Information processing apparatus and program

Abstract

To enhance empathy and trust from a real user while moving the conversation between a virtual user and the real user forward.SOLUTION: An information processing apparatus 1 is configured to: in making a request to a speech sentence generation apparatus (generative AI) 2 to generate a speech sentence of digital staff that has a conversation with a customer, transmit the changes of empathy score and trust score in addition to a conversation history, the empathy score indicating how much the customer empathizes with a speech of the digital staff, the trust score indicating how much the customer trusts the speech of the digital staff; calculating the empathy score when the speech of the digital staff is not in line with a topic, or calculating the trust score when the speech of the digital staff is in line with the topic; and increasing the sympathy score or the trust score when the customer agrees with the speech of the digital staff, or decreasing the empathy score or the trust score when the customer disagrees with the speech of the digital staff.SELECTED DRAWING: Figure 4

Description

Technical Field

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

Background Art

[0002] Currently, EC sites (online sales) such as apparel have become widespread, and it is assumed that in the future, there will be an increasing number of opportunities to open virtual stores in the shopping malls of virtual spaces (metaverse) on the Internet. In such a situation, in EC sites as well as in physical stores, it is highly regarded from the perspective of sales promotion to provide one-on-one customer service in which staff introduce recommended products through conversations with customers, dig out customer needs and materialize products desired by users, and respond to various inquiries.

[0003] Generally, the number of users visiting EC sites is much larger than the number of customers visiting physical stores. Therefore, it is not realistic for actual staff to serve each user visiting the EC site individually. Thus, it is assumed that virtual staff (referred to as virtual users or digital staff) is constructed on a computer, and instead of actual staff, digital staff serves customers (referred to as actual users) one-on-one.

[0004] The use of generative AI is expected for creating the speech content for digital staff to have conversations with customers via chat or the like. The conversation history between the digital staff and the customer is sent to the generative AI together with the generation request. The generative AI analyzes the conversation history and creates the speech sentence that the digital staff should say next. Since it is used in EC sites, the ultimate goal is sales, and for this purpose, high conversation skills and customer service capabilities like those of excellent actual staff are required. Excellent actual staff is good at advancing conversations to build an intimate relationship while gaining empathy and trust from customers, whether consciously or unconsciously.

[0005] However, it was not possible to break away from mechanical conversations by generating utterances based solely on the conversation history between the input digital staff and the customer.

Summary of the Invention

Problems to be Solved by the Invention

[0006] The purpose is to enhance the empathy and trust from the real user while the virtual user progresses the conversation with the real user.

[0007] The purpose is to build an intimate relationship with the real user while the virtual user progresses the conversation with the real user.

Means for Solving the Problems

[0008] The information processing apparatus according to the present embodiment includes a memory for storing a program, a processor for executing the program, and a communication interface for communicating with a user terminal and a speech generation apparatus (generative AI) via an electric public communication network, and is an information processing apparatus for a virtual user to have a conversation with a real user.

[0009] By executing the program, the processor functions as a first receiving means for receiving data of the real user's utterance from the user terminal, a first transmitting means for transmitting a generation request for the virtual user's utterance to the speech generation apparatus together with data of the utterance history of each of the real user and the virtual user, a second receiving means for receiving data of the virtual user's utterance from the speech generation apparatus, a second transmitting means for transmitting the received data of the virtual user's utterance to the user terminal, a means for determining whether the real user is positive or negative about the virtual user's utterance based on the real user's utterance, and an estimation means for calculating at least one of a empathy score representing the degree of empathy of the real user for the virtual user and a trust score representing the degree of trust of the real user for the virtual user based on the result of the determination.

[0010] The first transmission means transmits at least one of the transition of the calculated empathy score and the reliability score to the utterance generation device in addition to the generation request of the virtual user's utterance and the history of the utterances of the real user and the virtual user respectively.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Mode for Carrying Out the Invention

[0012] Hereinafter, an information processing apparatus according to an embodiment of the present invention will be described with reference to the drawings. The information processing apparatus according to the present embodiment is applied to a scenario where a virtual user (virtual user) constructed on a computer converses with an actual user. Here, a scenario will be described as an example in which a customer who has accessed an EC site converses with a virtual staff (digital staff) constructed on a computer in a chat room prepared on the EC site. In this example, the customer corresponds to an actual user, and the digital staff corresponds to a virtual user. Of course, the scenario where the actual user corresponds to the customer and the virtual user corresponds to the digital staff is not limited.

