Generation apparatus, generation method, and generation program

Abstract

To appropriately improve voice communication. [Solution] The generation device 100 extracts issues regarding the usage of voice communication by users who utilize voice communication via the telephone network for their business activities. Using the extracted issues and acquired information on the users' business activities, the generation device 100 inputs prompts into a large-scale language model that include commands to generate solutions corresponding to the issues and talk scripts in which the solutions are expressed based on a predetermined story, thereby generating solutions and talk scripts.

Description

Technical Field

[0001] The present invention relates to a generation device, a generation method, and a generation program.

Background Art

[0002] Voice communication via a telephone network may be used in business activities. For example, when a call is made to a predetermined telephone number, the call is connected to an operator, and appropriate responses are made according to requirements. However, in the above voice communication via a telephone network, events such as the call being made but not being connected to the operator may occur, which may lead to a decrease in customer satisfaction.

[0003] Therefore, as a conventional technique, there is known a technique for outputting an analysis result of call status data per unit time based on log data associating the start time, end time, and incoming / outgoing call numbers of calls between telephones (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, the conventional technique has a problem in appropriately realizing an improvement in voice communication. For example, the conventional technique outputs an analysis result of call status data per unit time such as the call volume. However, the conventional technique only outputs the analysis result, and it is difficult to generate a solution based on the analysis result or a talk script for appropriately proposing the solution to a user who uses voice communication in business activities.

Means for Solving the Problems

[0006] Therefore, in order to solve the above-mentioned problems and achieve the objective, the present invention is characterized by comprising: an extraction unit that extracts problems regarding the usage status of voice communication by users who use voice communication via a telephone network for their business activities; and a generation unit that inputs prompts including a command to generate a solution corresponding to the problem and a talk script in which the solution is expressed based on a predetermined story into a large-scale language model using the extracted problems and acquired information on the user's business activities, thereby generating the solution and the talk script. [Effects of the Invention]

[0007] The present invention has the effect of appropriately improving voice communication. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a diagram illustrating the overall process of the generating apparatus according to the embodiment. [Figure 2] Figure 2 shows the configuration of the generating apparatus according to this embodiment. [Figure 3] Figure 3 is a table diagram showing an example of usage information according to the embodiment. [Figure 4] Figure 4 shows an example of call volume data according to the embodiment. [Figure 5] Figure 5 shows an example of the call volume analysis results according to the embodiment. [Figure 6] Figure 6 is a table diagram showing an example of business information according to the embodiment. [Figure 7] Figure 7 shows an example of a candidate solution according to this embodiment. [Figure 8] Figure 8 is a table diagram showing an example of talk script information according to the embodiment. [Figure 9] Figure 9 shows an example of a prompt according to the embodiment. [Figure 10] Figure 10 shows an example of the solution and talk script output according to the embodiment. [Figure 11]FIG. 11 is a diagram showing a flowchart of processing by the generation device according to the embodiment. [Figure 12] FIG. 12 is a diagram showing an example of a computer that realizes the generation device according to the embodiment.

Embodiments of the Invention

[0009] Hereinafter, embodiments for implementing the present invention (hereinafter referred to as "embodiments") will be described with reference to the drawings. Note that each embodiment is not limited to the content described below.

[0010] <Overall Overview> (Background) When a call is made to a predetermined telephone number, the call is connected to an operator, and appropriate responses are made according to requirements. Thus, voice communication via a telephone network may be used in business activities.

[0011] In the above voice communication via a telephone network, in order to respond to a decrease in customer satisfaction, based on log data that associates the start time, end time, and incoming / outgoing telephone numbers of calls between telephones, a reference technique that outputs the analysis result of call status data per unit time is known.

[0012] However, the reference technique only outputs the analysis result, and it is difficult to generate a solution to the problem and a talk script for appropriately proposing the solution. There are problems in appropriately realizing the improvement of voice communication.

[0013] (Processing by Generation Device 100) Therefore, the generation device 100 according to the present embodiment generates a solution to a problem extracted based on the usage status of voice communication and a talk script for explaining the solution to the user.

