Information processing device and program

Abstract

This technology provides a method for ensuring that requests received from two or more users simultaneously are executed in the appropriate order. [Solution] The information processing device according to the first aspect of this disclosure acquires conversation data, which is a record of conversations between multiple users, analyzes the acquired conversation data to rank the users who have the initiative in the conversation, and when it receives requests from at least two or more of the multiple users at the same time, it determines the priority of the received requests based on the ranking result and has an agent carry out the requests according to the determined priority.

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus and a program.

Background Art

[0002] Patent Document 1 proposes a wearable terminal including an important word database, a topic word database, a microphone, and a camera. The proposed wearable terminal recognizes topic words and important words from voice information, recognizes the smiling time and the number of nods of a conversation partner per unit conversation time from video, analyzes the partner's receptivity to the conversation and conversation skill from the voice and video recognition results, and feeds back the receptivity and the conversation skill to a user. Patent Document 2 proposes an apparatus for analyzing human relationships between occupants.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of Invention

Problems to be Solved by the Invention

[0004] One of the objects of the present disclosure is to provide a technique for achieving execution of requests received from two or more users at the same time in an appropriate order.

Means for Solving the Problems

[0005] An information processing device according to a first aspect of this disclosure includes a control unit. The control unit is configured to acquire conversation data, which is a record of conversations between multiple users; to analyze the acquired conversation data to rank the users in terms of who has the initiative in the conversation; to determine the priority of the received requests based on the ranking results when requests are received from at least two or more of the multiple users at the same time; and to have an agent carry out the requests according to the determined priority.

[0006] The program relating to the second aspect of this disclosure is a program for causing a computer to execute an information processing method. The information processing method includes acquiring conversation data, which is a record of conversations between multiple users; analyzing the acquired conversation data to rank the users in terms of who has the initiative in the conversation; determining the priority of the received requests based on the ranking results when requests are received simultaneously from at least two of the multiple users; and causing an agent to carry out the requests according to the determined priority. A trained machine learning model may be used for at least one of the receiving of requests and the carrying out of the received requests. [Effects of the Invention]

[0007] According to this disclosure, it can be expected that requests received from two or more users at the same time will be fulfilled in the appropriate order. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 schematically illustrates an example of a scenario in which this disclosure applies. [Figure 2] Figure 2 schematically shows an example of the elements of conversation data analysis. [Figure 3] Figure 3 schematically illustrates one example of a method for determining priority. [Figure 4] Figure 4 schematically shows an example of the hardware configuration of an information processing device. [Figure 5]Figure 5 is a flowchart showing an example of the processing procedure for ranking. [Figure 6] Figure 6 is a flowchart showing an example of the processing procedure for fulfilling a request. [Modes for carrying out the invention]

[0009] In recent years, there has been progress in the development of agent technology that accepts and carries out requests based on user utterances. These agents are composed of, for example, trained machine learning models such as large-scale generative models. This agent technology allows for the fulfillment of various types of requests made through utterances. However, the inventors of this case have found the following problems with conventional agent technology: Specifically, if two or more users give requests to the agent at the same time, there is a possibility of a conflict in request acceptance. If a conflict occurs, the agent may fulfill requests in an inappropriate order, such as performing a request that could be performed later first.

[0010] In contrast, the information processing device according to the first aspect of this disclosure includes a control unit. The control unit is configured to acquire conversation data, which is a record of conversations between multiple users; to analyze the acquired conversation data to rank the users in terms of who has the initiative in the conversation; to determine the priority of the received requests based on the ranking results when requests are received from at least two or more of the multiple users at the same time; and to have an agent carry out the requests according to the determined priority.

[0011] In the first aspect of this disclosure, a ranking of conversational leadership among multiple users is determined based on conversations between them. When requests are received from two or more users simultaneously, the priority of the received requests is determined based on the ranking. The received requests are carried out according to the determined priority (i.e., requests are carried out in order of highest priority). Conversational leadership allows for an evaluation of the degree to which each user controls the conversational space. The higher the degree of control over the conversational space, the more likely it is to make major utterances, and thus more likely to give important requests to the agent (more likely to lead important requests). In other words, it is presumed that the more leadership a user has, the more important the requests they give. According to the first aspect of this disclosure, based on this perspective, by utilizing the results of ranking conversational leadership, the priority of requests received from two or more users simultaneously can be appropriately determined. As a result, it can be expected that requests received from two or more users simultaneously will be carried out in the appropriate order.

