Information processing systems, information processing methods, and programs

Abstract

To support the improvement of sales skills. [Solution] An information processing system that supports sales role-playing using a generation AI, which stores prompts for generating scenarios that differ for each negotiation phase, acquires negotiation information related to a negotiation specified by the user from an external system, determines the negotiation phase to be targeted for role-playing by identifying the negotiation phase of the negotiation based on the acquired negotiation information, and provides the generation AI with prompts corresponding to the determined negotiation phase to generate a scenario corresponding to that negotiation phase. As a result, the user can generate a scenario that matches the progress of the negotiation simply by specifying the negotiation, making scenario setting easy.

Description

Technical Field

[0006] , , ,

[0007] , ,

[0001] The present invention relates to an information processing system, an information processing method, and a program.

Background Art

[0002] Techniques for selling products using videos online have been proposed (see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Conventional techniques lack effective training methods for improving the skills of salespersons.

[0005] The present invention has been made in view of such a background, and an object thereof is to provide a technology capable of supporting the improvement of sales skills.

Means for Solving the Problems

[0006] The main invention of the present invention for solving the above problems is an information processing system that supports sales role-play using a generative AI, comprising: a prompt storage unit that stores a prompt for generating a scenario of the role-play; a negotiation phase determination unit that determines a negotiation phase to be the target of the role-play according to input information from a user; and a scenario generation unit that generates the scenario based on the determined negotiation phase and the prompt.

[0007] Regarding other problems disclosed in the present application and solutions therefor, they will be clarified by the embodiments of the invention and the drawings. [Effects of the Invention]

[0008] According to this invention, it is possible to support the improvement of sales skills. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows an example of the overall configuration of an information processing system. [Figure 2] This figure shows an example of the hardware configuration of management server 2. [Figure 3] This figure shows an example of the software configuration for management server 2. [Figure 4] This figure shows an example of the software configuration of user terminal 1. [Figure 5] This is a diagram illustrating the processing flow in an information processing system. [Modes for carrying out the invention]

[0010] <System Overview> The following describes an information processing system according to one embodiment of the present invention. The information processing system of this embodiment is a system that supports role-playing training for sales representatives by utilizing artificial intelligence (AI). In this system, the AI ​​plays the role of the sales partner, and the sales representative simulates a business negotiation while interacting with an avatar. The system automatically generates optimized scenarios for each negotiation phase, allowing sales representatives to conduct effective training in an environment close to an actual business negotiation. The AI ​​generates the avatar's behavior and dialogue, realistically reproducing the reactions and behavioral patterns of the sales partner. The role-playing scenario mainly defines the behavior of the sales partner and is dynamically adjusted according to the progress of the business negotiation.

[0011] Figure 1 shows an example of the overall configuration of an information processing system. The information processing system in this embodiment includes a management server 2. The management server 2 is connected to the user terminal 1 via a communication network. The communication network is, for example, the internet and is constructed using public telephone networks, mobile phone networks, wireless communication channels, Ethernet (registered trademark), etc.

[0012] User terminal 1 is a computer operated by the user. User terminal 1 can be, for example, a smartphone, a tablet computer, or a personal computer.

[0013] The management server 2 may be a general-purpose computer such as a workstation or personal computer, or it may be logically implemented through cloud computing.

[0014] <Management Server> Figure 2 shows an example of the hardware configuration of the management server 2. Note that the illustrated configuration is just one example, and other configurations are also possible. The management server 2 includes a CPU 201, memory 202, storage device 203, communication interface 204, input device 205, and output device 206. The storage device 203 stores various data and programs, such as a hard disk drive, solid-state drive, or flash memory. The communication interface 204 is an interface for connecting to a communication network, such as an adapter for connecting to Ethernet®, a modem for connecting to a public telephone network, a wireless communication device for wireless communication, or a USB (Universal Serial Bus) connector or RS232C connector for serial communication. The input device 205 is for inputting data, such as a keyboard, mouse, touch panel, button, or microphone. The output device 206 is for outputting data, such as a display, printer, or speaker. Furthermore, each functional unit of the management server 2, as described later, is realized by the CPU 201 reading programs stored in the storage device 203 into memory 202 and executing them, and each storage unit of the management server 2 is realized as part of the storage area provided by memory 202 and storage device 203.

[0015] Figure 3 shows an example of the software configuration of the management server 2. Note that the illustrated configuration is just one example, and other configurations are also possible. The management server 2 includes a prompt storage unit 231, a negotiation phase determination unit 211, a scenario generation unit 212, a customer information acquisition unit 213, a product information acquisition unit 214, a negotiation information acquisition unit 215, a role-playing execution unit 216, a role-playing input reception unit 217, a response generation unit 218, a response output unit 219, a termination determination unit 220, a scenario progress management unit 221, a prompt update unit 222, a dialogue management unit 223, a dialogue history recording unit 224, a scenario storage unit 232, a customer information storage unit 233, a product information storage unit 234, a negotiation information storage unit 235, a dialogue history storage unit 236, and an achievement element storage unit 237.

