiring a speech text. [Figure 16] It is a diagram showing an example of the configuration of a language model and an example of the machine learning process of the language model. [Figure 17] It is a diagram showing an example of an embedded expression acquisition process using an embedded part of a learned language model. [Figure 18] It is a diagram showing an example of acquiring an edge for generating a relationship graph. [Figure 19] It is a diagram showing an example of a relationship graph and an example of extracting positive and negative examples from the relationship graph. [Figure 20] It is a diagram showing an example of the embedded expression of each entity obtained by learning a graph neural network constituting a relationship graph. [Figure 21] It is a flowchart showing the processing contents of an embedded expression generation method in an embedded expression generation device. [Figure 22] It is a flowchart showing the processing contents of the machine learning of a language model. [Figure 23] It is a diagram showing the configuration of an embedded expression generation program.

Embodiments for Carrying Out the Invention

[0010] Embodiments of the parameter acquisition system according to the present invention will be described with reference to the drawings. Where possible, the same parts are denoted by the same reference numerals, and redundant descriptions are omitted.

[0011] Figure 1 shows the functional configuration of the parameter acquisition system according to this embodiment. The parameter acquisition system 1 of this embodiment is a system for acquiring parameters to be set for a character that operates in a virtual space, and is composed of a parameter acquisition device 30 as an example. The parameter acquisition system 1 may further include an embedded representation generation device 10.

[0012] The parameter acquisition device 30 is a device that acquires parameters to be set for a character that operates in a virtual space, and as shown in Figure 1, it functionally comprises an embedded expression input unit 31, a topic acquisition unit 32, a hobby acquisition unit 33, an attribute acquisition unit 34, and a setting information output unit 35. Each of these functional units 31 to 35 may be configured in a single device as illustrated in Figure 1, or they may be distributed across multiple devices.

[0013] The embedded expression generation device 10 is a device that generates embedded expressions of at least the user and the topic. In the example shown in Figure 1, the embedded expression generation device 10 is shown as a separate device from the parameter acquisition device 30, but it may be configured as an integral part of the parameter acquisition device 30. The functions of the embedded expression generation device 10 will be described later.

[0014] The block diagram shown in Figure 1 represents functional units. These functional blocks (components) are implemented by any combination of at least one of hardware and software. Furthermore, the method of implementing each functional block is not particularly limited. That is, each functional block may be implemented using one device that is physically or logically coupled, or it may be implemented using two or more physically or logically separated devices that are directly or indirectly connected (for example, using wired or wireless connections). A functional block may also be implemented by combining the above one device or the above multiple devices with software.

[0015] Functions include, but are not limited to, judgment, decision, judgment, calculation, calculation, processing, derivation, investigation, exploration, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, assumption, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating (mapping), and assigning. For example, a functional block (configuration part) that enables transmission is called a transmitting unit or transmitter. As mentioned above, the method of implementation is not particularly limited.

[0016] For example, the parameter acquisition device 30 in one embodiment of the present invention may function as a computer. Similarly, the embedded representation generation device 10 may function as a computer. Figure 2 shows an example of the hardware configuration of the parameter acquisition device 30 according to this embodiment. Similarly, the hardware configuration of the embedded representation generation device 10 is also shown in Figure 2. Physically, the parameter acquisition device 30 and the embedded representation generation device 10 may be configured as computer devices including a processor 1001, memory 1002, storage 1003, communication device 1004, input device 1005, output device 1006, bus 1007, etc.

[0017] In the following explanation, the term "device" can be replaced with "circuit," "device," "unit," etc. The hardware configuration of the parameter acquisition device 30 and the embedded representation generation device 10 may include one or more of the devices shown in the figure, or it may be configured to omit some of the devices.

[0018] Each function in the parameter acquisition device 30 and the embedded representation generation device 10 is realized by loading predetermined software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations and control communication by the communication device 1004 and the reading and / or writing of data in the memory 1002 and storage 1003.

[0019] The processor 1001 controls the entire computer, for example, by running the operating system. The processor 1001 may consist of a central processing unit (CPU) that includes interfaces with peripheral devices, control devices, arithmetic units, registers, etc. For example, the various functional units 31-35 shown in Figure 1 and the various functional units of the embedded representation generation device 10 may be implemented by the processor 1001.

[0020] Furthermore, the processor 1001 reads programs (program code), software modules, and data from the storage 1003 and / or communication device 1004 into the memory 1002 and executes various processes accordingly. The program used is one that causes the computer to execute at least a part of the operations described in the above embodiment. For example, the functional units 31-35 of the parameter acquisition device 30 and the functional units of the embedded representation generation device 10 may be stored in the memory 1002 and implemented by a control program that runs on the processor 1001. Although the above processes have been described as being executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented on one or more chips. The program may also be transmitted from a network via a telecommunications line.

[0021] Memory 1002 is a computer-readable recording medium and may consist of at least one of the following: ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), RAM (Random Access Memory), etc. Memory 1002 may also be called a register, cache, main memory, etc. Memory 1002 can store executable programs (program code), software modules, etc., for carrying out the parameter generation method according to one embodiment of the present invention.

[0022] The storage 1003 is a computer-readable recording medium and may consist of at least one of the following: an optical disc such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disc, a digital multipurpose disc, a Blu-ray® disc), a smart card, flash memory (e.g., a card, a stick, a key drive), a floppy® disk, a magnetic strip, etc. The storage 1003 may also be called an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, server, or other suitable medium including memory 1002 and / or storage 1003.

[0023] The communication device 1004 is hardware (transceiver / receiver device) for communicating between computers via a wired and / or wireless network, and is also referred to as a network device, network controller, network card, communication module, etc.

[0024] The input device 1005 is an input device that accepts input from an external source (e.g., a keyboard, mouse, microphone, switch, button, sensor, etc.). The output device 1006 is an output device that outputs to an external source (e.g., a display, speaker, LED lamp, etc.). The input device 1005 and the output device 1006 may be configured as an integrated unit (e.g., a touch panel).

[0025] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may consist of a single bus or different buses may be used for communication between devices.

[0026] Furthermore, the parameter acquisition device 30 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array), and some or all of each functional block may be realized by such hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.

[0027] Next, the various functions of the parameter acquisition device 30 will be described. The embedded expression input unit 31 acquires user embedded expressions, which are the embedded expressions of the user, and topic embedded expressions, which are the embedded expressions of the topic. User embedded expressions are represented by real number vectors and are embedded expressions that reflect the characteristics of the user. Topic embedded expressions are represented by real number vectors and are embedded expressions that reflect the characteristics of the topic. Furthermore, since the user embedded expressions and topic embedded expressions acquired by the embedded expression input unit 31 reflect the relationship between them using a predetermined method, it is possible to calculate the distance between the user and the topic.

[0028] Figure 3 is a schematic diagram illustrating the process of acquiring embedded representations. As shown in Figure 3, the embedded representation input unit 31 may acquire user embedded representations vu and topic embedded representations vt from the embedded representation generation device 10. As will be described later, the embedded representation generation device 10 generates user embedded representations vu and topic embedded representations vt that represent the characteristics of the user and topic, respectively, and appropriately reflect the relationship between the user and the topic.

[0029] The topic acquisition unit 32 acquires at least one topic in which the proximity of the user's embedded representation and the topic's embedded representation meets a predetermined condition. Specifically, the topic acquisition unit 32 may calculate the distance between the user's embedded representation and each of the topic's embedded representations of multiple topics, and acquire a predetermined number of topics with the highest calculated proximity.

[0030] Figure 4 shows an example of a topic obtained based on the distance between a user embedding expression and a topic embedding expression. In the example shown in Figure 4, the topic acquisition unit 32 calculates the distance between user A's user embedding expression vu and each of the topic embedding expressions vt of multiple topics obtained by the embedding expression input unit 31, and acquires the topics t11 (World Cup), t12 (Soccer), t13 (Japan National Team), and t14 (Goals) of topic embedding expressions vt that are closest in distance to the user. This makes it possible to extract topics that are likely to have a close relationship with the user.