[0012] The forms of the information processing apparatus described herein are not limited to those described above. As another form of the information processing apparatus described above, one aspect of the disclosure may be at least one of an information processing method, a program, and a machine-readable storage medium that stores such a program, which implement at least a part of each of the above components. Here, a machine-readable storage medium may be a non-temporary medium that stores information such as a program by electrical, magnetic, optical, mechanical, or chemical action. A non-temporary storage medium may include storage media (CDs, DVDs, semiconductor memory, etc.), auxiliary storage devices of a computer, external storage devices connected to a computer, etc.

[0013] For example, the program relating to the second aspect of this disclosure may be a program that causes a computer to execute an information processing method. The information processing method may include acquiring conversation data, which is a record of conversations between multiple users; analyzing the acquired conversation data to rank the users in terms of who has the initiative in the conversation; determining the priority of the received requests based on the ranking results when requests are received from at least two or more of the multiple users at the same time; and causing an agent to carry out the requests according to the determined priority.

[0014] For example, the information processing method relating to the third aspect of this disclosure may be performed by a computer. The information processing method may include: acquiring conversation data, which is a record of conversations between multiple users; analyzing the acquired conversation data to rank the users in terms of who has the initiative in the conversation; determining the priority of the received requests based on the ranking results when requests are received from at least two or more of the multiple users at the same time; and having an agent perform the requests according to the determined priority.

[0015] Hereinafter, embodiments relating to one aspect of this disclosure will be described with reference to the drawings. However, the embodiments described below are merely illustrative in all respects of this disclosure. Various improvements or modifications may be made without departing from the scope of this disclosure. In implementing this disclosure, specific configurations may be adopted as appropriate depending on the embodiment. In this embodiment, the data appearing is described in natural language, but more specifically, it is specified in pseudo-language, commands, parameters, machine code, electrical signals, etc., that can be recognized by machines such as computers.

[0016] [1. Application Examples] Figure 1 schematically shows an example of a scenario to which this disclosure applies. The information processing device 1 according to this embodiment is one or more computers configured to have an agent perform a request received from a user.

[0017] In the present embodiment, the information processing apparatus 1 acquires conversation data 20 which is a record of conversations held between a plurality of users U. By analyzing the acquired conversation data 20, the information processing apparatus 1 assigns a rank 25 representing the order of having the initiative in the conversation among the plurality of users U. Further, the information processing apparatus 1 accepts a request 30 made by utterance, and causes an agent A0 to execute the accepted request 30. The information processing apparatus 1 may accept requests 30 from at least two or more users U among the plurality of users U in the same period. When the information processing apparatus 1 accepts requests 30 from at least two or more users U among the plurality of users U in the same period, the information processing apparatus 1 determines the priority 50 of the accepted requests 30 based on the result of the assigned ranks 25. The information processing apparatus 1 causes the agent A0 to execute the requests 30 in accordance with the determined priorities 50. That is, the information processing apparatus 1 causes the agent A0 to execute the requests 30 in descending order of the priority 50.

[0018] Note that the method of accepting the request 30 is not particularly limited, and may be appropriately selected according to the embodiment. A known method may be adopted as the method of accepting the request 30. In one example, the information processing apparatus 1 may acquire utterance data 300 obtained by recording utterance with a microphone MA. The information processing apparatus 1 may identify the utterance content by performing speech analysis on the acquired utterance data 300. The information processing apparatus 1 may determine whether the request 30 is included by performing linguistic analysis on the identified utterance content. Known methods may be adopted as the methods of speech analysis and linguistic analysis. By using a large-scale generative model such as a large-scale speech model, the speech analysis and the linguistic analysis may be executed as an integrated process. When the request 30 is included in the utterance content, the information processing apparatus 1 may provide the request 30 to the agent A0, thereby causing the agent A0 to execute the processing of the request 30.

[0019] As described above, it is inferred that the higher the degree of initiative a user U has, the higher the importance of the request 30 given by that user U. According to the present embodiment, based on this viewpoint, by using the result of assigning the rank 25 of having the initiative in the conversation, when requests 30 are received from two or more users U in the same period, the priority 50 of the requests 30 received from each user U can be appropriately determined. As a result, it is possible to expect the requests 30 received from two or more users U in the same period to be executed in an appropriate order.