[0016] FIG. 4 is a diagram showing an example of the software configuration of the user terminal 1. The user terminal 1 includes an input unit 111, a display unit 112, a communication unit 113, and a storage unit 131.

[0017] <Management Server 2> Hereinafter, the functional units of the management server 2 will be described.

[0018] The prompt storage unit 231 stores prompts (prompt templates) for generating role-play scenarios for each negotiation phase. The prompt storage unit 231 stores prompts for the ice-breaking phase, prompts for the relationship-building phase, prompts for the needs and issues hearing phase, prompts for the solution proposal phase, prompts for the concern presentation phase, prompts for the price and condition negotiation phase, prompts for the closing phase, etc. Each prompt includes the personality setting of the business partner, the reaction pattern, the assumed speech content, the progress direction of the scenario, etc.

[0019] The scenario storage unit 232 stores the generated role-play scenarios. The scenario storage unit 232 stores the purpose of the scenario, the setting of the business partner, the assumed flow of the dialogue, the action pattern of the business partner in each scene, etc.

[0020] The business partner information storage unit 233 stores information about the business partner. The business partner information storage unit 233 stores the company name, industry, scale, position of the person in charge, past transaction history, the issues and concerns of the company, etc.

[0021] The product information storage unit 234 stores information about the products to be sold. The product information storage unit 234 stores the product name, functions, price, differentiation points from competing products, introduction cases, etc.

[0022] The negotiation information storage unit 235 stores information about the negotiation with the business partner. The negotiation information storage unit 235 stores the progress of the negotiation, past negotiation history, proposed content, requests and concerns from the partner, etc.

[0023] The dialogue history storage unit 236 stores user input and generated AI responses in chronological order during role-playing. The dialogue history storage unit 236 stores the content of the utterance, the time of the utterance, information about the ongoing scene, and the progress made in the scene, associating them with each other.

[0024] The achievement element storage unit 237 stores one or more achievement elements for each negotiation phase to determine the completion of that negotiation phase. An achievement element is an item that defines the dialogueal goal that the sales representative should achieve in that negotiation phase. For example, for the needs and challenges hearing phase, the achievement element storage unit 237 stores several achievement elements, such as the customer's current challenges being specifically articulated, the prioritization of the challenges being confirmed, the budget size being understood, and the decision-making process being understood. Each achievement element is associated with and stored the criteria used to determine the degree of achievement and the weighting coefficient used in determining the completion of the negotiation phase.

[0025] The negotiation phase determination unit 211 determines the negotiation phase to be used for role-playing based on the information input from the user. The negotiation phase determination unit 211 can also determine a negotiation phase explicitly specified by the user. Furthermore, the negotiation phase determination unit 211 can automatically determine a negotiation phase based on negotiation information acquired by the negotiation information acquisition unit 215. For example, if the negotiation information includes the information "first visit," the negotiation phase determination unit 211 will determine the icebreaker phase.

[0026] This section describes a specific example of how the sales phase determination unit 211 automatically determines the sales phase based on sales information. The sales phase determination unit 211 determines the sales phase by applying a predetermined response rule to at least one of the sales stage, progress rate, number of contacts with the sales partner, and most recent activity type included in the sales information acquired by the sales information acquisition unit 215 from an external system. For example, the sales phase determination unit 211 determines the icebreaker phase when the number of contacts is 0 or the sales stage is "Lead," the needs / issue hearing phase when the sales stage is "Hearing," and the price / terms negotiation phase when a quote has been submitted and the progress rate is above a predetermined value. If multiple response rules apply, the sales phase determination unit 211 determines one sales phase according to the priority order assigned to each response rule in advance.

[0027] Furthermore, users can specify a sales opportunity to be role-played. In this case, the sales opportunity information acquisition unit 215 acquires sales opportunity information related to the specified sales opportunity from an external system such as an SFA system or CRM system, and the sales opportunity phase determination unit 211 determines the sales opportunity phase to be role-played by identifying which sales opportunity phase the opportunity is currently in based on the acquired sales opportunity information. The scenario generation unit 212 generates a scenario corresponding to the sales opportunity phase using prompts corresponding to the identified sales opportunity phase. As a result, users only need to specify a sales opportunity to be role-played, and a scenario for the sales opportunity phase that matches the current progress of that sales opportunity is automatically generated, eliminating the need to manually set the persona of the sales partner or the sales opportunity phase, and making scenario setup easy.