[0014] Here, the overall image of the processing by the generation device 100 will be described. FIG. 1 is a diagram for explaining the overall image of the processing of the generation device 100 according to the embodiment. The generation device 100 shown in FIG. 1 is an example of a computer that provides a technology for realizing the information processing described below.

[0015] First, the generation device 100 extracts, using data regarding the call volume related to the use of voice communication (hereinafter, may be referred to as "call volume data") included in the information regarding the use of voice communication of a user who uses voice communication via a telephone network for business activities (hereinafter, may be referred to as "usage situation information"), the issues regarding the usage situation of the voice communication (FIG. 1(1-1)). Note that the above usage situation information includes information such as the company name (identification information) related to the user, the telephone number, and the call volume data (FIG. 1(1-2)).

[0016] The generation device 100 inputs a prompt (FIG. 1(2-3)) including a command to generate a solution according to the issue and a talk script in which the solution is expressed based on a predetermined story, using the issue regarding the usage situation of the voice communication extracted from the call volume data (FIG. 1(2-1)) and the information regarding the business activities of the acquired user (FIG. 1(2-2)), into a large language model to generate a solution and a talk script (FIG. 1(2-4)).

[0017] Note that the information regarding the business activities of the above user includes information such as the business content and organizational information in the business activities related to the user (FIG. 1(2-5)), and may be hereinafter referred to as "business activity information".

[0018] Then, the generation device 100 outputs the generated solution and talk script to a person in charge (hereinafter, may be simply referred to as the "person in charge") who proposes the solution to the target user (FIG. 1(3)).

[0019] In this way, the generation device 100 according to this embodiment generates solutions that address challenges in the usage of voice communication, and generates talk scripts to be used when proposing such solutions to users, thereby enabling efficient and effective solution proposals to users. As a result, it achieves the effect of appropriately improving voice communication.

[0020] <Description of Generator 100> The configuration of the generation device 100 according to this embodiment will now be described. Figure 2 is a diagram showing the configuration of the generation device 100 according to this embodiment. As shown in Figure 2, the generation device 100 has a communication unit 110, a storage unit 120, and a control unit 130.

[0021] Although not shown in Figure 2, the generation device 100 may be equipped with an input unit such as a keyboard or mouse to receive input such as operations from an administrator. Furthermore, the generation device 100 may be equipped with a display or the like to show the administrator information regarding the generation process of solutions and talk scripts.

[0022] (Communications Department 110) The communication unit 110 performs data communication related to the input of usage information collected from external sources, business activity information acquired from external sources, etc. The communication unit 110 also performs data communication related to the output of issues regarding the extracted voice communication usage status, information on solutions corresponding to those issues, and information on talk scripts. The communication unit 110 is implemented using a NIC (Network Interface Card), etc., and controls communication via telecommunication lines such as a LAN (Local Area Network) or the Internet. Furthermore, the communication unit 110 can be connected to the network via wired or wireless connections as needed, and can send and receive information bidirectionally with terminal devices, etc.

[0023] (Storage unit 120) The storage unit 120 stores data and programs used for various processes by the control unit 130, as well as various data acquired through the operation of the control unit 130. The storage unit 120 is implemented using semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or storage devices such as hard disks and optical discs. As shown in Figure 2, the storage unit 120 also includes a usage status information DB 121, a business activity information DB 122, a solution candidate DB 123, a talk script information DB 124, and a generation model DB 125.

[0024] (Usage Information DB121) The Usage Information DB121 is a database that stores information (usage information) regarding the usage of voice communications in a user's business activities. Specifically, the Usage Information DB121 stores information such as user identification information, telephone number, call volume data, and issues related to the extracted voice communication usage.

[0025] Here, an example of usage information stored in the usage information DB121 will be explained using Figure 3. Figure 3 is a table diagram showing an example of usage information according to the embodiment. The usage information DB121 stores information related to each of the following items, associating them with "No," which is information that identifies individual data included in user information, "identification information," "telephone number," "call volume data," "call volume analysis results," and "issues," in a table format or the like.