[0020] (User) The user U may be any person who can give the request 30 to the agent A0. The user U may be a real entity or a virtual entity. The virtual entity may include, for example, a robot, a video character, and the like. The scenario for receiving the request 30 from the user U is not particularly limited, and may be appropriately selected according to the embodiment. In one example, the information processing apparatus 1 may receive the request 30 from the user U in the vehicle MV. In this case, the plurality of users U may include at least one of a driver and a passenger. The type of the vehicle MV (number of wheels, power source, size, etc.) may be arbitrarily selected. The vehicle MV may include a manually driven vehicle and an autonomous driving vehicle.

[0021] (Agent) Agent A0 may consist of an application for performing request 30. Performing request 30 may involve performing information processing in accordance with request 30. The information processing in accordance with request 30 may include any type of information processing that a computer can perform. The configuration of Agent A0 is not particularly limited as long as it can perform request 30, and may be appropriately selected depending on the embodiment. In one example, Agent A0 may consist of an computational model such as a rule-based model or a trained machine learning model. A rule-based model is configured to derive an inference result (such as the result of performing request 30) from a given input according to rules. The rules may be given as appropriate. The rule-based model may include an expert system. A machine learning model is configured to have one or more computational parameters that can be adjusted by machine learning. One or more computational parameters are used for the computation of the desired inference (such as performing request 30). The machine learning model may consist of, for example, a neural network, a regression model, a decision tree model, a support vector machine, or other functional expressions. The machine learning method may be appropriately selected depending on the machine learning model adopted (e.g., backpropagation). The trained machine learning model may include large-scale generative models such as large-scale language models, large-scale visual language models, and large-scale speech models. Agent A0 may be a known agent.

[0022] Agent A0 may be deployed on the information processing device 1, or on another computer other than the information processing device 1. If Agent A0 is deployed on another computer, the information processing device 1 may be connected to the other computer as appropriate. The information processing device 1 may have Agent A0 perform the received request 30 by passing it on to the other computer. The information processing device 1 may obtain the result of Agent A0's performance of request 30 from the other computer.

[0023] (Conversation data) The composition of the conversation data 20 is not particularly limited, and can be appropriately selected depending on the embodiment, as long as it includes a record of a conversation between multiple users U. For example, the conversation record may include audio and images. For example, a microphone MA and an imaging device CA may be installed in the conversation space of multiple users U. As a specific example, in a scenario where a request 30 is received inside a vehicle MV, the vehicle MV may be equipped with a driver monitoring system including a microphone and an imaging device. The microphone and imaging device of this driver monitoring system may be used as at least part of the microphone MA and imaging device CA. However, the method of installing the microphone MA and imaging device CA is not limited to this example. Microphones and imaging devices other than those of the driver monitoring system may be used as the microphone MA and imaging device CA. The microphone MA and imaging device CA may be installed arbitrarily. Also, the number of microphone MAs and imaging device CAs may be arbitrarily selected. The information processing device 1 may directly acquire audio data 201 and image data 203 from the microphone MA and imaging device CA, or it may acquire them indirectly via a storage medium or the like. The conversation data 20 may consist of acquired audio data 201 and image data 203. The audio data 201 may be used in the format of an audio file, or it may be used after being converted to a text file format by speech analysis.

[0024] The audio and images reflect the content and context of the conversation. Therefore, according to one example of this embodiment, it can be expected that the analysis regarding leadership, described later, can be performed favorably because the recording of the conversation includes both audio and images. However, the configuration of the conversation data 20 is not limited to this example and may be modified as appropriate depending on the embodiment. In another example, the conversation data 20 may consist only of audio data 201. The conversation data 20 may further include any data other than the audio data 201 and image data 203.

[0025] (Analysis elements of conversation data) The degree to which a user controls the conversation space corresponds to the level of control they have over the conversation. A dominant user U has a higher ranking 25 for their leadership, while a subordinate user U has a lower ranking 25 for their leadership. As long as such ranking is possible, the analytical elements in the conversation data 20 are not particularly limited and may be appropriately selected depending on the embodiment.

[0026] Figure 2 schematically shows an example of the analysis elements of conversation data 20 according to this embodiment. In one example, analyzing conversation data 20 may include analyzing at least one of the relationships 211 between multiple users U appearing in the conversation data 20, the amount of knowledge 213 of each of the multiple users U, the amount of speech 215 of each of the multiple users U, and the speech characteristics 217 of each of the multiple users U. Accordingly, the ranking 25 of those who have the initiative in the conversation among the multiple users U may be assigned according to at least one of the analyzed relationships 211, amount of knowledge 213, amount of speech 215, and speech characteristics 217.