[0031] Furthermore, the topic acquisition unit 32 may acquire topics where the distance between the user's embedded representation vu and the topic's embedded representation vt is less than or equal to a predetermined value. Specifically, the topic acquisition unit 32 calculates the distance between user A's embedded representation vu and each of the topic's embedded representation vt acquired by the embedded representation input unit 31, and acquires topics whose calculated distance is less than or equal to a given threshold. This ensures that topics with a suitable proximity to the user are acquired.

[0032] Referring again to Figure 1, the hobby acquisition unit 33 acquires hobbies corresponding to the topic acquired by the topic acquisition unit 32, based on correspondence information that represents the correspondence between topics and hobbies. Specifically, the hobby acquisition unit 33 acquires hobbies corresponding to the topic acquired by the topic acquisition unit 32, based on a given list of hobbies that includes multiple hobby words representing hobbies, by referring to a thesaurus as correspondence information.

[0033] Figure 5 shows an example of a given hobby list that includes hobby terms. The hobby list may be pre-configured and stored in a predetermined storage means (e.g., storage 1003). As shown in Figure 5, the hobby list hl includes hobby terms that represent hobbies, such as "shopping," "music," "cooking," and "games," as a list.

[0034] A thesaurus is generally a dictionary of information that classifies and organizes words based on relationships such as hierarchical / subordinate relationships, part / whole relationships, synonymous relationships, and similar relationships. In this embodiment, the hobby acquisition unit 33 refers to a thesaurus that defines the relationships between multiple words, including at least hobby-related words and topic-related words, as corresponding information. The thesaurus may be pre-configured and stored in a predetermined storage means (for example, storage 1003).

[0035] The hobby acquisition unit 33 may refer to a thesaurus to extract hobby terms related to the topic term and generate a map as correspondence information showing the correspondence between the extracted topic term and hobby terms. Figure 6 is a diagram showing an example of a map that defines the correspondence between hobbies and topics. As shown in Figure 6, the map associates, for example, the topic term "World Cup" with hobby terms "sports," "soccer," and "World Cup."

[0036] The hobby acquisition unit 33 may refer to the map illustrated in Figure 6, extract hobby terms associated with topic terms representing topics acquired by the topic acquisition unit 32, and acquire hobbies represented by hobby terms.

[0037] Figure 7 shows an example of a thesaurus hierarchically structured, including topic words and hobby words, as an example of correspondence information. The thesaurus ts is information that hierarchically defines the relationships between hobby words h1, h21-h23 and topic word t2. The thesaurus ts may also be composed of topic words t21, t22, and t23 that indicate topics uttered by the user in the virtual space.

[0038] If the topic acquisition unit 32 acquires the topic "World Cup", the hobby acquisition unit 33 refers to the thesaurus ts and acquires the hobby word h21 "soccer" which is associated with the topic word t23 "World Cup" at a higher level.

[0039] When obtaining hobby terms by referring to a thesaurus with a hierarchical structure, the hobby acquisition unit 33 may further acquire hobby terms associated with the acquired hobby term at a higher or lower level. That is, in the example shown in Figure 7, the hobby acquisition unit 33 may further acquire hobby term h1 "sports" which is associated with the acquired hobby term h21 "soccer".

[0040] In this way, by referring to correspondence information composed of a thesaurus that defines the relationships between words representing hobbies and topics, corresponding hobby terms are extracted based on topic terms that represent topics closely related to the user. Therefore, hobbies represented by hobby terms can be output as hobby information closely related to the user.

[0041] The hobby acquisition unit 33 may acquire hobbies using the similarity between topic words and hobby words as correspondence information. Specifically, the hobby acquisition unit 33 refers to a given hobby list hl containing multiple hobby words that represent hobbies, and calculates the similarity between each topic word representing a topic acquired by the topic acquisition unit 32 and the hobby words included in the hobby list hl as correspondence information.

[0042] Figure 8 shows an example of calculating the similarity between topic words and hobby words as an example of correspondence information. As shown in Figure 8, the hobby acquisition unit 33 calculates the similarity sim between each of the topic words t31~t36,... that represent topics acquired by the topic acquisition unit 32 and the hobby words h31, h32,... included in the hobby list hl. The method for calculating the similarity between words is not limited, but the hobby acquisition unit 33 may calculate the similarity between topic words and hobby words using Word2Vec. According to Word2Vec, the similarity between hobby words and topic words can be calculated with high accuracy.

[0043] The hobby acquisition unit 33 then acquires hobbies corresponding to hobby words whose calculated similarity is equal to or greater than a given threshold. For example, if a given threshold for similarity is "0.7", the hobby acquisition unit 33 extracts the hobby word h31 "sports", which has a similarity of 0.8 to the topic word t32 "soccer", and acquires the hobby "sports" represented by the extracted hobby word "sports".

[0044] In this way, hobbies are obtained that are represented by hobby terms that have a high degree of similarity to topic terms that represent topics closely related to the user. Therefore, hobbies that are closely related to the user can be output as hobby information.

[0045] Figure 9 schematically illustrates the output of hobby information for setting as character parameters. As explained with reference to Figures 4 to 8, the topic acquisition unit 32 acquires topics t11 to t14 that have a close relationship with user A (ua) based on the distance between the user embedding representation vu and the topic embedding representation vt. The hobby acquisition unit 33 acquires hobbies H1 and H2 that correspond to topics t11 to t14 acquired by the topic acquisition unit 32, based on the correspondence information CI that represents the correspondence relationship between topics and hobbies.

[0046] The setting information output unit 35 then outputs the hobbies H1 and H2 acquired by the hobby acquisition unit 33 as hobby information HI for setting the parameters of the character corresponding to user A in the virtual space. The output method is not limited, and the setting information output unit 35 may set the parameters related to the hobbies of user A's character based on the hobby information HI. The setting information output unit 35 may also store the hobby information HI in a predetermined storage means.

[0047] Referring again to Figure 1, the parameter acquisition device 30 may further include an attribute acquisition unit 34. The attribute acquisition unit 34 acquires attribute information associated with hobbies acquired by the hobby acquisition unit 33 by referring to a given attribute list. The attribute list is, for example, information that has been pre-associated with hobbies and people's attribute information.

[0048] Figure 10 shows an example of an attribute list. The attribute list may be pre-configured and stored in a predetermined storage means (e.g., storage 1003). As shown in Figure 10, the attribute list stores attributes such as the user's age, gender, and occupation in association with hobbies. That is, the attribute list stores various attributes that a user with a particular hobby is likely to possess, in association with that hobby. For example, the attribute list stores attributes such as "20s," "male," and "university student" in association with the hobby "sports."

[0049] When the hobby acquisition unit 33 acquires the hobby "sports" as user A's hobby information, the attribute acquisition unit 34 acquires the attributes "20s", "male", and "university student" from the attribute list that are associated with the hobby "sports".

[0050] Figure 11 schematically illustrates the output of hobby information and attribute information for setting as character parameters. As shown in Figure 11, the topic acquisition unit 32 acquires topics t11 to t14 that have a close relationship with user A (ua) based on the distance between the user embedding representation vu and the topic embedding representation vt. The hobby acquisition unit 33 acquires hobbies H1 and H2 that correspond to topics t11 to t14 acquired by the topic acquisition unit 32, based on the correspondence information CI that represents the correspondence relationship between topics and hobbies.

[0051] Furthermore, the attribute acquisition unit 34 refers to the attribute list AL, which associates hobbies with attribute information, and acquires attribute information A1 to A3 associated with, for example, hobby information H1 "sports" from the hobby information acquired by the hobby acquisition unit 33.

[0052] The setting information output unit 35 then outputs the hobbies H1 and H2 acquired by the hobby acquisition unit 33 as hobby information HI for setting the parameter PM of the character corresponding to user A in the virtual space. Furthermore, the setting information output unit 35 outputs the attribute information AI, which includes attributes A1 to A3 acquired by the attribute acquisition unit 34, as information for setting the parameter PM of the character corresponding to user A in the virtual space.