[0013] As shown in FIG. 1, the information processing apparatus 1 according to the present embodiment is connected to a speech generation apparatus 2 that functions as a generation AI, an EC server 3, and a user terminal 4 via an electrical public communication network 5 such as an Internet line. The user terminal 4 is connected to the EC server 3, and a customer (actual user) accesses the EC site operated by the EC server 3. A chat room or the like is prepared on the EC site, and the customer purchases a product while conversing with the digital staff (virtual user). The information processing apparatus 1 directly receives the speech data of the customer input from the user terminal 4 from the user terminal 4 or indirectly via the EC server 3, creates a speech of the digital staff in response to the customer's speech, and directly transmits it to the user terminal 4 or indirectly via the EC server 3 to the user terminal 4. By repeating the said process, the conversation between the customer and the digital staff progresses on the chat room. The information processing apparatus 1 transmits a speech generation request to the speech generation apparatus 2 together with the ordered customer's speech and the digital staff's speech, that is, the conversation history between the customer and the digital staff.

[0014] In this embodiment, based on the customer's utterance, it is determined whether the customer is positive or negative about the digital staff's utterance, and based on the determination result, a sympathy score is calculated to estimate the degree to which the customer sympathizes with the digital staff's utterance, and a reliability score is calculated to estimate the degree to which the customer shows a sense of trust in the digital staff's utterance. Whether to apply the determination result of positive or negative regarding the customer's utterance to either the sympathy score or the reliability score is selected according to whether the digital staff's utterance follows the conversation topic. For example, if the e-commerce site is a clothing store, clothing is applied as the conversation topic.

[0015] When it is determined that the customer is positive about the digital staff's utterance, the reliability score or the sympathy score is increased, and when it is determined that the customer is negative, the reliability score or the sympathy score is decreased. Of course, when the reliability score, the sympathy score, and the intimacy score (to be described later) increase, it means that the customer's trust, sympathy, and intimacy with the digital staff increase.

[0016] By repeating this process, it is possible to determine how the sympathy score and the reliability score change as the conversation progresses, and further to identify what kind of utterance content of the digital staff caused this change.

[0017] Furthermore, in an actual human relationship, as the intimacy increases, the politeness of speech gradually decreases, and there is a tendency to gradually change from honorific language to friendly language. It is considered that the intimacy can be deepened by reducing the politeness of speech and changing from honorific language to friendly language. In other words, if the customer uses friendly language while the digital staff continues to use honorific language, the customer may lose the sense of intimacy.

[0018] Therefore, in the present embodiment, the politeness level of the digital staff's speech generated by the speech generation device 2 is changed according to the situation so that the customer follows the change in the politeness level of the words uttered by the customer. The politeness level of the words in the customer's speech is specified, and based on the change in the politeness level of the words in the specified customer's speech, the politeness level of the digital staff's words is determined. When the politeness level of the words in the customer's speech decreases, the politeness level of the digital staff's speech also decreases by one level from the current politeness level accordingly. When the politeness level of the words in the customer's speech increases, the politeness level of the digital staff's speech also increases by one level from the current politeness level. Thus, in the present embodiment, by changing the politeness level of the digital staff's words to match the politeness level of the customer's words, it becomes possible to increase the customer's sense of intimacy while suppressing a decrease in it.

[0019] Also, in the present embodiment, the customer's intimacy score for the digital staff is calculated based on the change in the politeness level of the words in the customer's speech. When the politeness level of the words in the customer's speech decreases, the customer's intimacy score for the digital staff is increased. When the politeness level of the words in the customer's speech increases, the customer's intimacy score for the digital staff is decreased.

[0020] When transmitting a speech generation request to the speech generation device 2, in addition to the conversation history between the customer and the digital staff, by transmitting the transitions of the empathy score, the reliability score, and the intimacy score respectively, the speech of the digital staff can be generated to follow the flow of the conversation and increase the empathy score and the reliability score without decreasing them.

[0021] As shown in FIG. 2, the information processing apparatus 1 as a conversation apparatus has a RAM 12, a ROM 13, a storage unit 14, an input device 15, a display 16, and a communication interface 17 connected to a processor 11 via a system bus 10. The processor 11 is composed of, for example, a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor 11 executes a conversation program loaded from the storage unit 14 and the ROM 13 into the RAM 12 to execute conversation processing between a customer and a digital staff member.

[0022] The RAM 12 functions as a main memory, a work area, etc. of the processor 11. The ROM 13 or the storage unit 14 stores a BIOS (Basic Input Output System), an operating system program (OS), a conversation processing program according to the present embodiment, programs for realizing other various functions, and various data required for those processes, which are executed by the processor 11.

[0023] The input device 15 consists of a keyboard (KB), a pointing device such as a mouse or a touch panel, etc. The display 19 is typically realized by an LCD (Liquid Crystal Display). In addition to the conversation processing program, the storage unit 14 stores data related to the conversation history between the customer and the digital staff member, the persona information and persona extension information of the customer illustrated in FIG. 9, etc.