[0027] Each element may be analyzed in any way. For example, computational models such as rule-based models and trained machine learning models may be used to analyze each element. The trained machine learning model may include large-scale generative models. Known analysis methods may be used to analyze each element. For example, the information processing device 1 may identify each user U that appears in the conversation data 20. Known methods such as voiceprint recognition and facial image recognition may be used to identify user U. The information processing device 1 may analyze each element from the data of each identified user U contained in the conversation data 20.

[0028] The relationship 211 represents the relationship between each user U. The type of relationship 211 may be defined as appropriate depending on the embodiment. For example, the type of relationship 211 may include kinship, friendship, work relationships, etc. Kinship relationships may include, for example, parent-child, husband-wife, relatives, etc. Work relationships may include, for example, superior-subordinate, colleague, customer, etc.

[0029] The information processing device 1 may identify the utterance period of each user U in the conversation data 20 from the identification result of each user U, and obtain the utterance content of each user U from the identified utterance period. The information processing device 1 may identify the utterance characteristics 217 of each user U from the obtained utterance content. If the conversation data 20 includes image data 203, the information processing device 1 may estimate the attributes of each user U, such as age and gender, from the image data 203. The information processing device 1 may estimate the relationship between each user U 211 from at least one of the utterance content, utterance characteristics 217, and attributes of each user U by using a computational model (rule-based model, trained machine learning model, etc.). Note that the method for analyzing the relationship between the users 211 is not limited to this example and may be appropriately modified depending on the embodiment. Known methods such as those described in Patent Document 2 may be used for analyzing the relationship between the users 211.

[0030] The relationships in the interpersonal relationship 211 and the ranking relationships may be defined as appropriate. For example, in the case of a hierarchical relationship in the interpersonal relationship 211, such as parent and child or boss and subordinate, it is presumed that a higher-ranking user U is more likely to dominate the conversation space than a lower-ranking user U. Therefore, in one example, if other factors are not considered, higher-ranking user U will have a higher rank 25, and lower-ranking user U Each user U may be assigned a rank 25 such that the rank 25 decreases as the user's rank decreases. If the relationship 211 is an equal relationship, each user U may be assigned a rank 25 such that the ranks 25 of users U in an equal relationship are equal. Differences in rank 25 may be made even among users U in an equal relationship based on attributes such as age and gender.

[0031] The information processing device 1 may store the results of its analysis (estimation) of the relationships 211 in a database, associated with the identification results of each user U. The database may be stored in the memory resources of the information processing device 1 or in an external storage device. For example, the information processing device 1 may assign a unique identifier to each user U. The information processing device 1 may extract data usable for identifying user U (voiceprint data, facial image data, etc.) from the data of each identified user U contained in the conversation data 20. The information processing device 1 may store the identifier of the target user U, the data usable for identification, and the analysis results of the relationships 211 in a database. In subsequent processing, the information processing device 1 may identify each user U by referring to the various data stored in the database, thereby identifying the relationships 211 between each user U. This is expected to speed up the analysis of relationships 211.

[0032] Knowledge level 213 indicates the degree to which knowledge is possessed. Knowledge level 213 may be evaluated as a discrete or continuous value. In a simple example, knowledge level 213 may be evaluated as a binary value (presence or absence of knowledge). Knowledge level 213 may be evaluated under a specific domain, or it may be evaluated without domain restrictions. The domain may be defined by conversational conditions such as the conversational environment or the types of requests 30 that can be accepted. For example, target terms for evaluating knowledge level 213 may be defined in advance. Target terms may include, for example, technical terms, related terms, etc. A list of target terms may be stored in the memory resources of the information processing device 1, or it may be incorporated into a program (program 81 described later). The information processing device 1 may count the number of target terms included in the utterances of each identified user U. The information processing device 1 may identify the knowledge level 213 of each user U according to the number of counted target terms. In another example, the information processing device 1 may provide the large-scale generative model with the utterances and estimation instructions of each user U, thereby causing the large-scale generative model to estimate the amount of knowledge 213 of each user U. When evaluating the amount of knowledge 213 under a specific domain, the estimation instructions provided to the large-scale generative model may include specifying the domain in which the amount of knowledge 213 will be evaluated.