[0028] The scenario generation unit 212 generates a scenario based on the determined negotiation phase and the prompts stored in the prompt storage unit 231. The scenario generation unit 212 can generate a scenario corresponding to a negotiation phase by setting the negotiation phase for a single prompt that generates a scenario corresponding to the negotiation phase. In addition, the scenario generation unit 212 can also obtain the prompt for the corresponding negotiation phase from the prompt storage unit 231, combine it with customer information and product information, and input it into a large-scale language model to generate a scenario. For example, in the needs and challenges hearing phase, the scenario generation unit 212 can generate a scenario such as "the customer gradually discloses information about their company's challenges, initially discussing only superficial issues, and whether they discuss deeper issues depends on how the salesperson asks questions."

[0029] The prompts for each negotiation phase stored in the prompt storage unit 231 may be configured as templates having insertion areas (slots). In this case, the scenario generation unit 212 reads the template corresponding to the determined negotiation phase and inserts the customer information obtained from the customer information storage unit 233, the product information obtained from the product information storage unit 234, the summary of negotiation information obtained from the negotiation information acquisition unit 215, and the achievement elements for the negotiation phase obtained from the achievement element storage unit 237 into the corresponding slots of the template, thereby synthesizing the input data for the generating AI. The scenario generation unit 212 provides the synthesized input data to the generating AI to generate a scenario that defines the behavior of the customer and the dialogue structure specific to the negotiation phase.

[0030] The customer information acquisition unit 213 acquires customer information about potential customers. The customer information acquisition unit 213 can acquire customer information through user input. In addition, the customer information acquisition unit 213 can also acquire customer information from external systems such as CRM (Customer Relationship Management) systems.

[0031] The product information acquisition unit 214 acquires product information related to products for sales partners. The product information acquisition unit 214 can acquire product information through user input. In addition, the product information acquisition unit 214 can also acquire product information from external systems such as product management systems.

[0032] The sales opportunity information acquisition unit 215 acquires sales opportunity information related to sales negotiations with sales partners from external systems. For example, the sales opportunity information acquisition unit 215 can acquire the progress status of sales negotiations, past sales negotiation history, and proposal details from SFA (Sales Force Automation) systems and CRM systems.

[0033] The role-playing execution unit 216 performs sales role-playing in a dialogue format between the user and the generated AI according to the scenario. Based on the generated scenario, the role-playing execution unit 216 controls the dialogue between the generated AI acting as the sales partner and the user, providing a realistic sales negotiation environment.

[0034] The role-playing input receiving unit 217 sequentially receives role-playing input from the user. The role-playing input receiving unit 217 can accept text input, voice input, or selection input from a list of options. In the case of voice input, the role-playing input receiving unit 217 converts it to text using speech recognition technology.

[0035] The response generation unit 218 prompts the generating AI to generate a role-playing response based on the received role-playing input and scenario. The response generation unit 218 considers the current scenario progress, user input, and the setting of the sales partner to cause the generating AI to generate an appropriate response as the sales partner. For example, in the price negotiation phase, the response generation unit 218 generates a response that "indicates budget constraints while hinting that it may be considered depending on the conditions."

[0036] The response output unit 219 outputs the generated role-playing response to the user, stores the role-playing response as history information, and uses it as reference information for generating the next prompt. The response output unit 219 can output the response by displaying text, outputting audio, or displaying avatar movements. The response output unit 219 records the outputted response in the dialogue history recording unit 224 and uses it as contextual information when generating the next response.

[0037] The termination determination unit 220 terminates the role-playing based on user input or predetermined termination conditions. The termination determination unit 220 terminates the role-playing when the user explicitly instructs termination, when the scenario is completed, or when a set time has elapsed.

[0038] The scenario progress management unit 221 divides the generated scenario into scenes or steps and manages its progress. Furthermore, the scenario progress management unit 221 determines the completion of the ongoing negotiation phase by following the procedure below. The scenario progress management unit 221 extracts the history associated with the ongoing negotiation phase from the history recorded in the dialogue history recording unit 224. Next, the scenario progress management unit 221 reads the achievement elements corresponding to the ongoing negotiation phase from the achievement element storage unit 237, evaluates the extracted history for each achievement element, and calculates a score representing the degree of achievement of that achievement element. The score is calculated, for example, by providing the history and the criteria for determining the achievement element to the generating AI as an evaluation prompt, and having it output a numerical value between 0 and 1 indicating the degree to which the achievement element has been met. The scenario progress management unit 221 calculates the sum by multiplying the score of each achieved element by the weight coefficient of that element. If the sum is equal to or greater than a predetermined threshold and the scores of each achieved element are equal to or greater than a predetermined lower limit, the unit determines that the ongoing negotiation phase is complete and proceeds to the next negotiation phase. In order to suppress false judgments, the scenario progress management unit 221 may perform the completion determination only after the number of back-and-forth exchanges in the ongoing negotiation phase has reached a predetermined number.