[0026] For example, as shown in Figure 3, the usage information DB121 stores identification information "A" identified by No. "1", telephone number "B", call volume data "C", call volume analysis results "D", and issue "E". The letters "A to E" listed for each item in the table diagram shown in Figure 3 are legends for the information contained in each item.

[0027] "Identification information" refers to information that identifies a user who uses voice calls for business activities, and includes, for example, information expressed by a combination of predetermined strings of characters, numbers, symbols, etc. Note that identification information may be modified by deleting or replacing information that could identify the individual user using publicly known technologies.

[0028] "Telephone number" refers to information that identifies the telephone used to mediate voice calls in business activities, and includes information that is represented by a predetermined combination of numbers, for example. "Call volume data" refers to data on call volume obtained based on the usage status of voice communications, and includes information such as total call volume, incomplete call volume (after-hours call volume, call volume outside designated area, call volume dropped midway, call volume encountered while busy, etc.), and completed call volume.

[0029] Here, using Figure 4, we will explain the call volume data acquired based on the usage status of voice communication. Figure 4 is a diagram showing an example of call volume data according to the embodiment. Figure 4 shows the end user 10 making the call and the connection destination provided for each request. <1> , destination <2> , destination <3> ...an example is shown where the two are connected via voice communication.

[0030] The "Total Call Volume" shown in (1-1) or (1-2) of Figure 4 represents the total amount of voice calls used per unit of time in business activities (the total number of calls made to the target destination). The Total Call Volume is the sum of "Completed Call Volume (2-2)" and "Incomplete Call Volume (Figure 4 (3-1))".

[0031] The "completed call volume" shown in (2-1) or (2-2) of Figure 4 refers to information indicating the amount of voice calls used in business activities that successfully connected to the target destination identified by the above telephone number (the number of calls that successfully connected to the target destination).

[0032] The "incomplete call volume" shown in (3-1) of Figure 4 refers to the number of calls that could not be connected to the target destination. Incomplete call volume is the sum of "out-of-hours call volume," "out-of-area call volume," "mid-call dropout volume," "busy call volume," and "other."

[0033] Figure 4 (4-1) or (4-2) shows "After-hours call volume," which is the number of calls made outside of the voice communication reception hours. Figure 4 (5-1) or (5-2) shows "Out-of-area call volume," which is the number of calls from outside the designated area. Figure 4 (6-1) or (6-2) shows "Mid-call dropout volume," which is the number of calls that were dropped midway through the call. Figure 4 (7-1) or (7-2) shows "Busy call volume," which is the number of calls for which there were no available lines in the subsequent stages. Figure 4 (8-1) shows "Other," which is the number of calls for which connections could not be made for reasons other than those mentioned above.

[0034] In addition, call volume data includes average call duration. "Average call duration" refers to information indicating the average duration of a single voice call used in business activities.

[0035] Returning to Figure 3, let's continue the explanation. The "call volume analysis results" are the results of the call volume analysis performed by the extraction unit 133 described later. These include information such as the total call volume, completed call volume, incomplete call volume, etc., aggregated based on predetermined classifications, and results such as call duration, presented in the form of a table or graph.

[0036] Here, an example of the "call volume analysis results" stored in the usage information DB121 will be explained using Figure 5. Figure 5 is a diagram showing an example of the call volume analysis results according to the embodiment. As shown in Figure 5, the usage information DB121 stores analysis results such as bar graphs (Figure 5(1)), pie charts (Figure 5(2)), and lists (Figure 5(3)) related to the aggregated call volume. Note that the display content and display format of the bar graphs, pie charts, and lists described above are merely examples, and the call volume analysis results stored in the usage information DB121 are not particularly limited.

[0037] In addition, the "Call Volume Analysis Results" include call volume data that meets pre-set conditions as a result of the call volume analysis. Specifically, the usage information DB121 stores call volume data that meets conditions such as "many abandoned calls + high abandoned call rate", "many busy calls + high busy encounter rate", "many after-hours calls + high after-hours rate", "long average call duration", and "(although the data appears normal) high call volume" as call volume problem data.