[0033] The relationship between knowledge level 213 and ranking may be defined as appropriate. It is presumed that a user U with a large knowledge level 213 is more likely to dominate the conversation space than a user U with a small knowledge level 213. Therefore, in one example, without considering other factors, the rank 25 of each user U may be assigned such that users with a large knowledge level 213 have a higher rank 25, and users with a small knowledge level 213 have a lower rank 25.

[0034] The speech volume 215 indicates the amount of speech given in a conversation. The speech volume 215 may be evaluated as a discrete or continuous value. For example, the information processing device 1 may determine the speech volume 215 of each identified user U according to at least one of the number of characters in the utterance, the number of words in the utterance, and the duration of the utterance.

[0035] The relationship between the amount of speech (215) and ranking may be defined as appropriate. It is presumed that a user U with a large amount of speech (215) is more likely to dominate the conversation space than a user U with a small amount of speech (215). Therefore, in one example, without considering other factors, the ranking (25) of each user U may be assigned such that users with a large amount of speech (215) have a higher rank (25), and users with a small amount of speech (215) have a lower rank (25).

[0036] Speech characteristics 217 indicate the features of speech. Speech characteristics 217 may include, for example, tone of voice, intonation, speech rate, etc. Tone of voice may include speech style such as the presence or absence of honorifics. As described above, speech Characteristics 217 may also be used in the analysis of interpersonal relationships 211. Interpersonal relationships 211 may be reflected in speech characteristics 217. Therefore, when assigning ranks 25, the results of the analysis of speech characteristics 217 may be used in the same way as for interpersonal relationships 211. For example, if other factors are not considered, the rank 25 of each user U may be assigned such that the user U estimated to be higher in rank based on speech characteristics 217 will have a higher rank 25, and the user U estimated to be lower in rank will have a lower rank 25.

[0037] Based on interpersonal relationships 211, knowledge level 213, speaking volume 215, and speech characteristics 217, the degree to which one dominates the conversational space can be appropriately evaluated. Therefore, according to one example of this embodiment, by using at least one of interpersonal relationships 211, knowledge level 213, speaking volume 215, and speech characteristics 217, it is possible to appropriately rank the degree to which one has the initiative in the conversation.

[0038] The collection of conversation data 20 may or may not overlap with the acceptance of request 30, at least partially. That is, conversation data 20 may or may not overlap with the utterance data 300, which includes request 30, at least partially. Furthermore, while accepting request 30, the information processing device 1 may continue to collect conversation data 20 and continuously update the ranking 25 of those who have the initiative in the conversation in real time. In one example, the result of assigning the ranking 25 may be configured to directly show the ranking 25, or it may be configured to indirectly show the ranking 25 based on the degree to which one has the initiative in the conversation, etc. In a simple example, the result of assigning the ranking 25 may be configured to show whether or not one has the initiative as a binary value.

[0039] (request) The types of requests 30 are not particularly limited and may be appropriately selected depending on the embodiment. Requests 30 may include, for example, information retrieval (including questions), task management, machine control, application control, etc. Information retrieval may include questions. Task management may include setting alerts or timers, registering or confirming schedules, creating to-do lists, etc. The target of machine control may be arbitrarily selected. For example, when a request 30 is received inside a vehicle MV, machine control may include controlling the vehicle MV, such as steering the steering wheel, operating the accelerator, operating the brakes, operating the turn signals, operating the wipers, operating the air conditioning, turning the lights on / off, and communication processing. Application control may include, for example, playing / stopping content, starting / stopping a target application, etc. Application control may include, for example, adjusting the volume, adjusting the screen brightness, etc. For example, in a scenario where a request 30 is received inside a vehicle MV, the vehicle MV may be equipped with a navigation device. The information processing device 1 may be a navigation device or may be connected to a navigation device. In this case, application control may include controlling the navigation device, such as setting a destination, providing route guidance, checking travel time, and searching for nearby facilities.

[0040] (same period) Accepting requests 30 at the same time means accepting requests 30 simultaneously, or at times when they may conflict. For example, the scope of "simultaneous" can be defined by the time range in which the execution of requests 30 received from two or more users U may overlap.