[0024] As shown in FIG. 3, by executing the conversation processing program, the processor 11 functions as a control unit 21, a natural language analysis processing unit 22, a sensitivity estimation processing unit 23, a reliability estimation processing unit 24, an intimacy estimation processing unit 25, a purchase possibility estimation processing unit 26, a persona information update unit 27, a persona extension information update unit 28, a word usage conversion processing unit 29, a speech generation request processing unit 30, a reception processing unit 31, and a transmission processing unit 32.

[0025] The data of the customer's speech input from the user terminal 4 is received by the reception processing unit 31 via the EC server 3. Note that the data of the customer's speech input from the user terminal 4 may be directly received by the reception processing unit 31 from the user terminal 4 without passing through the EC server 3. The natural language analysis processing unit 22 performs natural language analysis processing on the data of the customer's speech received via the reception processing unit 31, decomposes it into parts of speech, extracts the string of nouns in particular, and performs syntactic analysis. Also, the natural language analysis processing unit 22 performs natural language analysis processing on the speech of the digital staff transmitted to the EC server 3 via the transmission processing unit 32, decomposes it into parts of speech, extracts the string of nouns, and performs syntactic analysis. The parts of speech decomposed, the nouns extracted, and the syntactic analysis results by the natural language analysis processing unit 22 are used in the respective processes of the empathy degree estimation processing unit 23, the reliability estimation processing unit 24, the persona information update unit 27, and the persona extended information update unit 28.

[0026] When the content of the digital staff's speech does not follow the theme of the conversation, the empathy degree estimation processing unit 23 executes the calculation of the empathy degree score. When the content of the digital staff's speech follows the theme of the conversation, the empathy degree estimation processing unit 23 does not execute the calculation of the empathy degree score. The theme of the conversation is, for example, the apparel field if the EC site is an apparel store. Specifically, if the nouns included in the digital staff's speech are related to apparel, it is determined that the content of the digital staff's speech follows the theme of the conversation. If the nouns included in the digital staff's speech are not related to apparel and are irrelevant to apparel, it is determined that the content of the digital staff's speech does not follow the theme of the conversation. This determination process may create an apparel-related term dictionary in advance and store it in the storage unit 14, and check whether the nouns included in the digital staff's speech match the dictionary and are listed in the apparel-related term dictionary, or it may be estimated using a learned model in which the suitability of the nouns and the conversation theme is machine-learned as teacher data.

[0027] When the customer is positive about the digital staff's statement, a predetermined value is added to the current value of the empathy score to increase the empathy score. When the customer is negative about the digital staff's statement, a predetermined value is subtracted from the current value of the empathy score to decrease the empathy score. Typically, the addition value and the subtraction value are the same value, but the absolute value of the addition value may be higher or lower than the absolute value of the subtraction value. The determination process for identifying whether the customer's statement is positive or negative with respect to the digital staff's statement may be to create a dictionary in advance that covers a plurality of positive expressions and a plurality of negative expressions and store it in the storage unit 14, and determine whether the customer's statement includes any positive expression or negative expression, or it may be estimated using a learned model that has been machine-learned with a plurality of expressions and the possibility of being positive or negative as teacher data.

[0028] When the content of the digital staff's statement is along the theme of the conversation, the reliability estimation processing unit 24 executes the calculation of the reliability score. When the content of the digital staff's statement is not along the theme of the conversation, the reliability score is not calculated. When the customer makes a positive statement about the digital staff's statement, a predetermined value is added to the current value of the reliability score to increase the reliability score. When the customer sends a negative statement about the digital staff's statement, a predetermined value is subtracted from the current value of the reliability score to decrease the reliability score. Typically, the addition value and the subtraction value are the same value, but the absolute value of the addition value may be higher or lower than the absolute value of the subtraction value.

[0029] The intimacy estimation processing unit 25 compares the diction of the customer's speech with that of the digital staff's speech, and calculates an intimacy score representing the degree of intimacy of the customer towards the digital staff based on the comparison result. Specifically, when the politeness level of the diction of the customer's speech decreases, a predetermined value is added to the current value of the intimacy score to increase the intimacy score. When the politeness level of the diction of the customer's speech increases, a predetermined value is subtracted from the current value of the intimacy score to decrease the intimacy score. Typically, the added value and the subtracted value are the same, but the absolute value of the added value may be higher or lower than the absolute value of the subtracted value.

[0030] Typically, for the determination of the politeness level of the diction, a dictionary regarding a plurality of inflectional expressions corresponding to each of a plurality of levels with different politeness levels from honorific language to friendly language is created in advance and stored in the storage unit 14. The inflectional expression of the customer's speech is collated with the dictionary, and the politeness level of the corresponding inflectional expression is specified to perform the determination.