[0038] Returning to Figure 3, let's continue the explanation. The "issue" is an issue concerning the usage status of voice communications, which is extracted based on call volume data by the extraction unit 133 described later, and stores the issue corresponding to the call volume issue data.

[0039] For example, the usage information DB121 stores the issue "Many people drop out because the guidance is complex and difficult to understand" which corresponds to the call volume issue data "Many calls are abandoned midway + high call abandonment rate". The usage information DB121 also stores the issue "Insufficient number of lines, insufficient operator work" which corresponds to the call volume issue data "Many calls are busy + high busy encounter rate". The usage information DB121 also stores the issue "High demand for after-hours service" which corresponds to the call volume issue data "Many after-hours calls + high after-hours rate". The usage information DB121 also stores the issue "Operators are overworked, customer harassment is occurring" which corresponds to the call volume issue data "Long average call duration".

[0040] (Business activity information DB122) Returning to Figure 2, let's continue the explanation. The Business Activity Information DB122 is a database that stores information about the user's business activities (business activity information), including information about the user's organization (organizational information) and information about the user's work content (work content).

[0041] For example, the business activity information DB122 contains information such as the organization's overview, business policies, organizational philosophy, and areas of focus in business activities (e.g., cost reduction, customer satisfaction, sales development, etc.) related to the user. In addition, the business activity information DB122 stores information about the business content performed at the organization to which the voice communication is connected, and information about the type of connection method used for the voice communication at that destination, as business content.

[0042] Here, an example of business information among the business activity information stored in the business activity information DB122 will be explained using Figure 6. Figure 6 is a table diagram showing an example of business information according to the embodiment. As shown in Figure 6, the business activity information DB122 stores information related to each of the following items in a table format, such as "No," which is information that identifies individual data included in the business information, "Business," which indicates the content of the business performed in the organization set as the connection destination for voice communication, and "Business analysis result," which indicates the result of a predetermined analysis performed on the business.

[0043] For example, as shown in Figure 6, the business activity information DB122 stores the following in association: task No. "1" "Handling tasks related to login methods and operations on the XX page", task No. "1-1" which is under task No. "1" "Regarding authentication methods", and task No. "1-2" which is under task No. "1" "Other".

[0044] Furthermore, the Business Activity Information DB122 stores either "Standard / Non-Standard" or "Non-Standard" as the business analysis result for each business operation. It also stores either "Required" or "Not Required" for "Identity Verification," indicating whether identity verification is necessary. Finally, it stores either "Yes" or "No" for "Handling of Personal Information," indicating whether personal information is handled.

[0045] In other words, the business activity information DB122 stores information as business information, summarizing the content of each task performed at the destination connected via voice call.

[0046] (Solution candidate DB123) The Solution Candidate DB123 is a database that stores solution candidates used in the problem extraction process by the extraction unit 133 described later. Specifically, the Solution Candidate DB123 stores information such as hypothetical problems regarding the usage status of voice communication (hypothetical problems) and candidate solutions to the problems corresponding to those problems (solution candidates).

[0047] Here, an example of a solution candidate stored in the solution candidate DB123 will be explained using Figure 7. Figure 7 is a diagram showing an example of a solution candidate according to the embodiment. As shown in Figure 7, the solution candidate DB123 stores information related to each item, such as "No," which is information that identifies the individual data included in the solution candidate, "Hypothetical Problem," and "Proposed Solution," in a table format or the like.

[0048] For example, as shown in Figure 7, Solution Candidate DB123 is identified by No. "1" as the hypothetical problem "complex guidance / high dropout rate" and the proposed solution " <1> Connection guidance corrected. <2> It remembers the "requirements for AI automated response". In the above proposed solution, if the call volume analysis result is calculated as "many abandoned calls + high abandoned call rate" based on usage information, the hypothesized problem is that "many people drop out because the guidance is complex / difficult to understand", so <1> Connection guidance corrected. <2> This means that "AI automated response requirement distribution" is being proposed as a solution. Note that explanations for the solution candidates identified by No. "2 to 5" will be omitted.