[0053] The output configuration is not limited, and the setting information output unit 35 may set parameters related to the hobbies and attributes of user A's character based on the hobby information HI and attribute information AI. The setting information output unit 35 may also store the hobby information HI and attribute information AI in a predetermined storage means.

[0054] The topic acquisition unit 32 may acquire topics that are close in distance to all users active in the virtual space. Specifically, the topic acquisition unit 32 may acquire topics that are commonly extracted by a predetermined number of users or more. The topic acquisition unit 32 may also extract a predetermined number of topics that are closest in distance to all users, or to a predetermined percentage or a predetermined number of users among all users. The topic acquisition unit 32 outputs the topics extracted for all users as hobby information to be set as parameters related to the hobbies of characters active in the virtual space. This hobby information output based on all users may, for example, be set as parameters for an NPC that does not correspond to a specific user. Furthermore, the attribute acquisition unit 34 may acquire attribute information by referring to an attribute list based on the hobby information output based on all users, and output it as information to be set as parameters for an NPC that does not correspond to a specific user.

[0055] Figure 12 is a flowchart showing the processing details of the parameter acquisition method in the parameter acquisition device 30.

[0056] In step S31, the embedded expression input unit 31 acquires a user embedded expression, which is a user's embedded expression, and a topic embedded expression, which is a topic's embedded expression.

[0057] In step S32, the topic acquisition unit 32 acquires at least one topic in which the proximity of the user's embedded representation and the topic's embedded representation meets a predetermined condition.

[0058] In step S33, the hobby acquisition unit 33 acquires hobbies corresponding to the topic acquired by the topic acquisition unit 32, based on correspondence information that represents the correspondence between the topic and the hobby.

[0059] In step S34, the setting information output unit 35 outputs the hobbies acquired by the hobby acquisition unit 33 as hobby information for setting as parameters of the character corresponding to the user in the virtual space.

[0060] In step S35, the attribute acquisition unit 34 refers to a given attribute list and acquires attribute information associated with the hobby acquired by the hobby acquisition unit 33.

[0061] In step S36, the setting information output unit 35 outputs the attribute information acquired by the attribute acquisition unit 34 as information for setting the parameters of the character corresponding to the user in the virtual space.

[0062] Next, with reference to Figure 13, a parameter acquisition program for causing the computer to function as the parameter acquisition device 30 of this embodiment will be described. Figure 13 is a diagram showing the configuration of the parameter acquisition program. The parameter acquisition program P3 is composed of a main module m30 that comprehensively controls the parameter acquisition process in the parameter acquisition device 30, an embedded expression input module m31, a topic acquisition module m32, a hobby acquisition module m33, an attribute acquisition module m34, and a setting information output module m35. Each of the modules m31 to m35 realizes the respective functions for each of the functional units 31 to 35.

[0063] The parameter acquisition program P3 may be transmitted via a transmission medium such as a communication line, or it may be stored in a recording medium M3, as shown in Figure 13.

[0064] According to the parameter acquisition device 30, parameter acquisition method, and parameter acquisition program P3 of this embodiment described above, the distance between the user and the topic can be calculated based on user embedding representations and topic embedding representations, which represent the characteristics of the user and the topic, respectively. Therefore, topics that meet predetermined conditions for proximity to the user can be acquired. Then, based on the corresponding information, hobbies corresponding to the topic are acquired. Consequently, the acquired hobbies have a certain degree of proximity to the user. Since the acquired hobbies are output as hobby information, it becomes possible to apply the hobby information to the parameters set for the user's character.

[0065] (Embedded representation generation device) Next, the embedded representation generation device 10 shown in Figure 1 will be described. The embedded representation generation device 10 can obtain embedded representations of entities in which the relationships between different entities are appropriately represented. Figure 14 is a diagram showing the functional configuration of the embedded representation generation device 10 according to this embodiment. The embedded representation generation device 10 of this embodiment is a device that generates embedded representations of at least users and topics.

[0066] As shown in Figure 14, the embedded expression generation device 10 functionally comprises a speech log acquisition unit 11, a speech recognition unit 12, a text acquisition unit 13, an emotion acquisition unit 14, a language understanding unit 15, a topic extraction unit 16, an embedded expression acquisition unit 17, a relationship extraction unit 18, a relationship learning unit 19, an embedded expression output unit 20, and a link prediction unit 21. Each of these functional units 11 to 21 may be configured in a single device as illustrated in Figure 14, or they may be distributed across multiple devices.

[0067] The block diagram shown in Figure 14 represents functional units. These functional blocks (components) are realized by any combination of at least one of hardware and software. Furthermore, the method of realizing each functional block is not particularly limited. That is, each functional block may be realized using one device that is physically or logically coupled, or it may be realized using two or more physically or logically separated devices that are directly or indirectly connected (for example, using wired or wireless connections). A functional block may also be realized by combining the above one device or the above multiple devices with software.

[0068] Next, the various functions of the embedded expression generation device 10 will be described. The speech log acquisition unit 11 acquires a speech log that represents the content of the user's speech. The speech recognition unit 12 converts the speech log into text if it is in the form of speech. The text acquisition unit 13 acquires speech text, which is text that represents the content of the user's speech, based on the speech log. The emotion acquisition unit 14 acquires emotion information that represents the user's emotions at the time the user's speech was uttered, based on the speech audio or the user's facial expression, and associates the acquired emotion information with the speech text that represents the content of the speech.

[0069] Referring to Figure 15, the processing details of the speech log acquisition unit 11, speech recognition unit 12, text acquisition unit 13, and emotion acquisition unit 14 will be explained in detail. Figure 15 is a diagram that schematically illustrates the process of acquiring speech text.

[0070] The speech log acquisition unit 11 may acquire a speech log representing the content of the user's speech in text form based on input via an input device 41, such as a keyboard or touch panel. Alternatively, the speech log acquisition unit 11 may acquire a speech log representing the content of the user's speech in audio data form based on voice input via a microphone 42, for example.

[0071] The speech log acquired by the speech log acquisition unit 11 may be audio or text (chat) representing the content of the user's speech in a predetermined virtual space. The predetermined virtual space may, for example, be a virtual space known as a metaverse. The user's speech may be speech made by an avatar in a virtual space such as a metaverse, and the speech log acquisition unit 11 may acquire the speech log representing the avatar's speech in the form of audio or text.

[0072] The speech recognition unit 12 converts speech into text when the speech log acquisition unit 11 acquires a speech log of speech characteristics. The speech recognition unit 12 may convert the speech log consisting of speech into text by any method, for example, by using well-known speech recognition technology.

[0073] The text acquisition unit 13 acquires speech text, which is text representing the content of the user's utterance, based on the speech log. If the speech log is acquired in text form by the speech log acquisition unit 11, the text acquisition unit 13 acquires the text representing the speech log as speech text. If the speech log is acquired in voice form by the speech log acquisition unit 11, the text acquisition unit 13 acquires the speech log converted into text by the speech recognition unit 12 as speech text. The text acquisition unit 13 then sends the acquired speech text t1 to the language understanding unit 15.

[0074] The emotion acquisition unit 14 acquires emotion information representing the user's emotions at the time the user utters a speech, based, for example, on the user's speech voice acquired via the microphone 42, or on an image representing the user's facial expression acquired via the camera 43.

[0075] The emotion acquisition unit 14 may acquire the user's emotional information from the spoken voice by any method, for example, by using well-known emotion recognition technology. Furthermore, the emotion acquisition unit 14 may acquire the user's emotional information from an image representing the user's facial expression by any method, for example, by using well-known facial expression recognition technology.

[0076] Furthermore, the source of emotional information is not limited to the user's facial expressions and spoken voice; the emotional information acquisition unit 14 may also acquire it from the state of the user's avatar when they speak in the virtual space.

[0077] Emotional information includes categories such as "joy," "anger," "sadness," and "surprise," and certain specific emotional categories such as "happy" and "calm" can be classified as positive emotions.

[0078] The emotion acquisition unit 14 associates emotion information acquired from the user's facial expressions and voice during speech with the speech text t1 that represents the content of the speech. Therefore, the language understanding unit 15 can acquire the speech text t1 to which the emotion information is associated.