[0041] (Priority) Priority 50 determines the order in which two or more requests 30 received at the same time are carried out. Requests 30 with a higher priority 50 are carried out first. The priority 50 of a request 30 may be determined as appropriate based on the result of assigning a ranking 25 to determine who has the initiative in the conversation. For example, the higher the ranking 25 (i.e., the greater the degree of initiative in the conversation) of user U, the higher the priority 50 of that user U's request 30 will be, and the lower the ranking 25 (i.e., the lower the user U's request 30 will be, the lower the priority 50 of that user U's request 30 will be. The priority 50 may be determined solely from the ranking 25, or it may be determined by further referring to other information in addition to the ranking 25.

[0042] Figure 3 schematically shows an example of a method for determining priority 50 according to this embodiment. In one example, the information processing device 1 may determine the urgency level 40 of the received request 30. Determining the priority 50 of the received request 30 based on the result of assigning rank 25 may be configured by determining the priority 50 of the received request 30 based on the result of assigning rank 25 and the result of determining the urgency level 40. Urgency level 40 is an example of other information.

[0043] The urgency level of 40 may be determined appropriately from the utterance of request 30. A computational model such as a rule-based model or a trained machine learning model may be used to determine the urgency level of 40. The trained machine learning model may include a large-scale generative model. For example, the urgency level of request 30 may be determined depending on whether or not specific terms are included in the utterance of request 30. If the utterance of request 30 includes terms related to urgency, such as words indicating urgency (e.g., "prioritize") or words indicating highly urgent processing, the urgency level of request 30 may be determined to be high. On the other hand, if the utterance of request 30 does not include terms related to urgency, the urgency level may be determined to be low. Highly urgent processing may be set appropriately depending on the embodiment.

[0044] Furthermore, rank 25 and urgency 40 may be appropriately reflected in the determination of priority 50. For example, information processing device 1 may reflect rank 25 and urgency 40 in the determination of priority 50 without assigning any preference. In another example, information processing device 1 may prioritize either rank 25 or urgency 40 and reflect the prioritized indicator in the determination of priority 50. For example, information processing device 1 may determine the priority 50 of request 30 such that the higher the urgency 40, the higher the priority 50, and the lower the urgency 40, the lower the priority 50. If the urgency 40 is the same, information processing device 1 may determine the priority 50 of request 30 such that requests from user U with a higher rank 25 have a higher priority 50, and requests from user U with a lower rank 25 have a lower priority 50. According to one example of this embodiment, by further reflecting the urgency 40 of the request 30 in the determination of the priority 50, in addition to the rank 25 that has the initiative in the conversation, it is possible to expect that requests 30 received from two or more users U at the same time will be carried out in a more appropriate order.

[0045] [2 Example Configurations] Figure 4 schematically shows an example of the hardware configuration of the information processing device 1 according to this embodiment. In one example, the information processing device 1 may be configured as a computer in which a control unit 11, a storage unit 12, an external interface 13, an input device 14, and an output device 15 are electrically connected.

[0046] The control unit 11 is configured to perform information processing based on the program and various data. For example, the control unit 11 includes a hardware processor such as a CPU (Central Processing Unit), RAM (Random Access Memory), and ROM (Read Only Memory). Good. The control unit 11 (CPU) is an example of a processor resource.

[0047] The storage unit 12 is configured to hold arbitrary data. For example, the storage unit 12 may include a hard disk drive, a solid-state drive, semiconductor memory, etc. The storage unit (and RAM, ROM) is an example of a memory resource. In one example, the storage unit 12 may store various information such as a program 81. Program 81 is a program that causes the information processing device 1 to execute information processing related to the fulfillment of request 30 by agent A0 (Figures 5 and 6 described later). Program 81 includes a series of instructions for said information processing.

[0048] In one example, program 81 may be stored in a storage medium 91 instead of or together with the storage unit 12. The storage medium 91 is configured to store various types of information (stored programs, etc.) by electrical, magnetic, optical, mechanical, or chemical means so that a machine such as a computer can read the information. The storage unit 12 and the storage medium 91 are examples of non-temporary storage media. The information processing device 1 retrieves program 81 from the storage medium 91. It is advantageous. The storage medium 91 may be a disk-type storage medium (CD, DVD, etc.) or a non-disk-type storage medium such as semiconductor memory (flash memory, etc.). Any drive device may be used to read the information stored in the storage medium 91. The type of drive device may be selected according to the storage medium 91. The drive device may be connected to the information processing device 1 in any way. The storage medium 91 may include an external storage device that can be connected to the information processing device 1.