[0049] (Talk script information DB124) The Talk Script Information DB124 is a database that stores information (talk script information) related to talk scripts generated by the generation unit 134, which will be described later. Specifically, the Talk Script Information DB124 stores information such as the name of the target user who will use the talk script (identification information), the generated talk script (talk script), the results of discussions regarding improvements to the generated talk script and improvement information (improvement information), and the improved talk script based on the discussion information (improved talk script). Note that the "identification information" may be the same information as the "identification information" stored in the Usage Status Information DB121, as explained using Figure 3.

[0050] Here, an example of talk script information stored in the talk script information DB124 will be explained using Figure 8. Figure 8 is a table diagram showing an example of talk script information according to the embodiment. The talk script information DB124 stores information related to each of the following items in a table format, associating it with "No," which is information that identifies individual data included in the talk script information, "identification information," "talk script," "improvement information," and "improvement talk script."

[0051] For example, as shown in Figure 8, the talk script information DB124 stores identification information "A", talk script "F", improvement information "G", and improvement talk script "H", all identified by No. "1". The letters "A, and F through H" listed for each item in the table diagram shown in Figure 8 are legends for the information contained in each item.

[0052] (Generative model DB125) The generative model DB125 is a database that stores predetermined models used by the generation unit 134 (described later) to generate solutions and talk scripts. For example, the generative model DB125 can store large-scale language models as generative models.

[0053] Specifically, the generation device 100 according to this embodiment can use at least one of the following as a large-scale language model: "ChatGPT®", a large-scale language model possessing general-purpose knowledge, and "tsuzumi®", a predetermined large-scale language model on which adapter tuning is performed (see, for example, References 1 and 2).

[0054] (Reference 1):ChatGPT(OpenAI),<URL:https: / / openai.com / chatgpt> ,<Searched on September 2, 2020> (Reference 2): NTT version of large-scale language model "tsuzumi",<URL:https: / / www.rd.ntt / research / LLM_tsuzumi.html> ,<Searched on September 2, 2020>

[0055] (Control unit 130) Now, let's return to Figure 2 and continue the explanation. The control unit 130 has an internal memory for temporarily storing programs and processing data that define various processing procedures of the generation device 100, and is realized by electronic circuits such as a CPU (Central Processing Unit) and an MPU (Micro Processing Unit), and integrated circuits such as an ASIC (Application Specific Integrated Circuit) and an FPGA (Field Programmable Gate Array). As shown in Figure 2, the control unit 130 has a collection unit 131, a receiving unit 132, an extraction unit 133, a generation unit 134, and an output unit 135.

[0056] (Collection Section 131) The collection unit 131 collects information that the generation device 100 uses to perform predetermined processing. The collection unit 131 then stores the collected information in the storage unit 120.

[0057] Specifically, the collection unit 131 collects the target call volume data from an information processing device or the like that has a function to store call volume data related to the user's voice communication usage status. The collection unit 131 then stores the collected call volume data in the usage status information DB 121.

[0058] In addition, the collection unit 131 collects information regarding discussions on improvements made to the generated solutions and talk scripts (hereinafter sometimes referred to as "discussion information") and stores it in the talk script information DB 124 as information regarding improvements to the solutions and talk scripts (hereinafter sometimes referred to as "improvement information"). For example, the collection unit 131 collects log data from phone calls, emails, chats, video conferences, etc., in which discussions about the generated solutions and talk scripts took place, as discussion information. Next, the collection unit 131 extracts improvement information from the collected log data. Then, the collection unit 131 stores the collected improvement information in the talk script information DB 124.

[0059] (Reception desk 132) The reception unit 132 receives commands from the user via the display unit, input unit, and communication unit 110 described above, regarding the extraction of issues, solutions, and generation of talk scripts.

[0060] (Extraction part 133) The extraction unit 133 extracts issues from the call volume data that meet predetermined conditions. For example, the extraction unit 133 extracts issues in the following manner.