[0079] The language understanding unit 15 performs machine learning on a language model composed of an encoder-decoder model. Figure 16 shows an example of the configuration of the language model and the machine learning processing of the language model. The language model md is an encoder-decoder model composed of a neural network, and includes an embedding unit en (encoder) and a decoding unit de (decoder).

[0080] The configuration of the language model md is not limited, but it may be an encoder-decoder model consisting of pairs of recurrent neural networks such as seq2seq, or it may be composed of transformers such as T5 (Text-to-Text Transfer Transformer).

[0081] The embedding unit en encodes the input text and outputs an embedding representation that represents the characteristics of the text. The decoding unit de decodes the embedding representation, which includes at least the output from the embedding unit en, and outputs the decoded text dt. In the description of the input and output of the language model, "text" refers to vector data obtained by converting text using a predetermined method, or to vector data that represents text and is output.

[0082] The language understanding unit 15 inputs a first user utterance text, which represents the content of one user's utterance, from among the utterance texts representing the content of the user's utterance, into the embedding unit en, thereby obtaining the user utterance embedding representation output from the embedding unit en.

[0083] In the example shown in Figure 16, the language understanding unit 15 inputs the first user utterance text ut1 ("Tonight's dinner is...") from the utterance text ut ("Tonight's dinner is...") which represents the content of user A's utterance and is used as training data for the language model md, into the embedding unit en. The language understanding unit 15 then obtains the user utterance embedding representation ebs, which has been encoded and output by the embedding unit en.

[0084] Here, the language understanding unit 15 acquires user embeddings, which are the user's embedded representations. For example, the embedding representation generation device 10 may further include a user embedding representation management unit 22. The user embedding representation management unit 22 may generate and manage initial user embeddings before learning. The user embedding representation management unit 22 may also manage user embeddings during the learning process. The user embedding representation management unit 22 may be configured as a functional unit of the embedding representation generation device 10 shown in Figure 14, or it may be configured as a separate device.

[0085] The user embedding representation is represented by a real-valued vector. The initial user embedding representation may be a random real-valued vector, or it may be a real-valued vector consisting of feature quantities that reflect some characteristics of the user. In the embedding representation generation device 10 of this embodiment, the method for obtaining the initial user embedding representation is not limited and may be any well-known method.

[0086] The language understanding unit 15 generates a composite embedding expression by combining a user utterance embedding expression and a user embedding expression which is the embedding expression of that particular user. The language understanding unit 15 may also generate a composite embedding expression by concatenating the user utterance embedding expression and the user embedding expression. In the example shown in Figure 16, the language understanding unit 15 obtains the user embedding expression ebu of user A from the user embedding expression management unit 22, and concatenates the user utterance embedding expression ebs, which is the embedding expression of the first user utterance text ut1, with the user embedding expression ebu of user A to generate a composite embedding expression ebl. Then, the language understanding unit 15 inputs the composite embedding expression ebl to the decoding unit de, and obtains the decoded text dt which is decoded by the decoding unit de.

[0087] The language understanding unit 15 performs machine learning to adjust the language model and user embedding representation so that the error between the second user utterance text following the first user utterance text and the decoded text in the utterance text is reduced. In the example shown in Figure 16, the language understanding unit 15 adjusts the language model md and user embedding representation ebu so that the error between the second user utterance text ut2 (curry) following the first user utterance text ut1 in the utterance text ut ("Tonight's dinner is" "curry") and the decoded text dt is reduced.

[0088] Furthermore, the language understanding unit 15 may perform machine learning to adjust the language model md and user embedding representations using speech text associated with emotion information representing predetermined positive emotions. As mentioned above, speech text ut can be accompanied by emotion information representing the user's emotions at the time the speech text was uttered. In such cases, the language understanding unit 15 may perform machine learning to adjust the language model md and user embedding representations using speech text ut associated with emotion information representing positive emotions such as "happy" or "calm" as training data.

[0089] In this way, by using speech texts associated with emotional information representing positive emotions in machine learning, it is possible to use combinations of first and second user speech texts that are likely to be expressed when the user is experiencing positive emotions as training data. By performing machine learning using such training data, it is possible to obtain an embedding unit and a user embedding expression that can generate topic embedding expressions that reflect a favorable relationship between the user and topic words, etc.

[0090] The language model md, which includes a pre-trained neural network, can be understood as a program that is loaded or referenced by a computer, causing the computer to perform predetermined processes and realize predetermined functions.

[0091] In other words, the trained language model md of this embodiment is used in a computer equipped with a CPU and memory. Specifically, the computer's CPU operates in accordance with instructions from the trained language model md stored in memory, performing calculations on the input data input to the input layer of the neural network, for example, based on trained weight coefficients (parameters) and response functions corresponding to each layer, and outputting the result (probability) from the output layer.

[0092] Referring again to Figure 14, the topic extraction unit 16 extracts topic words from the utterance text, which are words or phrases that represent the topic in the user's utterance. The method applied to the extraction of topic words is not limited, and the topic extraction unit 16 can extract topic words by using, for example, well-known methods such as morphological analysis and text mining.

[0093] The embedding expression acquisition unit 17 inputs a topic word into a trained embedding unit and acquires a topic embedding expression output from the embedding unit. Figure 17 shows an example of the embedding expression acquisition process using the embedding unit of a trained language model. As shown in Figure 17, the embedding expression acquisition unit 17 acquires a topic embedding expression ebt by inputting the topic word tp extracted by the topic extraction unit 16 into the trained embedding unit en. The trained embedding unit en can output a suitable topic embedding expression that appropriately reflects the characteristics of the topic in response to the input topic word.

[0094] Furthermore, the embedding representation acquisition unit 17 may further acquire location embedding representations output from the embedding unit en by inputting location text representing a location into the learned embedding unit en. The location text may be, for example, the name of the location and a descriptive text explaining the location. This allows for obtaining location embedding representations that suitably reflect the characteristics of the location.

[0095] The relationship extraction unit 18 generates a relationship graph with at least the user and topic as nodes, based on the user's speech history (speech log) and action history. The relationship extraction unit 18 may also generate a relationship graph that further includes location as a node.

[0096] The relationship extraction unit 18 extracts relationships between nodes based on the user's utterances and actions, and draws edges based on the extracted relationships. In this embodiment, the relationship extraction unit 18 generates a relationship graph based on the user's utterance history and action history in a predetermined virtual space.

[0097] Figure 18 shows an example of edge acquisition for generating a relation graph. As shown in Figure 18, the relation extraction unit 18 acquires the user's utterance history hs (utterance log and utterance text, etc.) in a virtual space such as the metaverse. The relation extraction unit 18 extracts the actual user interaction r1 from the user's utterance history hs and assigns it as the edge ed1 between the user's nodes in the relation graph.

[0098] Furthermore, the relationship extraction unit 18 extracts the user's utterance record r2 of a topic word from the user's utterance history hs and assigns it as an edge ed2 that connects the user's node and the topic word's node.

[0099] Furthermore, the relationship extraction unit 18 acquires the user's behavior history ha in the virtual space. The relationship extraction unit 18 then extracts the user's visit record r3 to a location from the user's behavior history ha and assigns it as an edge ed3 that connects the user's node and the node of that location.

[0100] The relation learning unit 19 obtains the learned embedding representation for each node by training a graph neural network that uses the learned user embedding representation and topic embedding representation, respectively, as features of the user and topic nodes in the relation graph.

[0101] Alternatively, the relation learning unit 19 may obtain learned embedding representations for each node in a relation graph that further includes location nodes, by training a graph neural network of the relation graph using location embedding representations as features of the location nodes.

[0102] Specifically, the relation learning unit 19 associates the trained user embedding representation ebu, obtained by machine learning by the language understanding unit 15, and the topic embedding representation ebt, obtained by the embedding representation acquisition unit 17, with each user and topic node in the relation graph as features. In addition, the relation learning unit 19 associates the location embedding representation obtained by the embedding representation acquisition unit 17 with the location node in the relation graph as a feature.