[0039] The prompt update unit 222 can update prompts in accordance with the progress of the scenario. Specifically, when the scenario progress management unit 221 determines that the ongoing negotiation phase is complete, the prompt update unit 222 identifies the next negotiation phase, reads the prompt corresponding to that next negotiation phase from the prompt storage unit 231, and updates the prompt provided to the generating AI with the read prompt. As a result, the negotiation phase automatically transitions according to the progress of the conversation without the sales representative having to explicitly instruct the switching of negotiation phases, and scenarios optimized for each negotiation phase are continuously provided.

[0040] The dialogue management unit 223 inputs the updated prompt to the generation AI, obtains the role-playing response from the generation AI's output, and returns the obtained response to the scenario progress management unit 221 to control the transition to the next scene. The dialogue management unit 223 manages the coordination with the generation AI and also performs quality checks on responses and corrects inappropriate responses.

[0041] The dialogue history recording unit 224 stores user input and generated AI responses in chronological order, associating them with at least the most recent utterance pair and the ongoing scene information. The dialogue history recording unit 224 can record not only the content of the utterances but also additional information such as the emotions at the time of the utterance and the degree of the sales partner's reaction.

[0042] <User Terminal 1> The following describes the functional components of user terminal 1.

[0043] The memory unit 131 is the storage device of the user terminal 1. The memory unit 131 stores data entered by the user and role-playing history information.

[0044] The input unit 111 accepts data input from the user. The input unit 111 accepts input such as sales partner information, product information, selection of negotiation phase, and statements made during role-playing. The input unit 111 supports various input methods, including text input, voice input, and selection input via touch operation.

[0045] The display unit 112 displays information received from the management server 2 to the user. The display unit 112 displays the generated scenario, the sales partner's avatar, the conversation content, the progress of the role-playing, etc. The display unit 112 is equipped with an avatar display function to visually represent the emotions and reactions of the sales partner.

[0046] The communication unit 113 communicates with the management server 2. The communication unit 113 sends user input data to the management server 2 and receives role-playing responses and scenario information from the management server 2.

[0047] Figure 5 is a diagram illustrating the processing flow in an information processing system.

[0048] The management server 2 receives customer information, product information, and negotiation phase selection information from the user terminal 1 (S101). The negotiation phase determination unit 211 determines the target negotiation phase based on the received input information (S102). The scenario generation unit 212 obtains prompts corresponding to the determined negotiation phase from the prompt storage unit 231 (S103), and generates a role-playing scenario by providing customer information, product information, and prompts to a large-scale language model (S104). The management server 2 sends the generated scenario to the user terminal 1 and starts role-playing (S105). The role-playing input reception unit 217 receives role-playing input from the user (S106), and the response generation unit 218 assigns prompts to the generation AI based on the received input and scenario to generate a role-playing response (S107). The response output unit 219 outputs the generated response to the user terminal 1 and stores it as history information in the dialogue history recording unit 224 (S108). The termination determination unit 220 determines the end of the role-playing based on user input or predetermined termination conditions (S109). If it is not the end, it returns to step S106; if it is the end, it completes the process (S110).

[0049] As described above, the information processing system of this embodiment enables effective role-playing training for sales representatives by utilizing generating AI. By automatically generating scenarios optimized for each phase of a sales negotiation, it is possible to provide practical training tailored to each stage, from icebreakers and relationship building to needs assessment, solution proposals, addressing concerns, price negotiations, and closing. Customized scenarios that take into account sales partner information and product information enable training in an environment close to actual sales negotiations, efficiently supporting the skill improvement of sales representatives. Furthermore, by recording and analyzing the dialogue history, it is possible to understand the strengths and areas for improvement of sales representatives and promote continuous skill improvement. In addition, by managing the progress of the scenario, step-by-step learning is realized, and training at an appropriate difficulty level according to the sales representative's level can be provided.

[0050] Although these embodiments have been described above, they are intended to facilitate understanding of the present invention and are not intended to limit its interpretation. The present invention can be modified and improved without departing from its spirit, and equivalents thereof are also included.

[0051] For example, the processing performed by each functional unit of the management server 2 described above may be executed by any of the functional units. Furthermore, different functional units may be added to perform some of the processing performed by each of the functional units described above. Also, the functional units of the management server 2 may be distributed across multiple computers.

[0052] Furthermore, the information stored in each memory unit of the management server 2 may be stored in any of the memory units. That is, the information stored in the multiple memory units mentioned above may be stored in a single memory unit, or a portion of the information stored in one memory unit may be stored in another memory unit.