[0049] If the configuration is adopted in which agent A0 is deployed to the information processing device 1, the data of agent A0 may be stored in memory resources such as the storage unit 12, or it may be incorporated into the program 81. If the configuration is adopted in which an arithmetic model is used for each process, the data of the arithmetic model may be stored in memory resources such as the storage unit 12, or it may be incorporated into the program 81. If the configuration is adopted in which the analysis results of the human relationships 211 are stored as past history in a database deployed to the information processing device 1, that database may be stored in memory resources such as the storage unit 12.

[0050] The external interface 13 is configured to connect to an external device by wire or wireless connection. The external interface 13 may be, for example, a USB (Universal Serial Bus) port, a dedicated port, a communication port, etc. The type and number of external interfaces 13 may be determined as appropriate depending on the embodiment. If the external interface 13 includes a communication port, the communication network standard may be arbitrarily selected. The information processing device 1 may be connected to at least one of the microphone MA and the imaging device CA via the external interface 13.

[0051] The input device 14 is configured to accept information input. The input device 14 may consist of, for example, a mouse, keyboard, touch panel, touchpad, or control. The output device 15 is configured to output information. The output device 15 may consist of, for example, a display or speaker. The information processing device 1 may be operated using the input device 14 and the output device 15. The input device 14 and the output device 15 may be directly connected to the information processing device 1, or they may be indirectly connected via the external interface 13. The input device 14 and the output device 15 may be integrated at least in part by a touch panel display or the like.

[0052] Regarding the specific hardware configuration of the information processing device 1, components can be omitted, replaced, and added as appropriate depending on the embodiment. For example, the control unit 11 may include multiple hardware processors. Hardware processors include microprocessors, FPGAs (Field-Programmable Gate Arrays), DSPs (Digital Signal Processors), and ECs. The information processing device 1 may consist of a U (Electronic Control Unit), a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), etc. At least one of the external interface 13, input device 14, and output device 15 may be omitted. The program 81 may be stored in an external storage device such as a NAS (Network Attached Storage). An external storage device is also an example of a non-temporary storage medium. The information processing device 1 may consist of multiple computers. In this case, the hardware configuration of each computer may or may not be the same. The information processing device 1 may consist of a computer designed specifically for the services provided, as well as a general-purpose server device, a general-purpose PC (Personal Computer). ), notebook PCs, terminal devices, etc. Terminal devices may include user terminals such as mobile phones (smartphones, etc.) and tablet devices. In the context of vehicle MVs, terminal devices may include in-vehicle terminals.

[0053] [3 Examples of operation] Figure 5 is a flowchart showing an example of the processing procedure for ranking by the information processing device 1 according to this embodiment. Figure 6 shows the processing procedure for fulfilling a request by the information processing device 1 according to this embodiment. This flowchart shows an example of the processing procedure. The control unit 11 of the information processing device 1 executes instructions contained in the program 81 held in the storage unit 12. As a result, the information processing device 1 operates as a computer capable of performing the following information processing. The control unit 11 is configured to perform the following information processing. The following processing procedure is an example of an information processing method performed by a computer. However, the following processing procedure is merely an example, and each step may be modified as much as possible.

[0054] (Ranking) As shown in Figure 5, in step S101, the control unit 11 acquires conversation data 20, which is a record of a conversation between multiple users U. In one example, the conversation data 20 may consist of audio data 201 and image data 203. Thus, the conversation record may include both audio and images. Once the conversation data 20 is acquired, the control unit 11 proceeds to the next step S102.

[0055] In step S102, the control unit 11 analyzes the acquired conversation data 20 to assign a ranking 25 to the users U who have the upper hand in the conversation. For example, the control unit 11 may analyze at least one of the following: the relationships between the users U 211, the amount of knowledge each of the users U 213, the amount of speech each of the users U 215, and the speech characteristics each of the users U 217, as shown in the conversation data 20. Then, the control unit 11 may assign a ranking 25 to the users U who have the upper hand in the conversation, according to at least one of the analyzed relationships 211, amount of knowledge 213, amount of speech 215, and speech characteristics 217. Once the ranking 25 is assigned, the control unit 11 proceeds to the next step S103.

[0056] In step S103, the control unit 11 outputs information regarding the results of assigning rank 25 to each user U. The output destination and the content of the information to be output may be appropriately selected depending on the embodiment. In one example, the control unit 11 may store the results of assigning rank 25 to each user U in the storage unit 12 or RAM for use in the process of determining the priority 50. Once the output process is complete, the control unit 11 terminates the ranking processing procedure related to this example of operation.