[0031] It is assumed that when the empathy score, reliability score, and intimacy score increase, the likelihood of the customer purchasing the product increases. Therefore, the purchase likelihood estimation processing unit 26 calculates a purchase likelihood score (comprehensive score) representing the likelihood of the customer purchasing the product based on the calculated empathy score, reliability score, and intimacy score. For example, the purchase likelihood score is calculated by adding to the current value of the purchase likelihood score the change amount of each of the empathy score, reliability score, and intimacy score, or a value obtained by multiplying the change amount by a predetermined coefficient such as 0.5. Note that the purchase likelihood score may be calculated by simply adding the empathy score, reliability score, and intimacy score, or by weighted adding the empathy score, reliability score, and intimacy score. The calculation method can be arbitrarily changed.

[0032] The persona information update unit 27 generates new persona information regarding the place of origin, hobbies, etc. based on the part-of-speech and syntactic analysis results decomposed by the natural language analysis processing unit 22, and updates the existing persona information regarding the customer by adding it thereto. The persona information is initially the user information registered on the EC site. As the conversation progresses, new persona information is generated and added. The persona information is stored in the storage unit 14.

[0033] The persona extension information update unit 28 generates persona extension information associating the content of the digital staff's conversation, which is considered to have induced the change, with the change in the empathy score and the reliability score. Here, the persona extension information is generated as information that can identify whether the nouns in the digital staff's utterance lead to an increase or decrease in the empathy score and the reliability score. For example, as persona extension information related to a certain customer, an increase in the reliability score is associated with "blue" and "jacket" and generated. The persona extension information is appropriately generated as the conversation progresses and accumulated in the storage unit 14.

[0034] The diction conversion processing unit 29 converts the diction of the digital staff's utterance sentence received from the utterance sentence generation device 2 into the diction of the set politeness level. The digital staff's utterance sentence received from the utterance sentence generation device 2 is initially often mechanical, business-like, and dull. As it is, the intimacy level is low. Therefore, a dictionary for converting into multiple levels of diction with different politeness levels from honorific language to friendly language is created in advance and stored in the storage unit 14. The diction conversion processing unit 29 converts the words and phrases of the digital staff's utterance sentence received from the utterance sentence generation device 2 into the set diction using the said dictionary.

[0035] As will be described later, the politeness level of the customer's language is identified. When the politeness level changes during the conversation process, the politeness level of the language used by the digital staff is changed following the change. For example, when the politeness level of the customer's language decreases, that is, when the intimacy of the customer towards the digital staff increases, the politeness level of the language regarding the digital staff's speech is decreased by one level from the current politeness level. When the politeness level of the customer's language increases, that is, when the intimacy of the customer towards the digital staff decreases, the politeness level of the language regarding the digital staff's speech is increased by one level from the current politeness level. By changing the politeness level of the language used by the digital staff following the change in the politeness level of the customer's language in this way, the intimacy of the customer towards the digital staff can be increased, and the decrease in the intimacy of the customer towards the digital staff can be suppressed.

[0036] The utterance generation request processing unit 30 transmits the data of the conversation history between the customer and the digital staff including the customer's previous utterance to the utterance generation device 2 together with the request for generating the utterance of the digital staff in response to the customer's utterance. Further, in addition to the conversation history data, the utterance generation request processing unit 30 transmits the persona information, persona extension information, the transition of the empathy score, the transition of the reliability score, the transition of the intimacy score, and the transition of the purchase possibility score to the utterance generation device 2. By referring to these pieces of information, the utterance generation device 2 can generate the utterance of the digital staff so as to follow the flow of the conversation and, moreover, without decreasing the empathy score and the reliability score and being able to increase them.

[0037] Figure 4 shows the flow of the conversation process of the information processing apparatus 1. Along with the start of the conversation, the empathy estimation processing unit 23, the reliability estimation processing unit 24, the intimacy estimation processing unit 25, and the purchase possibility estimation processing unit 26 initialize their respective scores under the control of the control unit 21 (S1). For example, the empathy score, the reliability score, and the intimacy score are initialized to zero values, and the purchase possibility score is initialized to 50 as the median value. When the conversation is repeated with the same customer and digital staff, the values of the empathy score, the reliability score, the intimacy score, and the purchase possibility score at the end of the previous conversation may be carried over.

[0038] Data of the customer's speech is received from the EC server 3 via the reception processing unit 31 (S2). Data of the digital staff's speech is read from the storage unit 14. The customer's speech and the digital staff's speech are subjected to natural language analysis processing (S3). Part-of-speech analysis is performed, nouns are extracted, and syntactic analysis is executed.