[0053] <Example 1> In the embodiment described above, an example was shown where the sales negotiation phase is determined before generating a scenario. However, it is also possible to generate a continuous role-playing scenario that combines multiple sales negotiation phases. In this case, the scenario generation unit 212 generates an integrated scenario that includes a series of sales negotiation flows from icebreaker to relationship building, needs assessment, solution proposal, concern management, price negotiation, and closing. The scenario progress management unit 221 monitors the completion conditions of each phase and controls the system to automatically move to the next phase when the conditions are met. For example, if the sales representative has a sufficient understanding of the client's challenges during the needs assessment phase, the scenario progress management unit 221 determines that it is time to move to the solution proposal phase, and the prompt update unit 222 updates the prompt to one appropriate for the proposal phase. This allows the sales representative to receive comprehensive training throughout the entire sales negotiation process.

[0054] <Modification 2> The above-described embodiment demonstrates role-playing through text-based dialogue, but a voice-interactive role-playing system combining speech recognition and speech synthesis technologies may also be implemented. In this case, the role-playing input receiving unit 217 converts the user's voice input into text using a speech recognition engine, and further analyzes the intonation, speaking speed, and emotional tone of the voice to utilize in evaluating sales skills. The response output unit 219 converts the generated role-playing response into speech using a speech synthesis engine and outputs it to the user. During speech synthesis, the voice quality and speaking style are adjusted according to the setting of the sales partner; for example, a calm, low voice is used for the role of an older executive, and a bright, high-pitched voice is used for the role of a younger salesperson. The dialogue history recording unit 224 records both voice data and text data so that areas for improvement in the salesperson's speaking style and presentation skills can be identified in later analysis.

[0055] <Variation 3> The embodiment described above shows a role-playing scenario with a single sales partner, but it is also possible to generate a role-playing scenario in the form of a meeting with multiple sales partners. In this case, the scenario generation unit 212 generates a scenario that includes multiple sales partner characters with different roles, such as decision-makers, technical staff, and finance staff. Each character has different interests and concerns; for example, the technical staff member is set to focus on functional details, the finance staff member on cost, and the decision-maker on return on investment. The response generation unit 218 generates responses from the multiple characters from their respective positions in response to the user's statements, and the response output unit 219 displays or outputs audio for each character separately. The scenario progress management unit 221 individually manages the level of understanding and interest of each character and controls the system to proceed to the next phase only when the agreement of all parties is obtained. This makes it possible to provide advanced sales training that corresponds to the complex decision-making processes in actual companies.

[0056] <Modification 4> The embodiment described above shows scenario generation based on predefined negotiation phases, but it is also possible to generate individually optimized scenarios by analyzing the sales representative's past performance and weaknesses. In this case, the management server 2 further includes a sales performance analysis unit, which analyzes the sales representative's past role-playing history, actual negotiation results, customer feedback, etc., to identify the sales representative's strengths and weaknesses. The scenario generation unit 212 adjusts the difficulty and focus of the scenarios so that the identified weaknesses can be trained intensively. For example, for a sales representative who is not good at price negotiation, multiple price negotiation scenarios with progressively increasing difficulty are generated so that they can be trained step by step, starting with basic price offers, responding to tough price reduction requests, and dealing with price competition with competitors. The prompt update unit 222 adjusts the difficulty of the scenarios in real time according to the sales representative's proficiency level, maintaining an appropriate challenge level.

[0057] <Modification 5> The above-described embodiment demonstrated real-time interactive role-playing, but a function to provide real-time feedback and advice to the salesperson's statements may be added. In this case, the management server 2 further includes a feedback generation unit, which analyzes each statement made by the salesperson to immediately identify areas for improvement and positive aspects. For example, if a salesperson interrupts a client to begin explaining their company's products, the feedback generation unit generates advice such as, "You can build better relationships by listening to the client until they finish speaking before responding." The response output unit 219 displays the feedback information to the user along with the response from the client, allowing the salesperson to improve their sales techniques while continuing the conversation. The feedback generation unit also generates a comprehensive evaluation report after the conversation ends and provides a scoring function that allows for the quantitative measurement of the salesperson's growth. The conversation history recording unit 224 records the feedback information in association with the salesperson's improvement status, enabling the measurement of continuous learning effects.

[0058] <Variation 6> In the embodiment described above, an example was shown in which the score of the achievement elements is used to determine the completion of the negotiation phase. However, the score may also be used to adjust the prompt for the next negotiation phase. In this case, the prompt update unit 222 updates the prompt for the next negotiation phase to one with a higher difficulty level for the sales partner, the higher the score of the achievement elements calculated in the ongoing negotiation phase. For example, for a sales representative with a high level of achievement in the needs and issues hearing phase, the prompt update unit 222 updates the prompt to one that specifies a sales partner who will make stronger price reduction demands in the price and conditions negotiation phase. This makes it possible to dynamically provide role-playing with a difficulty level that matches the proficiency of the sales representative.