[0103] Then, the relational learning unit 19 learns a graph neural network of relational graphs, using the embedding representations as features for each node, thereby modifying the features and weights of each node and obtaining the learned embedding representations for each node.

[0104] The relation learning unit 19 can learn relation graphs using well-known graph neural network learning methods. The learning of relation graphs will be briefly explained with reference to Figure 19. Figure 19 shows an example of a relation graph and an example of extracting positive and negative examples from the relation graph.

[0105] The relation graph gn illustrated in Figure 19 includes nodes n1 to n5, each corresponding to either a user, topic, or location. The relation learning unit 19 randomly samples a node of interest. In the example shown in Figure 19, node n2 is sampled as the node of interest.

[0106] The relation learning unit 19 extracts a positive example graph g1 and a negative example graph g2 from the relation graph gn. The positive example graph g1 includes node n2, which is the node of interest, and nodes n1 and n5, which are connected to node n2 by edges. The negative example graph g2 includes node n2, which is the node of interest, and nodes n3 and n4, which are not connected to node n2 by edges. Note that the negative example graph g2 does not need to include all nodes that are not connected to the node of interest by edges.

[0107] The following describes an example of learning a relational graph (gn). However, since the learning process for graph neural networks is a well-known technique, the explanation will be brief.

[0108] First, let's explain the learning process using the positive example graph g1. Based on the positive example graph g1, the relational learning unit 19 extracts an adjacency matrix A in which the nodes included in the graph are represented as rows and columns, and the connection relationships with the node of interest, node n2, via edges are represented as elements.

[0109] Furthermore, the relational learning unit 19 extracts a diagonal matrix I in which the nodes included in the graph are used as rows and columns, and the self-loops of the nodes are used as elements. Then, if the real vector representing the features of a node is denoted as node feature quantity X, the features of each node are expressed by the following equation as the sum (convolution) of the features of related nodes represented by the adjacency matrix A and the features of the node itself represented by the diagonal matrix I. (A+I)·X

[0110] The relational learning unit 19 multiplies the features of each convolved node by a weight W, as shown by the following equation, and then inputs this into the activation function f to obtain the output H. H(positive example) = f((A+I)·X·W) Then, the relational learning unit 19 learns weights and features such that the output H (positive example) obtained based on the positive example graph g1 becomes 1.

[0111] The relation learning unit 19 similarly obtains the output H(negative examples) based on the negative example graph g2. Then, the relation learning unit 19 learns the weights and features so that the output H(negative examples) obtained based on the negative example graph g2 becomes 0.

[0112] Referring again to Figure 14, the embedding representation output unit 20 outputs the embedding representation of each node that has undergone learning by the relation learning unit 19. Figure 20 is a diagram showing an example of the embedding representation of each entity obtained by learning the graph neural network that constitutes the relation graph. As shown in Figure 20, the embedding representation output unit 20 outputs the embedding representation EB of entities 1, 2, 3, 4, 5, ... corresponding to each node of the relation graph gn, based on the learning gm of the graph neural network targeting the relation graph gn by the relation learning unit 19.

[0113] The resulting embedding representation of each node is a real vector that appropriately reflects the characteristics of each entity corresponding to each node, as well as the relationships between entities, making it possible to calculate the distance between entities. Therefore, each node in the relational graph corresponds to a different type of entity, such as a user, topic, or location, and it becomes possible to calculate the distance between different types of entities.

[0114] The manner in which the embedded expression is output by the embedded expression output unit 20 is not limited and may include storage by a predetermined storage means, transmission to a predetermined device, display on a predetermined display device, etc.

[0115] Referring again to Figure 14, the link prediction unit 21 calculates the distance between nodes based on the learned embedding representation of each node, and calculates link prediction information indicating the likelihood of edges being formed between each node based on the calculated distance between nodes.

[0116] Specifically, the link prediction unit 21 determines, for example, whether the distance between nodes, calculated as the distance between real vectors, is less than or equal to a given threshold. If the link prediction unit 21 determines that the distance between nodes is less than or equal to the threshold, it outputs link prediction information indicating that it predicts the existence of an edge between those nodes.

[0117] Thus, by training the graph neural network gm on the relational graph gn, an embedding representation expressed as a real vector, which can calculate the distance between entities of different types, is obtained. Therefore, link prediction information is calculated that allows for the evaluation of the likelihood of edges being formed between each node in the graph. Consequently, it becomes possible to predict whether there is a relationship of a certain degree or higher between the entities corresponding to each node.

[0118] Furthermore, the link prediction unit 21 outputs link prediction information that indicates each node whose distance from the node is less than or equal to the given threshold, based on a given threshold for the distance between nodes.

[0119] Specifically, the link prediction unit 21 determines, for example, whether the distance between nodes calculated as the distance between real number vectors is less than or equal to a given threshold, and outputs information indicating the entity corresponding to the node whose distance is determined to be less than or equal to the threshold as link prediction information. If at least one of the entities corresponding to the node whose distance is determined to be less than or equal to the threshold is a user, the user may be provided with information indicating the other entity as recommendation information.

[0120] Figure 21 is a flowchart showing the processing details of the embedded representation generation method in the embedded representation generation device 10.

[0121] In step S1, the text acquisition unit 13 acquires speech text, which is text representing the content of the user's utterance, based on the speech log.

[0122] In step S2, the language understanding unit 15 performs machine learning on a language model composed of an encoder-decoder model. The processing details of step S2 will be explained with reference to Figure 22.

[0123] Figure 22 is a flowchart showing the processing steps of the language model's machine learning. In step S21, the language understanding unit 15 inputs a first user utterance text, representing the utterance of one user from among the spoken texts, into the embedding unit en.

[0124] In step S22, the language understanding unit 15 obtains the user utterance embedding representation ebs encoded and output by the embedding unit en.

[0125] In step S23, the language understanding unit 15 generates a composite embedding expression "ebl" by combining the user utterance embedding expression and the user embedding expression, which is the embedding expression of that user. The language understanding unit 15 then inputs the composite embedding expression ebl to the decoding unit de.

[0126] In step S24, the language understanding unit 15 obtains the decoded text dt, which has been decoded by the decoding unit de.

[0127] In step S25, the language understanding unit 15 performs machine learning to adjust the language model and user embedding representation so that the error between the second user utterance text following the first user utterance text and the decoded text in the utterance text is reduced.

[0128] In step S26, the language understanding unit 15 determines whether or not to terminate the machine learning of the language model. If it is determined that the machine learning of the language model should be terminated, the process proceeds to step S27. On the other hand, if it is not determined that the machine learning of the language model should be terminated, the processes in steps S21 to S25 are repeated using the utterance texts (first and second user utterance texts) as training data.

[0129] In step S27, the language understanding unit 15 outputs the trained language model and user embedding representations. The language understanding unit 15 may, for example, store the trained language model in a predetermined storage means. The language understanding unit 15 may also store the trained user embedding representations in a predetermined storage means, or have them managed by the user embedding representation management unit 22.

[0130] Referring again to Figure 21, in step S3, the topic extraction unit 16 extracts topic words from the utterance text, which are words or phrases that represent the topic in the user's utterance.

[0131] In step S4, the embedding expression acquisition unit 17 inputs the topic word into the learned embedding unit en and acquires the topic embedding expression output from the embedding unit en. Here, the embedding expression acquisition unit 17 may further acquire a location embedding expression output from the embedding unit en by inputting location text representing a place into the learned embedding unit en.

[0132] In step S5, the relationship extraction unit 18 generates a relationship graph with at least the user and topic as nodes, based on the user's speech history (speech log) and action history. The relationship extraction unit 18 may also generate a relationship graph that further includes location as a node.

[0133] In step S6, the relation learning unit 19 performs training of a graph neural network, using the trained user embedding representation and topic embedding representation as features of the user and topic nodes in the relation graph. The relation graph used for training may further include locations as nodes, and location embedding representations may be used as features of the location nodes.

[0134] In step S7, the relation learning unit 19 learns a graph neural network of relation graphs where the embedding representations are the features of each node, thereby modifying the features and weights of each node and obtaining the learned embedding representations for each node.