[0039] Using the natural language analysis result, the empathy estimation processing unit 23 executes empathy estimation processing to calculate the empathy score (S4), the reliability estimation processing unit 24 executes reliability estimation processing to calculate the reliability score (S5), and the intimacy estimation processing unit 25 executes intimacy estimation processing to calculate the intimacy score (S6). Each process will be described later. Also, the purchase possibility estimation processing unit 26 calculates a purchase possibility score representing the possibility that the customer purchases a product based on the empathy score, the reliability score, and the intimacy score (S7). The method of calculating the purchase possibility score representing the possibility that the customer purchases a product on the EC site from the empathy score, the reliability score, and the intimacy score is arbitrary. For example, the purchase possibility score is calculated by adding to the current value of the purchase possibility score a value obtained by multiplying the change amount of each of the empathy score, the reliability score, and the intimacy score or the change amount by a predetermined coefficient such as 0.5.

[0040] Next, the persona extension information update unit 28 generates persona extension information associating nouns in the digital staff's utterance made immediately before the customer's utterance indicating the change with the changes in the empathy score, reliability score, and intimacy score respectively (S8). Thereby, when any of the customer's empathy, reliability, and intimacy changes, the nouns in the digital staff's utterance that caused the change can be recognized. For example, when the customer's reliability score increases, it can be determined that the increase is caused by "blue" and "jacket", in other words, the customer is interested in "blue" and "jacket", or the customer's preferences can be identified.

[0041] In the next step S9, the utterance generation request processing unit 30 generates an utterance generation request together with the conversation history, persona information, persona extension information, the transition of the empathy score, the transition of the reliability score, the transition of the intimacy score, and the transition of the purchase probability score. The transmission processing unit 32 transmits the utterance generation request together with the data of the conversation history, persona information, persona extension information, the transition of the empathy score, the transition of the reliability score, the transition of the intimacy score, and the transition of the purchase probability score to the utterance generation device 2 via the communication interface 17 (S9).

[0042] The reception processing unit 31 receives the data of the digital staff's utterance from the utterance generation device 2 via the communication interface 17 (S10). The diction conversion processing unit 29 converts the diction of the digital staff's utterance into the set polite diction (S11). The transmission processing unit 32 transmits the data of the digital staff's utterance with the converted diction to the EC server 3 via the communication interface 17 (S12). When the conversation is continuing (S13, NO), it returns to step S2 and waits for the next customer's utterance. When the conversation has ended (S13, YES), the conversation processing ends.

[0043] FIG. 5 shows the flow of the empathy estimation process in step S4 of FIG. 4. Refer to FIGS. 10 and 11 for an example of the transitions of the empathy score, reliability score, intimacy score, purchase probability score, and the change in diction.

[0044] Based on the analysis result of the natural language analysis processing unit 22, the empathy degree estimation processing unit 23 determines whether the content of the digital staff's speech follows the theme of the conversation (S21). For example, if the EC site is a clothing store, the theme of the conversation is the clothing field. Typically, the nouns included in the digital staff's speech are collated with a clothing-related term dictionary to check whether they are listed in the clothing-related term dictionary. When the nouns included in the digital staff's speech match any of the terms listed in the clothing-related term dictionary, it is determined that the content of the digital staff's speech follows the theme of the conversation (S21, YES). When the nouns included in the digital staff's speech do not match any of the terms listed in the clothing-related term dictionary, it is determined that the content of the digital staff's speech does not follow the theme of the conversation (S21, NO). When it is determined that the content of the digital staff's speech follows the theme of the conversation (S21, YES), the system waits for the next speech of the digital staff. When it is determined that the content of the digital staff's speech does not follow the theme of the conversation (S21, NO), the system proceeds to the next step S22 and starts the operation process of the empathy score. Note that when it is determined that the digital staff's speech follows the theme of the conversation, that is, when the digital staff's speech is within the scope of the theme of the conversation, the empathy score is not calculated, but the reliability score is calculated. In other words, when the digital staff's speech is casual conversation, the empathy score is calculated, and when the digital staff's speech is not casual conversation and is in line with the theme to be consulted originally, the reliability score is calculated.

[0045] In step S22, the natural language analysis processing unit 22 analyzes the customer's speech. In step S23, the empathy level estimation processing unit 23 distinguishes whether the customer's speech is affirmative, negative, or neither with respect to the digital staff's speech based on the analysis result of the natural language analysis processing unit 22. Typically, by collating the predicates included in the customer's speech with a dictionary that covers a plurality of pre-prepared affirmative expressions and a plurality of negative expressions, and checking whether they are listed in the dictionary, it is distinguished whether the customer's speech is affirmative, negative, or neither.