[0059] <Example 7> In the embodiments described above, when transitioning between negotiation phases, the content of the dialogue in the previous negotiation phase may be carried over to the next negotiation phase. In this case, when transitioning between negotiation phases, the scenario progress management unit 221 generates summary information from the history of the previous negotiation phase recorded in the dialogue history recording unit 224, including issues disclosed by the sales partner, agreed-upon matters, unresolved concerns, and prices and conditions already presented. The prompt update unit 222 updates the prompt by inserting this summary information into the prompt slot corresponding to the next negotiation phase. As a result, the sales partner's character setting, issues, and previously mentioned conditions are consistently maintained across negotiation phases, and a scenario is generated in which multiple negotiation phases are aligned as a single continuous negotiation.

[0060] <Differentiation Example 8> In the embodiment described above, an example was shown in which the system moves to the next negotiation phase when the completion of the negotiation phase is determined. However, the scenario progress management unit 221 may perform multiple types of transitions based on the calculated score of the achievement elements. For example, the scenario progress management unit 221 may (a) determine that the dialogue is stalled because the sum of the scores does not meet a predetermined lower limit, continue the ongoing negotiation phase, or return to the previous negotiation phase and regenerate a scenario that focuses on the missing achievement elements; (b) skip negotiation phases that are determined to have already been completed based on the negotiation information acquired by the negotiation information acquisition unit 215; and (c) branch to the negotiation phase corresponding to the reaction (e.g., the concern presentation phase or the price / terms negotiation phase) regardless of the order of the negotiation phases if a specific reaction (e.g., strong price concerns) is detected from the history recorded in the dialogue history recording unit 224. The prompt update unit 222 updates the prompt to the negotiation phase to be transitioned to according to these transitions.

[0061] <Modification 9> In the embodiment described above, an example of generating a scenario based on predetermined negotiation phases was shown. However, multiple playbooks defining sets of negotiation phases and their order may be stored and selected according to the situation. In this case, the management server 2 further includes a playbook storage unit that stores multiple playbooks defining sets of negotiation phases and their progression order for each industry, product, or sales method. The negotiation phase determination unit 211 selects a playbook based on customer information, product information, or a sales method specified by the user, and determines each negotiation phase according to the order defined by the selected playbook. This provides role-playing with a negotiation phase configuration that is suitable for the product and sales style.

[0062] <Variation 10> In the embodiment described above, the expected amount of dialogue for each negotiation phase may be managed. In this case, the expected number of dialogue turns or time is associated with and stored for each negotiation phase. The scenario progress management unit 221 monitors the number of turns or time elapsed in the ongoing negotiation phase, and if the associated upper limit is reached but the completion conditions for the negotiation phase are not met, it switches to a scenario that promotes the progress of the negotiation phase, such as the sales partner prompting a conclusion, or provides hints to the user. The scenario progress management unit 221 may also prevent the transition to a negotiation phase before the associated lower limit is reached, thereby suppressing hasty transitions.

[0063] <Variation 11> In the embodiment described above, the prompt storage unit 231 may store multiple prompts of different difficulty levels for the same negotiation phase. In this case, the scenario generation unit 212 selects and uses one prompt from the multiple prompts corresponding to the negotiation phase based on the user's skill level or the score of the achievement elements of the negotiation phase in the user's past role-playing. For example, the scenario generation unit 212 selects a prompt with easy salesperson behavior for users with low past scores in the negotiation phase, and selects a prompt with more advanced salesperson behavior for users with high scores.

[0064] <Variation 12> In the embodiment described above, an example was shown in which prompts stored in advance by the prompt storage unit 231 are used. However, prompts may also be generated from achievement elements stored in the achievement element storage unit 237. In this case, the scenario generation unit 212 provides the generation AI with the achievement elements for the negotiation phase and their judgment criteria stored in the achievement element storage unit 237, and automatically generates prompts that specify that the sales partner engages in dialogue to elicit the achievement elements from the sales representative. This makes it easy to expand the configuration of negotiation phases, as a corresponding prompt can be obtained by defining the achievement elements for the negotiation phase, even when adding a new negotiation phase.

[0065] <Example 13> In the embodiment described above, if the sum of the scores of the achievement elements in the completion determination of an ongoing negotiation phase does not reach a predetermined threshold, the scenario progress management unit 221 may regenerate a different scenario for that negotiation phase that focuses on the achievement elements with low scores, and repeat the same negotiation phase. This makes it possible to provide iterative training on a negotiation phase basis.