[0135] In step S8, the embedding representation output unit 20 outputs the embedding representation of each node that has been learned by the relation learning unit 19.

[0136] Next, with reference to Figure 23, an embedded expression generation program for causing a computer to function as the embedded expression generation device 10 of this embodiment will be described. Figure 23 is a diagram showing the configuration of the embedded expression generation program. The embedded expression generation program P1 is composed of a main module m10 that comprehensively controls the embedded expression generation process in the embedded expression generation device 10, a speech log acquisition module m11, a speech recognition module m12, a text acquisition module m13, an emotion acquisition module m14, a language understanding module m15, a topic extraction module m16, an embedded expression acquisition module m17, a relationship extraction module m18, a relationship learning module m19, an embedded expression output module m20, and a link prediction module m21. Each of the modules m11 to m21 realizes the respective functions for each of the functional units 11 to 21.

[0137] The embedded expression generation program P1 may be transmitted via a transmission medium such as a communication line, or it may be stored in a recording medium M1, as shown in Figure 23.

[0138] According to the embedded expression generation device 10, embedded expression generation method, and embedded expression generation program P1 of this embodiment described above, a language model composed of an encoder-decoder model uses a pair of first and second user utterance texts as training data, inputs the first user utterance text into the embedding unit, synthesizes the user utterance embedding expression obtained from inputting the first user utterance text into the embedding unit, and inputs the synthesized embedding expression obtained from the user embedding expression into the decoding unit. The language model and user embedding expression are machine-learned so that the error between the decoded text output from the decoding unit and the second user utterance text is reduced, thereby obtaining an embedding unit (encoder) that outputs a suitable topic embedding expression in response to the input of a topic word, and a user embedding expression that appropriately reflects the characteristics of the user. A relationship graph is generated with the user and topic as nodes, and edges are drawn between the nodes based on the history of the user's utterances and actions. By learning a graph neural network that uses the topic embedding expression obtained by inputting a topic word into the embedding unit and the learned user embedding expression as features of the topic word and the user, respectively, learned topic embedding expressions and user embedding expressions that appropriately reflect the characteristics of the topic word and the user are obtained. Since the obtained topic embedding representations and user embedding representations reflect the relationships between those entities, it is possible to calculate the distance between the user and the topic.

[0139] The invention described herein can be understood, for example, as follows:

[0140] A parameter acquisition system relating to the first aspect of this disclosure is a parameter acquisition system for acquiring parameters to be set for a character to operate in a virtual space, comprising: a topic acquisition unit that acquires at least one topic whose proximity to a user embedding representation, which is an embedding representation of a user represented by a real number vector, and a topic embedding representation, which is an embedding representation of a topic represented by a real number vector, meets a predetermined condition; a hobby acquisition unit that acquires hobbies corresponding to the topics acquired by the topic acquisition unit based on correspondence information representing the correspondence relationship between topics and hobbies; and a setting information output unit that outputs the hobbies acquired by the hobby acquisition unit as hobby information for setting the parameters of a character corresponding to a user.

[0141] Based on the above aspects, the distance between the user and the topic can be calculated based on user embeddings and topic embeddings that represent the characteristics of the user and the topic, respectively. Therefore, topics that meet the predetermined conditions for proximity to the user can be acquired. Then, based on the corresponding information, hobbies corresponding to the topic are acquired. Consequently, the acquired hobbies have a certain degree of closeness to the user. Since the acquired hobbies are output as hobby information, it becomes possible to apply the hobby information to the parameters set for the user's character.

[0142] In the parameter acquisition system relating to the second aspect, the hobby acquisition unit may refer to a thesaurus as correspondence information that defines the relationships between multiple words, including hobby terms representing hobbies and topic terms representing topics, in the parameter acquisition system relating to the first aspect, and acquire hobbies corresponding to hobby terms associated with topic terms corresponding to topics acquired by the topic acquisition unit.

[0143] Based on the above aspects, by referring to correspondence information composed of a thesaurus that defines the relationship between words representing hobbies and topics, corresponding hobby terms are extracted based on topic terms that represent topics closely related to the user. Therefore, hobbies represented by hobby terms can be output as hobby information closely related to the user.

[0144] In the parameter acquisition system relating to the third aspect, the hobby acquisition unit may, in the parameter acquisition system relating to the first aspect, refer to a given hobby list containing multiple hobby words representing hobbies, calculate the similarity between each topic word representing a topic acquired by the topic acquisition unit and the hobby words included in the hobby list as correspondence information, and acquire hobbies corresponding to hobby words whose calculated similarity is equal to or greater than a given threshold.

[0145] Based on the above aspects, hobbies are obtained that are represented by hobby terms that have a high degree of similarity to topic terms that represent topics closely related to the user. Therefore, hobbies closely related to the user can be output as hobby information.

[0146] In the parameter acquisition system relating to the fourth aspect, the hobby acquisition unit in the parameter acquisition system relating to the third aspect may calculate the similarity between topic words and hobby words using Word2Vec.

[0147] Based on the above aspects, the similarity between hobby-related terms and topic terms included in the hobby list can be calculated with high accuracy.

[0148] The parameter acquisition system relating to the fifth aspect further includes an attribute acquisition unit that, in the parameter acquisition system relating to any one of the first to fourth aspects, refers to a given attribute list that associates hobbies with a person's attribute information and acquires attribute information associated with hobbies acquired by the hobby acquisition unit, and the setting information output unit may output the attribute information acquired by the attribute information as information for setting the parameters of a character corresponding to the user.

[0149] Based on the above aspects, attribute information associated with hobbies obtained by referencing the attribute list can be retrieved, and therefore, attribute information corresponding to the user can be output as information for setting character parameters. Thus, in addition to hobbies, attribute information can be set as character parameters.

[0150] In the parameter acquisition system relating to the sixth aspect, in the parameter acquisition system relating to any one of the first to fifth aspects, the topic acquisition unit may acquire a predetermined number of topics in which the proximity between the user's embedded expression and the topic's embedded expression is high, or acquire topics in which the distance between the user's embedded expression and the topic's embedded expression is less than or equal to a predetermined amount.

[0151] Based on the above aspects, topics that have a suitable level of closeness to the user can be obtained. Hobby information suitable for setting as parameters for the user's character can also be obtained.

[0152] The parameter acquisition system relating to the seventh aspect further comprises an embedded expression input unit that acquires user embedded expressions and topic embedded expressions from an embedded expression generation device that generates user and topic embedded expressions, in the parameter acquisition system relating to any one of the first to sixth aspects, wherein the embedded expression generation device is a language understanding unit that learns a language model composed of an encoder-decoder model including an embedding unit and a decoding unit, the embedding unit outputs an embedded expression that represents the features of the input text, the decoding unit decodes the embedded expression which includes at least the output from the embedding unit, the first user utterance text which represents the content of an utterance of one user from among the utterance text which represents the content of an utterance of a user inputs to the embedding unit, the decoded text which is output from the decoding unit by inputting a composite embedded expression which is a combination of the user utterance embedded expression and the user embedded expression of the said one user inputs to the decoding unit, and the second user utterance text which follows the first user utterance text and the decoded text in the utterance text The system may also include: a language understanding unit that performs machine learning to adjust the language model and user embedding representations so that the error with the Kist is small, and the user embedding representations are either the initial user embedding representations before learning or user embedding representations during the learning process; a topic extraction unit that extracts topic words, which are words or phrases that represent the topic in the user's utterance, from the utterance text; an embedding representation acquisition unit that inputs the topic words into a learned embedding unit and obtains topic embedding representations output from the embedding unit; a relationship extraction unit that generates a relationship graph, which is a graph in which at least the user and the topic are nodes, the history of dialogue between users are edges connecting the users, and the history of the user's utterance of topic words are edges connecting the user and the topic; a relationship learning unit that obtains learned embedding representations for each node by learning a graph neural network, which uses the learned user embedding representations and topic embedding representations as features of the user and topic nodes in the relationship graph; and an embedding representation output unit that outputs the learned embedding representations for each node to an embedding representation input unit.