[0057] (Fulfillment of request) As shown in Figure 6, in step S301, the control unit 11 receives a request 30 from user U via speech. In step S302, the control unit 11 determines whether it received requests 30 from at least two or more users U at the same time. If it received requests 30 from two or more users U at the same time, the control unit 11 proceeds to the next step S303. On the other hand, if it did not receive requests 30 from two or more users U at the same time (i.e., it received a request 30 from only one user U), the control unit 11 proceeds to step S311.

[0058] In step S303, the control unit 11 determines the urgency level 40 of the received request 30. In step S304, the control unit 11 determines the priority level 50 of the received request 30 based on the result of assigning rank 25 and the result of determining the urgency level 40. In step S305, the control unit 11 has agent A0 carry out the request 30 according to the determined priority level 50. That is, the control unit 11 has agent A0 carry out the requests 30 in order of highest priority level 50. Once the execution of the requests 30 from each user U is complete, the processing procedure related to request execution in this example operation is terminated.

[0059] In step S311, the control unit 11 instructs agent A0 to carry out the received request 30. Once the execution of the received request 30 is completed, the processing procedure related to request execution in this example is terminated. In response to receiving the request 30 from user U in the processing of step S301, the control unit 11 processes steps S303 to S305 or step S311. This can be executed in real time.

[0060] (Features) According to this embodiment, the processing in steps S101 and S102 allows for assigning a ranking 25 to determine which user U has the upper hand in the conversation. In step S304, by utilizing the result of assigning the ranking 25, it is possible to appropriately determine the priority 50 of the requests 30 received from each user U when requests 30 are received from two or more users U at the same time. As a result, in step S305, it is possible to expect the requests 30 received from two or more users U at the same time to be executed in an appropriate order.

[0061] [4. Variant] While embodiments of this disclosure have been described in detail above, the above description is merely illustrative in all respects. The processes and means described in this disclosure can be freely combined and implemented as long as no technical inconsistencies arise. Various improvements or modifications may be made to the above embodiments as appropriate.

[0062] Furthermore, with respect to each processing step of the information processing device 1 described above, at least one of the following may be omitted, replaced, or added, depending on the embodiment. For example, the processing in step S303 may be omitted. Accordingly, in step S304, the control unit 11 may determine the priority 50 of the received request 30 based on the result of assigning rank 25. [Explanation of symbols]

[0063] 1... Information processing device, 11... Control unit, 20...Conversation data, 25...Ranking 211...Human relationships, 213...Amount of knowledge, 215...Volume of speech, 217...Speech characteristics, 30...Request, 300...Speech data, 40...Urgency, 50...Priority, A0...Agent

Claims

1. To obtain conversation data, which is a record of conversations that took place between multiple users. By analyzing the acquired conversation data, a ranking is determined of the users who have the upper hand in the conversation. When requests are received simultaneously from at least two of the aforementioned multiple users, the priority of the received requests is determined based on the results of the ranking. In accordance with the determined priorities, the agent shall carry out the aforementioned request. A control unit configured to perform the following actions: Information processing device.

2. Analyzing the conversation data includes analyzing at least one of the relationships between the multiple users, the amount of knowledge each of the multiple users possesses, the amount of speech each of the multiple users provides, and the speech characteristics of each of the multiple users, as expressed in the conversation data. The ranking of who takes the lead in the conversation among the multiple users is determined according to at least one of the analyzed relationships, knowledge levels, speaking volume, and speech characteristics. The information processing apparatus according to claim 1.

3. The record of the aforementioned conversation includes audio and images. The information processing apparatus according to claim 1.

4. The control unit is configured to further perform the task of determining the urgency of the received request. Determining the priority of the received requests based on the results of the ranking constitutes determining the priority of the received requests based on the results of the ranking and the determination of urgency. The information processing apparatus according to claim 1.

5. A program that causes a computer to execute an information processing method, The aforementioned information processing method is To obtain conversation data, which is a record of conversations that took place between multiple users. By analyzing the acquired conversation data, a ranking is determined of the users who have the upper hand in the conversation. If requests are received simultaneously from at least two of the aforementioned multiple users, the priority of the received requests is determined based on the ranking results. In accordance with the determined priorities, the agent shall carry out the aforementioned request. including, program.

Images