[0046] When the customer's speech is affirmative, the empathy level estimation processing unit 23 adds a predetermined value I1 to the current value of the empathy level score to increase the empathy level score (S24). When the customer's speech is negative, the empathy level estimation processing unit 23 subtracts a predetermined value D1 from the current value of the empathy level score to decrease the empathy level score (S25). When the customer's speech is neither affirmative nor negative, assuming that the customer is indifferent to the digital staff's speech, the empathy level estimation processing unit 23 subtracts a predetermined value D2 whose absolute value is larger than the predetermined value D1 from the current value of the empathy level score to further decrease the empathy level score (S26).

[0047] FIG. 6 shows the flow of the reliability estimation process in step S5 of FIG. 4. The reliability estimation processing unit 24 determines whether the content of the digital staff's speech follows the theme of the conversation based on the analysis result of the natural language analysis processing unit 22 (S31). The determination method is the same as that in step S21. When it is determined that the content of the digital staff's speech does not follow the theme of the conversation (S31, YES), it waits for the next digital staff's speech. When it is determined that the content of the digital staff's speech follows the theme of the conversation (S31, NO), it proceeds to the next step S32 and starts the calculation process of the reliability score.

[0048] In step S32, the natural language analysis processing unit 22 analyzes the customer's speech. In step S33, the reliability estimation processing unit 24 distinguishes whether the customer's speech is affirmative, negative, or neither with respect to the digital staff's speech based on the analysis result of the natural language analysis processing unit 22. The method of this distinction is the same as (S1) S23.

[0049] When the customer's speech is affirmative, the reliability estimation processing unit 24 adds a predetermined value I11 to the current value of the reliability score to increase the reliability score (S34). When the customer's speech is negative, the reliability estimation processing unit 24 subtracts a predetermined value D11 from the current value of the reliability score to decrease the reliability score (S35). When the customer's speech is neither affirmative nor negative, assuming the customer is indifferent to the digital staff's speech, the reliability estimation processing unit 24 subtracts a predetermined value D12 whose absolute value is larger than the predetermined value D11 from the current value of the reliability score to further decrease the reliability score (S36).

[0050] Fig. 7 shows the flow of the intimacy estimation processing in step S6 of Fig. 4. The intimacy estimation processing unit 25 identifies the politeness level of the customer's speech (S41). For the determination of the politeness level of the speech, the suffix of the customer's speech is collated with a dictionary regarding a plurality of suffix expressions related to each of a plurality of levels with different politeness levels from honorific language to friendly language, and the politeness level of the suffix expression listed in the dictionary that matches the suffix of the customer's speech is confirmed.

[0051] The intimacy estimation processing unit 25 determines whether the identified politeness level of the customer's speech has changed from the politeness level of the previous customer's speech (S42). When the politeness level has not changed (S42, NO), it returns to step S41 and waits for the next customer's speech. When the politeness level has changed (S42, YES), it is determined whether the identified politeness level of the customer's speech has decreased compared to the politeness level of the previous customer's speech (S43).

[0052] When it is determined that the politeness level of the speech of a specific customer has decreased compared to the politeness level of the speech of the previous customer (S43, YES), a predetermined value I21 is added to the current value of the intimacy score to increase the intimacy score (S44). When it is determined that the politeness level of the speech of the specific customer has become higher than the politeness level of the speech of the previous customer (S43, NO), a predetermined value D21 is subtracted from the current value of the intimacy score to decrease the intimacy score (S45).

[0053] Figure 8 shows the flow of the speech style change process. The speech style change process is executed in parallel with the conversation process shown in Figure 4. The speech style conversion processing unit 29 first initially sets the speech style of the digital staff's speech to "5" as the median of the prepared politeness levels, for example, at 11 levels, prior to the conversation (S51). When receiving the customer's speech, the politeness level of the customer's speech is specified (S52). Then, the speech style conversion processing unit 29 determines whether the politeness level of the specified customer's speech has changed from the politeness level of the previous customer's speech (S53).

[0054] When the politeness level has not changed (S53, NO), the process returns to step S52 and waits for the next customer's speech. When the politeness level has changed (S53, YES), it is determined whether the politeness level of the specified customer's speech has decreased compared to the politeness level of the previous customer's speech (S54).

[0055] When it is determined that the politeness level of the specified customer's speech has decreased compared to the politeness level of the previous customer's speech (S54, YES), that is, when the intimacy of the customer towards the digital staff has increased, the speech style of the digital staff is accordingly changed to a lower politeness level by one step (S55). When it is determined that the politeness level of the specified customer's speech has increased compared to the politeness level of the previous customer's speech (S54, NO), that is, when the intimacy of the customer towards the digital staff has decreased, the speech style of the digital staff is accordingly changed to a higher politeness level by one step (S56).

[0056] When the conversation is ongoing (S57, NO), return to step S52 and wait for the next customer's statement. When the conversation has ended (S57, YES), the word usage change process ends.