[0153] As described above, a language model composed of an encoder-decoder model uses a pair of first and second user utterance texts as training data. The first user utterance text is input to the embedding unit, and the resulting user utterance embedding expression is synthesized with the user embedding expression. This composite embedding expression is then input to the decoding unit. The language model and user embedding expression are machine-learned to minimize the error between the decoded text output from the decoding unit and the second user utterance text. This results in an embedding unit (encoder) that outputs a suitable topic embedding expression in response to the input of a topic word, as well as a user embedding expression that appropriately reflects the user's characteristics. A relationship graph is then generated with the user and topic as nodes, and edges are drawn between the nodes based on the user's utterance and behavior history. By inputting a topic word into the embedding unit, the topic embedding expression and the learned user embedding expression are used as feature quantities for the topic word and the user, respectively. This learning process of a graph neural network then yields a learned topic embedding expression and a user embedding expression that appropriately reflects the characteristics of the topic word and the user. The resulting topic embedding representations and user embedding representations reflect the relationships between those entities, making it possible and preferable to calculate the distance between the user and the topic.

[0154] Furthermore, if we consider the embedded representation generation device in the parameter acquisition system relating to the seventh aspect as the embedded representation generation device relating to the first aspect, then the embedded representation generation device relating to the first aspect has the following other aspects.

[0155] The embedding representation generation device relating to the second aspect further includes an emotion acquisition unit that, in the embedding representation generation device relating to the first aspect, acquires emotion information representing the user's emotions at the time the user utters an utterance based on the voice of the utterance or the user's facial expression, and associates the acquired emotion information with the utterance text representing the content of the utterance, and the language understanding unit may perform machine learning to adjust the language model and user embedding representation using the utterance text to which emotion information representing a predetermined positive emotion is associated.

[0156] Based on the above aspects, speech texts representing utterances made by users when they are likely to be experiencing positive emotions are used in machine learning. Therefore, the combination of first and second user speech texts that constitute the training data is a combination that is likely to be expressed when the user is experiencing positive emotions. By performing machine learning using such training data, an embedding unit and a user embedding expression capable of generating topic embedding expressions that reflect a favorable relationship between the topic word and the user can be obtained.

[0157] In the embedding representation generation device relating to the third aspect, in the embedding representation generation device relating to the first or second aspect, the embedding representation acquisition unit further acquires location embedding representations output from the embedding unit by inputting location text representing a place into the learned embedding unit, the relationship extraction unit generates a relationship graph which is a graph in which at least the user, topic and place are nodes, the history of dialogue between users are edges connecting users, the history of the user uttering topic words are edges connecting the user and the topic, and the history of the user visiting a place are edges connecting the user and the place, and the relationship learning unit may obtain learned embedding representations for each node by learning a graph neural network which uses each of the learned user embedding representations, topic embedding representations and location embedding representations as features of the user, topic and place nodes in the relationship graph.

[0158] Based on the above aspects, by inputting location text into the embedding section of a trained language model, a location embedding representation that appropriately reflects the characteristics of the location can be obtained. Then, a relationship graph is generated with the user, topic, and location as nodes, and edges are drawn between the nodes based on the user's utterance and behavior history. By training a graph neural network that uses the topic embedding representation, location embedding representation, and trained user embedding representation as features of the topic word, location, and user, respectively, trained topic embedding representations, location embedding representations, and user embedding representations that appropriately reflect the characteristics of the topic word, location, and user can be obtained. Since the relationships between these entities are reflected in the obtained topic embedding representations, location embedding representations, and user embedding representations, it is possible to calculate the distance between the user and the topic and location.

[0159] In the fourth aspect of the embedding representation generation device, the embedding representation generation device for any one of the first to third aspects may further include a link prediction unit that calculates the distance between nodes based on the learned embedding representation of each node, and calculates link prediction information indicating the possibility of edges being drawn between each node based on the calculated distance between nodes.

[0160] Based on the above aspects, training a graph neural network on a relational graph yields an embedding representation expressed as real vectors, which allows for the calculation of distances between entities of different types. This enables the calculation of link prediction information that allows for the evaluation of the likelihood of edges being formed between each node in the graph. Consequently, it becomes possible to predict whether there is a relationship of a certain degree or greater between the entities corresponding to each node.

[0161] In the embedded representation generation device relating to the fifth aspect, the link prediction unit in the embedded representation generation device relating to the fourth aspect may output information indicating each node whose distance between nodes is less than or equal to a given threshold, based on a given threshold for the distance between nodes, as link prediction information.

[0162] Based on the aspects described above, it becomes possible to obtain information about entities that have a relationship of a certain degree or higher, based on information indicating nodes whose distance from other nodes is below a given threshold.

[0163] In the embedded representation generation device relating to the sixth aspect, in the embedded representation generation device relating to any one of the first to fifth aspects, the utterance text may be obtained based on an utterance log of speech or text representing the content of the user's utterance in a predetermined virtual space.

[0164] Based on the above aspects, in a virtual space, it is easy to obtain audio or text representing user utterances, thus facilitating the acquisition of spoken text.

[0165] In the embedded representation generation device relating to the seventh aspect, in the embedded representation generation device relating to any one of the first to sixth aspects, the relationship extraction unit may generate a relationship graph based on the user's speech history and action history in a predetermined virtual space.

[0166] Based on the above aspects, in a virtual space, it is easy to obtain the user's speech history and action history, making it easy to generate relational graphs.

[0167] Although the present disclosure has been described in detail above, it will be clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the invention as defined by the claims. Therefore, the descriptions herein are for illustrative purposes only and are not intended to be restrictive in any way.

[0168] Information notification is not limited to the embodiments described herein and may be carried out by other means. For example, information notification may be carried out by physical layer signaling (e.g., DCI (Downlink Control Information), UCI (Uplink Control Information)), upper layer signaling (e.g., RRC (Radio Resource Control) signaling, MAC (Medium Access Control) signaling, broadcast information (MIB (Master Information Block), SIB (System Information Block))), other signals, or combinations thereof. RRC signaling may also be called RRC messages, and may be, for example, RRC Connection Setup messages, RRC Connection Reconfiguration messages, etc.

[0169] Each aspect / embodiment described herein may be applied to systems utilizing LTE (Long Term Evolution), LTE-A (LTE-Advanced), SUPER 3G, IMT-Advanced, 4G, 5G, FRA (Future Radio Access), W-CDMA®, GSM®, CDMA2000, UMB (Ultra Mobile Broadband), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, UWB (Ultra-WideBand), Bluetooth®, and other appropriate systems, and / or next-generation systems extended based thereon. Furthermore, multiple systems may be applied in combination (for example, a combination of at least one of LTE and LTE-A with 5G).

[0170] The processing procedures, sequences, flowcharts, etc., of each aspect / embodiment described herein may be reordered, provided they are consistent with each other. For example, the methods described herein present the elements of various steps in an exemplary order and are not limited to that specific order.

[0171] The specific operations described in this disclosure as being performed by a base station may, in some cases, be performed by its upper node. In a network consisting of one or more network nodes having a base station, it is clear that various operations performed for communication with a terminal can be performed by the base station and at least one other network node (for example, an MME or S-GW, but not limited to these). Although the above example illustrates a case where there is one other network node besides the base station, it may also be a combination of multiple other network nodes (for example, an MME and an S-GW).

[0172] Information, etc. (see the "Information, Signals" section) can be output from a higher layer (or lower layer) to a lower layer (or higher layer). Input and output may also occur via multiple network nodes.

[0173] Input and output information may be stored in a specific location (e.g., memory) or managed in a management table. Input and output information may be overwritten, updated, or appended to. Output information may be deleted. Input information may be sent to other devices.

[0174] The determination may be made by a value represented by 1 bit (0 or 1), by a boolean value (true or false), or by a numerical comparison (for example, a comparison with a predetermined value).

[0175] Each aspect / embodiment described herein may be used individually, in combination, or switched between as needed during implementation. Furthermore, notification of specific information (e.g., notification that "X is") is not limited to explicit notification, but may also be implicit (e.g., by not providing such notification).