[0066] <Example 14> In the above-described variation 1, an example was shown of generating an integrated scenario that includes multiple negotiation phases. However, the scenario generation unit 212 may also generate a scenario for each negotiation phase sequentially at the time of transition to that negotiation phase. In this case, when generating a scenario for the target negotiation phase, the scenario generation unit 212 refers to the previous dialogue history and the score of the achievement elements, so it can generate a scenario that is consistent with the content of the dialogue that actually took place up to that point.

[0067] <Example 15> In the embodiment described above, the opportunity information acquisition unit 215 may continue to acquire opportunity information from an external system even while role-playing is in progress. In this case, when the acquired opportunity information is updated, the opportunity phase determination unit 211 re-evaluates the opportunity phase based on the updated opportunity information and changes the ongoing opportunity phase as necessary. The prompt update unit 222 updates the prompt in accordance with the change.

[0068] <Variation 16> In the embodiment described above, the management server 2 may further include a setting unit that sets or changes the achievement elements stored in the achievement element storage unit 237, their judgment criteria, weight coefficients, and thresholds used for completion determination, in response to input from a user or administrator. This makes it possible to adjust the criteria for completing the negotiation phase for each company or product.

[0069] <Example 17> In the embodiment described above, the display unit 112 may display to the user the ongoing negotiation phase, the score calculated for each achievement element of that negotiation phase, and the overall progress of the negotiation phase. This allows the user to perform role-playing while understanding which achievement elements of which negotiation phase they are lacking.

[0070] <Disclosure Items> Furthermore, this disclosure also includes the following configurations. [Item 1] An information processing system that uses generative AI to support sales role-playing, A prompt storage unit that stores prompts for generating the aforementioned role-playing scenario, A negotiation phase determination unit that determines the negotiation phase to be the subject of the role-playing based on user input information, A scenario generation unit that generates the scenario based on the determined negotiation phase and prompt, An information processing system equipped with the following features. [Item 2] A sales partner information acquisition unit that acquires customer information about sales partners, A product information acquisition unit that acquires product information regarding products for the aforementioned sales partner, Furthermore, The information processing system described in item 1, wherein the scenario generation unit generates the scenario by providing the counterparty information, the product information, and the prompt to a large-scale language model. [Item 3] A sales partner information acquisition unit that acquires customer information about sales partners, A business negotiation information acquisition unit that acquires business negotiation information regarding business negotiations with the aforementioned business partner from an external system, Furthermore, The aforementioned negotiation phase determination unit is an information processing system according to item 1, which determines the negotiation phase based on the negotiation information. [Item 4] The system further includes a role-playing execution unit that performs a sales role-play in the form of a dialogue between the user and the generated AI according to the aforementioned scenario. The aforementioned role-playing execution unit is A role-playing input receiving unit that sequentially receives role-playing input from users, A response generation unit that generates a role-playing response by providing prompts to the generating AI based on the received role-playing input and the scenario, A response output unit outputs the generated role-playing response to the user, stores the role-playing response as history information, and uses it as reference information for generating the next prompt, An termination determination unit that terminates the role-playing based on user input or predetermined termination conditions, An information processing system as described in item 1, comprising the features described in item 1. [Item 5] An information processing method that uses generative AI to support sales role-playing, The steps include storing prompts for generating the aforementioned role-playing scenario, A step of determining the sales negotiation phase to be the subject of the role-playing based on information input from the user, A step of generating the scenario based on the determined negotiation phase and the prompt, A method of information processing performed by a computer. [Item 6] An information processing method that uses generative AI to support sales role-playing, The steps include storing prompts for generating the aforementioned role-playing scenario, A step of determining the sales negotiation phase to be the subject of the role-playing based on information input from the user, A step of generating the scenario based on the determined negotiation phase and the prompt, A program that causes a computer to execute something. [Explanation of symbols]

[0071] 1 User terminal 2 Management Server

Claims

1. An information processing system that uses generational AI to support sales role-playing, A prompt storage unit that stores prompts for generating the role-playing scenario, which differs for each negotiation phase, A business opportunity information acquisition unit obtains business opportunity information related to a business opportunity specified by the user from an external system, A business opportunity phase determination unit determines the business opportunity phase to be the subject of the role-playing by identifying the business opportunity phase of the designated business opportunity based on the business opportunity information obtained, A scenario generation unit reads the prompt corresponding to the determined negotiation phase from the prompt storage unit, provides the read prompt to the generation AI, and generates the scenario corresponding to the negotiation phase. For each negotiation phase, an achievement element storage unit stores one or more achievement elements for determining the completion of the negotiation phase. A dialogue history recording unit records the user's role-playing input and the AI's generated responses in chronological order, in relation to the ongoing negotiation phase. A scenario progress management unit evaluates the history associated with the ongoing negotiation phase from the history recorded in the dialogue history recording unit for each achievement element, calculates a score representing the degree of achievement of the achievement element, and determines the completion of the negotiation phase and moves to the next negotiation phase if the calculated score meets predetermined conditions. A prompt update unit updates the prompt to the next negotiation phase in accordance with the transition of the negotiation phase by the scenario progress management unit, An information processing system equipped with the following features.