[0057] As described above, according to this embodiment, the empathy score, reliability score of the customer with respect to the statement of the digital staff, and the intimacy score of the customer with respect to the digital staff are calculated, and the changes in these empathy scores and reliability scores are transmitted to the statement generation device 2 together with the conversation history and the statement generation request. Thus, the statement generation device 2 can generate statements of the digital staff that enhance the empathy and reliability and suppress the decrease in empathy and reliability.

[0058] Also, according to this embodiment, by changing the politeness level of the digital staff's word usage following the change in the politeness level of the customer's statement, the intimacy can be enhanced and the decrease in intimacy can be suppressed.

[0059] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and their modifications are included in the scope and gist of the invention, as well as in the invention described in the claims and its equivalent scope.

Explanation of Reference Numerals

[0060] 1... Information processing device, 2... Spoken sentence generation device (generative AI), 3... EC server, 4... User terminal, 11... Processor, 10... System bus, 12... RAM, 13... ROM, 14... Storage unit, 15... Input device, 16... Display, 17... Communication interface, 21... Control unit, 22... Natural language analysis processing unit, 23... Empathy estimation processing unit, 24... Reliability estimation processing unit, 25... Intimacy estimation processing unit, 26... Purchase probability estimation processing unit, 27... Persona information update unit, 28... Persona extended information update unit, 29... Diction conversion processing unit, 30... Spoken sentence generation request processing unit, 31... Reception processing unit, 32... Transmission processing unit.

Claims

1. An information processing apparatus for a virtual user to have a conversation with an actual user, comprising: a memory for storing a program, a processor for executing the program, and a communication interface for communicating with a user terminal and a speech generation device (generative AI) via an electrical public communication network, wherein the processor, by executing the program, serves as a first receiving means for receiving data of the speech of the actual user input from the user terminal, serves as a first transmitting means for transmitting a generation request for the speech of the virtual user, together with data of the speech history of each of the actual user and the virtual user, to the speech generation device, serves as a second receiving means for receiving data of the speech of the virtual user from the speech generation device, serves as a second transmitting means for transmitting the received data of the speech of the virtual user to the user terminal, serves as a means for determining whether the actual user is affirmative or negative with respect to the speech of the virtual user based on the speech of the actual user, functions as an estimation means for calculating at least one of a sympathy score representing the degree of sympathy of the actual user for the virtual user and a reliability score representing the degree of trust of the actual user in the virtual user based on the result of the determination, wherein the first transmitting means transmits, in addition to the generation request for the speech of the virtual user and the data of the speech history, data of at least one of the calculated sympathy score and reliability score to the speech generation device. An information processing apparatus.

2. The processor further functions as a means for determining whether the speech of the virtual user follows the theme of the conversation, wherein the estimation means calculates the reliability score when the speech of the virtual user follows the theme of the conversation, and calculates the sympathy score when the speech of the virtual user does not follow the theme of the conversation. The information processing apparatus according to Claim 1.

3. The estimation means increases the reliability score or the sympathy score when it is determined that the actual user is affirmative with respect to the speech of the virtual user, and decreases the reliability score or the sympathy score when it is determined that the actual user is negative with respect to the speech of the virtual user. The information processing apparatus according to Claim 1.

4. The processor further functions as a means for updating the persona information of the actual user from the content of the speech of the actual user, The information processing apparatus according to claim 1, wherein the first transmission means transmits the updated persona information to the utterance generation device in addition to at least one of the generation request, the history of the utterance, and at least one of the calculated empathy score and reliability score.

5. The processor further functions as means for creating persona extension information by associating the content of the virtual user's utterance with a change in the empathy score or the reliability score. The information processing apparatus according to claim 1, wherein the first transmission means transmits the created persona extension information to the utterance generation device in addition to at least one of the generation request of the virtual user's utterance, the history of the utterance, and at least one of the calculated empathy score and reliability score.

6. The processor further functions as a diction conversion means for converting the diction of the virtual user's utterance received from the utterance generation device into other diction with different levels of politeness.

7. The processor further functions as means for specifying the level of politeness of the diction of the real user's utterance. The information processing apparatus according to claim 6, wherein the diction conversion means changes the diction of the virtual user's utterance to a diction with a one-step lower level of politeness when the level of politeness of the diction of the real user's utterance decreases, and changes the diction of the virtual user's utterance to a diction with a one-step higher level of politeness when the level of politeness of the diction of the real user's utterance increases.

8. Means for specifying the level of politeness of the diction of the real user's utterance; The processor further functions as a closeness estimation means for calculating a closeness score representing the degree of closeness of the real user to the virtual user when the level of politeness of the diction of the real user's utterance changes.

9. The information processing apparatus according to claim 8, wherein the closeness estimation means increases the closeness score when the level of politeness of the diction of the real user's utterance decreases, and decreases the closeness score when the level of politeness of the diction of the real user's utterance increases.