[0176] Software should be broadly interpreted to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, execution threads, procedures, functions, and so on, whether they are called software, firmware, middleware, microcode, hardware description languages, or by any other name.

[0177] Furthermore, software, instructions, etc., may be transmitted and received via a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies such as coaxial cable, fiber optic cable, twisted pair, and digital subscriber lines (DSL) and / or wireless technologies such as infrared, radio, and microwave, these wired and / or wireless technologies are included in the definition of a transmission medium.

[0178] The information, signals, etc. described in this disclosure may be represented using any of the various different techniques. For example, the data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0179] In addition, terms described in this disclosure and / or terms necessary for understanding this specification may be replaced with terms having the same or similar meaning.

[0180] The terms “system” and “network” as used herein are interchangeable.

[0181] Furthermore, the information, parameters, etc., described herein may be expressed as absolute values, relative values ​​from a given value, or by corresponding other information. For example, wireless resources may be indicated by an index.

[0182] The names used for the parameters described above are not restrictive in any way. Furthermore, the formulas and other expressions using these parameters may differ from those expressly disclosed in this disclosure. Various channels (e.g., PUCCH, PDCCH, etc.) and information elements can be identified by any suitable name, and therefore, the various names assigned to these various channels and information elements are not restrictive in any way.

[0183] As used in this disclosure, the terms “determining” and “determining” may encompass a wide variety of actions. “Determining” may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiry (e.g., searching in a table, database, or other data structure), and ascertaining. “Determining” may also include, for example, receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, and accessing (e.g., accessing data in memory). Furthermore, "judgment" and "decision" can include considering something as having been "judged" or "decided" after resolving, selecting, choosing, establishing, comparing, etc. In other words, "judgment" and "decision" can include considering something as having been "judged" or "decided" after some action. Also, "judgment (decision)" can be reinterpreted as "assuming," "expecting," or "considering."

[0184] As used in this disclosure, the phrase "based on" does not mean "based solely on" unless otherwise specified. In other words, the phrase "based on" means both "based solely on" and "based at least on."

[0185] Where the designations “first,” “second,” etc., are used herein, no reference to those elements shall generally limit the quantity or order of those elements. These designations may be used herein as a convenient way to distinguish between two or more elements. Thus, references to the first and second elements shall not imply that only two elements may be employed therein, or that the first element must precede the second element in any way.

[0186] To the extent that “include,” “including,” and their variations are used herein or in the claims, these terms are intended to be inclusive, as is the term “comprising.” Furthermore, the term “or” as used herein or in the claims is not intended to be exclusive OR.

[0187] In this disclosure, if articles are added through translation, such as a, an, and the in English, this disclosure may include the fact that the noun following these articles is plural.

[0188] In this disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "combine" may be interpreted similarly to "different." [Explanation of symbols]

[0189] 1...Parameter acquisition system, 10...Embedded expression generation device, 11...Speech log acquisition unit, 12...Speech recognition unit, 13...Text acquisition unit, 14...Emotion acquisition unit, 15...Language understanding unit, 16...Topic extraction unit, 17...Embedded expression acquisition unit, 18...Relationship extraction unit, 19...Relationship learning unit, 20...Embedded expression output unit, 21...Link prediction unit, 22...Embedded expression management unit, 30...Parameter acquisition device, 31...Embedded expression input unit, 32...Topic acquisition unit, 33...Hobby acquisition unit, 34...Attribute acquisition unit, 35...Setting information output unit, 41...Input device, 42...Microphone, 43...Camera, M1...Recording medium, m10...Main module, m11...Speech log acquisition module, m1 2...Speech recognition module, m13...Text acquisition module, m14...Emotion acquisition module, m15...Language understanding module, m16...Topic extraction module, m17...Embedded expression acquisition module, m18...Relationship extraction module, m19...Relationship learning module, m20...Embedded expression output module, m21...Link prediction module, M3...Recording medium, m30...Main module, m31...Embedded expression input module, m32...Topic acquisition module, m33...Hobby acquisition module, m34...Attribute acquisition module, m35...Setting information output module, P1...Embedded expression generation program, P3...Parameter acquisition program.

Claims

1. A parameter acquisition system that acquires parameters to be set for a character that operates in a virtual space, A topic acquisition unit acquires at least one topic whose proximity to a user embedding representation, which is an embedding representation of a user as a real number vector, and a topic embedding representation, which is an embedding representation of a topic as a real number vector, satisfies a predetermined condition. A hobby acquisition unit acquires hobbies corresponding to topics acquired by the topic acquisition unit, based on correspondence information that represents the relationship between topics and hobbies. A setting information output unit outputs the hobbies acquired by the hobby acquisition unit as hobby information for setting the parameters of the character corresponding to the user, A parameter acquisition system equipped with the following features.

2. The hobby acquisition unit refers to a thesaurus as correspondence information that defines the relationships between multiple words, including hobby terms representing hobbies and topic terms representing topics, and acquires hobbies corresponding to hobby terms associated with topic terms corresponding to topics acquired by the topic acquisition unit. The parameter acquisition system according to claim 1.

3. The hobby acquisition unit refers to a given hobby list containing multiple hobby terms that represent hobbies, calculates the similarity between each topic term representing a topic acquired by the topic acquisition unit and a hobby term included in the hobby list as correspondence information, and acquires hobbies corresponding to hobby terms whose calculated similarity is equal to or greater than a given threshold. The parameter acquisition system according to claim 1.

4. The hobby acquisition unit calculates the similarity between the topic word and the hobby word using Word2Vec. The parameter acquisition system according to claim 3.

5. The system further includes an attribute acquisition unit that references a given attribute list relating hobbies to a person's attribute information and acquires attribute information associated with the hobbies acquired by the hobby acquisition unit, The setting information output unit outputs the attribute information obtained from the attribute information as information for setting the parameters of the character corresponding to the user. The parameter acquisition system according to claim 1.

6. The aforementioned topic acquisition unit, The user embedding representation of the user and the topic embedding representation of the topic are obtained in order of proximity, or The user embedding representation of the user and the topic embedding representation of the topic are to be acquired if the distance between them is less than or equal to a predetermined amount. The parameter acquisition system according to claim 1.

7. The system further comprises an embedded expression input unit that acquires the user embedded expression and the topic embedded expression from an embedded expression generation device that generates at least user and topic embedded expressions, The embedded representation generation device is A language understanding unit that learns a language model composed of an encoder-decoder model including an embedding unit and a decoding unit, The aforementioned embedding unit outputs an embedding representation that shows the characteristics of the input text. The decoding unit decodes the embedded representation which includes at least the output from the embedding unit. By inputting a first user utterance text representing the content of one user's utterance from among the utterance texts representing the content of the user's utterance into the embedding unit, a user utterance embedding representation output from the embedding unit is obtained; by inputting a composite embedding representation obtained by combining the user utterance embedding representation and the user embedding representation of the said one user into the decoding unit, a decoded text output from the decoding unit is obtained; and machine learning is performed to adjust the language model and the user embedding representation so that the error between the second user utterance text following the first user utterance text and the decoded text in the utterance text is reduced. The user embedding representation is an initial user embedding representation before learning or a user embedding representation during the learning process, and is a language comprehension unit. A topic extraction unit extracts topic words from the aforementioned utterance text, which are words or phrases that represent the topic of the user's utterance. An embedding expression acquisition unit inputs the aforementioned topic word into the embedding unit which has already learned the topic word and acquires the topic embedding expression output from the embedding unit, A relationship extraction unit generates a relationship graph based on the user's utterance history and action history, in which at least the user and the topic are nodes, the history of dialogue between users are edges connecting the users, and the history of the user uttering the topic word is an edge connecting the user and the topic. A relational learning unit obtains a learned embedding representation for each node by training a graph neural network that uses the learned user embedding representation and the topic embedding representation, respectively, as features of the user and topic nodes in the relational graph. The system includes an embedding representation output unit that outputs the learned embedding representation of each node to the embedding representation input unit. The parameter acquisition system according to claim 1.

Images