2. The aforementioned business negotiation information acquisition unit acquires the business negotiation information from the aforementioned external system, which is a sales support system or a customer relationship management system. The information processing system according to claim 1, wherein the negotiation phase determination unit identifies the negotiation phase by applying at least one of the negotiation stage, progress rate, and number of contacts included in the negotiation information to a predetermined correspondence rule.

3. The information processing system according to claim 1, wherein the scenario progress management unit provides the history and the criteria for determining the achievement elements to the generating AI as evaluation prompts, and calculates the score by quantifying the output evaluation results.

4. The information processing system according to claim 1, wherein the prompt update unit updates the prompt for the next negotiation phase to one that has a different difficulty level for the sales partner, according to the score calculated in the ongoing negotiation phase.

5. The information processing system according to claim 1, wherein the prompt is a template having slots, and the scenario generation unit inserts customer information regarding sales partners, product information regarding products, and a summary of the business negotiation information into the corresponding slots of the template to synthesize input data for the generating AI.

6. A sales partner information acquisition unit that acquires customer information about sales partners, A product information acquisition unit that acquires product information regarding products for the aforementioned sales partner, Furthermore, The information processing system according to claim 1, wherein the scenario generation unit generates the scenario by providing the counterparty information, the product information, and the prompt to a large-scale language model.

7. The system further includes a role-playing execution unit that performs a sales role-play in the form of a dialogue between the user and the generated AI according to the aforementioned scenario. The role-playing execution unit is A role-playing input receiving unit that sequentially receives role-playing input from users, A response generation unit that generates a role-playing response by providing prompts to the generating AI based on the received role-playing input and the scenario, A response output unit outputs the generated role-playing response to the user, stores the role-playing response as history information, and uses it as reference information for generating the next prompt, An termination determination unit that terminates the role-playing based on user input or predetermined termination conditions, The information processing system according to claim 1, comprising:

8. An information processing method that uses generative AI to support sales role-playing, A step of storing prompts for generating the aforementioned role-playing scenarios, which differ for each negotiation phase, The steps include obtaining information about a business opportunity specified by the user from an external system, and The steps include determining the negotiation phase to be used for the role-playing by identifying the negotiation phase of the designated negotiation based on the negotiation information obtained, The steps include: reading the prompt corresponding to the determined negotiation phase, providing the read prompt to the generating AI to generate the scenario corresponding to the negotiation phase; For each negotiation phase, the step of storing one or more achievement elements for determining the completion of the negotiation phase, The process involves recording the user's role-playing input and the AI's responses in chronological order, in relation to the ongoing sales negotiation phase. The steps include: evaluating the history associated with the ongoing negotiation phase from the recorded history for each achievement element, calculating a score representing the degree of achievement of said achievement element, determining the completion of the negotiation phase and moving to the next negotiation phase if the calculated score meets predetermined conditions; The steps include updating the prompt to correspond to the next negotiation phase in accordance with the transition of the negotiation phase by the aforementioned scenario progress management unit, A method of information processing performed by a computer.

9. A program that uses generative AI to support sales role-playing, A step of storing prompts for generating the aforementioned role-playing scenarios, which differ for each negotiation phase, The steps include obtaining information about a business opportunity specified by the user from an external system, and The steps include determining the negotiation phase to be used for the role-playing by identifying the negotiation phase of the designated negotiation based on the negotiation information obtained, The steps include: reading the prompt corresponding to the determined negotiation phase, providing the read prompt to the generating AI to generate the scenario corresponding to the negotiation phase; For each negotiation phase, the step of storing one or more achievement elements for determining the completion of the negotiation phase, The process involves recording the user's role-playing input and the AI's responses in chronological order, in relation to the ongoing sales negotiation phase. The steps include: evaluating the history associated with the ongoing negotiation phase from the recorded history for each achievement element, calculating a score representing the degree of achievement of said achievement element, determining the completion of the negotiation phase and moving to the next negotiation phase if the calculated score meets predetermined conditions; The steps include updating the prompt to correspond to the next negotiation phase in accordance with the transition of the negotiation phase by the aforementioned scenario progress management unit, A program that causes a computer to execute something.

Images