10. The processor further functions as means for calculating an overall score from the empathy score, the reliability score, and the closeness score.

11. An information processing apparatus having a memory for storing a program, a processor for executing the program, and a communication interface for communicating with a user terminal and a speech generation device (generation AI) via an electrical public communication network, and for a virtual user to have a conversation with an actual user, By executing the program, the processor Means for receiving data of the speech of the actual user input from the user terminal, Means for transmitting a generation request for the speech of the virtual user to the speech generation device together with data of the speech history of each of the actual user and the virtual user, Means for receiving data of the speech of the virtual user from the speech generation device, Means for converting the diction of the received speech of the virtual user, An information processing apparatus that functions as means for transmitting the converted data of the speech of the virtual user to the user terminal.

12. Means for specifying the politeness level of the diction regarding the speech of the actual user, The processor further functions as means for changing the politeness level of the diction regarding the speech of the virtual user based on the specified politeness level of the diction regarding the speech of the actual user. The information processing apparatus according to claim 11.

13. The means for changing the diction lowers the politeness level of the diction regarding the speech of the virtual user when the politeness level of the diction regarding the speech of the actual user decreases, and raises the politeness level of the diction regarding the speech of the virtual user when the politeness level of the diction regarding the speech of the actual user increases. The information processing apparatus according to claim 12.

14. An information processing apparatus having a memory for storing a program, a processor for executing the program, and a communication interface for communicating with a user terminal and a speech generation device (generation AI) via an electrical public communication network, and for a virtual user to have a conversation with an actual user, By executing the program, the processor Means for receiving data of the speech of the actual user input from the user terminal, Means for transmitting a generation request for the speech of the virtual user to the speech generation device together with data of the speech history of each of the actual user and the virtual user, Means for receiving data of the speech of the virtual user from the speech generation device, means for transmitting the received data of the virtual user's utterance to the user terminal; means for executing a determination as to whether the real user is positive or negative about the virtual user's utterance based on the real user's utterance; means for calculating at least one of a sympathy score representing the degree of sympathy of the real user with respect to the virtual user and a reliability score representing the degree of trust of the real user with respect to the virtual user based on the result of the determination; An information processing apparatus that functions as means for generating, in association with the content of the virtual user's utterance, at least one change in the sympathy score and the reliability score as persona expansion information regarding the real user.

15. A computer having a communication interface for communicating with a user terminal and a speech generation device (generation AI) via an electric public communication network, for a virtual user to have a conversation with a real user, first receiving means for receiving data of the real user's utterance input from the user terminal; first transmission means for transmitting a generation request for the virtual user's utterance to the speech generation device together with data of the utterance history of each of the real user and the virtual user; second receiving means for receiving the virtual user's utterance from the speech generation device; second transmission means for transmitting the received data of the virtual user's utterance to the user terminal; means for executing a determination as to whether the real user is positive or negative about the virtual user's utterance based on the real user's utterance; A program that functions as estimation means for calculating at least one of a sympathy score representing the degree of sympathy of the real user with respect to the virtual user and a reliability score representing the degree of trust of the real user with respect to the virtual user based on the result of the determination, wherein the first transmission means transmits, in addition to the generation request for the virtual user's utterance and the data of the utterance history, data of at least one of the calculated sympathy score and reliability score to the speech generation device.

16. A computer having a communication interface for communicating with a user terminal and a speech generation device (generation AI) via an electric public communication network, for a virtual user to have a conversation with a real user, means for receiving data of the real user's utterance from the user terminal; means for transmitting a request for generating a statement of the virtual user to the statement generation device together with data on the history of statements of each of the real user and the virtual user; means for receiving data on a statement of the virtual user from the statement generation device; means for converting the diction of the received statement of the virtual user; A program that functions as means for transmitting the converted data on the statement of the virtual user to the user terminal.

17. A computer for a virtual user to have a conversation with a real user, having a communication interface for communicating with a user terminal and a statement generation device (generative AI) via an electric public communication network, means for receiving data on a statement of the real user from the user terminal; means for transmitting a request for generating a statement of the virtual user to the statement generation device together with data on the history of statements of each of the real user and the virtual user; means for receiving data on a statement of the virtual user from the statement generation device; means for transmitting the received data on the statement of the virtual user to the user terminal; means for executing a determination as to whether the real user is affirmative or negative with respect to the statement of the virtual user based on the statement of the real user; means for calculating at least one of a sympathy score representing the degree of sympathy of the real user with respect to the virtual user and a reliability score representing the degree of trust of the real user with respect to the virtual user based on the result of the determination; A program that functions as means for generating, in association with the content of the statement of the virtual user, a change in at least one of the sympathy score and the reliability score as persona expansion information regarding the real user.

Images