[0061] The extraction unit 133 retrieves usage information stored in the usage information DB 121. Next, the extraction unit 133 uses the retrieved usage information to perform call volume analysis, including numerical aggregation and predetermined statistical processing, based on known techniques.

[0062] The extraction unit 133 extracts call volume problem data from the call volume data calculated by call volume analysis that meets conditions such as "many abandoned calls + high abandoned call rate", "many busy calls + high busy encounter rate", "many after-hours calls + high after-hours rate", "long average call duration", and "(although the data doesn't seem to be a problem) high call volume". The extraction unit 133 then stores the extracted call volume problem data as "call volume problem data" in the solution candidate DB 123.

[0063] Next, the extraction unit 133 extracts issues using the extracted call volume issue data. The extraction unit 133 also stores the extracted issues as "issues" in the solution candidate DB 123.

[0064] (Generation unit 134) The generation unit 134 inputs the issues extracted from the call volume data of the user's business activities by the extraction unit 133, the user's business activity information, and prompts including commands to generate solutions and talk scripts corresponding to those solutions to the large-scale language model, thereby generating solutions and talk scripts. Specifically, the generation unit 134 uses the prompts described below to cause the large-scale language model to acquire predetermined information and generate solutions and talk scripts.

[0065] The above prompt includes a command to cause the large-scale language model to acquire as prior knowledge a hypothetical problem associated with call volume data and proposed solutions (solutions) corresponding to that hypothetical problem. The above prompt also includes a command to cause the large-scale language model to acquire as prior knowledge at least one of the following: information on the user's voice communication usage, which is acquired as user business activity information, and information on the user's business activities (job content, organizational information).

[0066] The generation unit 134 uses the above prompt to cause the large-scale language model to retrieve the hypothetical problems and proposed solutions stored in the solution candidate DB 123, and inputs the retrieved hypothetical problems and proposed solutions into the large-scale language model as prior knowledge.

[0067] Furthermore, the generation unit 134 uses the above prompt to cause the large-scale language model to acquire business activity information, including business information and organizational information, from the user's organization's homepage or a server device that displays the homepage. The generation unit 134 then stores the business activity information acquired by the large-scale language model in the business activity information DB 122.

[0068] Furthermore, the generation unit 134 uses the above prompts to identify proposed solutions corresponding to the hypothetical problems based on the extracted problems, and generates a solution using the identified proposed solutions. The above prompts include instructions to generate a talk script in which the generated solution is expressed based on a predetermined story.

[0069] (Explanation of an example prompt) Here, an example of a prompt to be input to the large-scale language model in this embodiment will be explained using Figure 9. Figure 9 is a diagram showing an example of a prompt according to the embodiment. Figure 9 shows an example of a prompt to cause the large-scale language model to generate a solution and talk script corresponding to the problem based on the results of the problem and call volume analysis, business information and organizational information.

[0070] The generation unit 134 inputs the "prompt" shown in Figure 9(1) into the large-scale language model to acquire prior knowledge and generate solutions and talk scripts. The prompt shown in Figure 9(1) includes the following:

[0071] As shown in Figure 9 (1-1), the prompt includes the role definition "You are a salesperson for the automated telephone response system." By using the prompt that includes the "role definition" as described above, the generation unit 134 can generate solutions and talk scripts in the large-scale language model that are tailored to the areas and perspectives desired by the person in charge.

[0072] As shown in (1-2) of Figure 9, the prompt includes the processing instruction "Execute the task according to the following constraints." The generation unit 134, by constraining the content of the solution and talk script to be generated using the prompt which includes the "processing instruction based on processing constraints" as described above, can cause the large-scale language model to generate the solution and talk script desired by the person in charge.

[0073] In the example shown in Figure 9, the prompt includes the first to third tasks (from (1-3) to (1-5) in Figure 9).

[0074] For example, the first task includes the instruction, "Understand the business operations and organizational information," and the constraints, "Obtain business and organizational information for company XX, and read the information regarding the business operations in detail. In particular, focus on understanding the policies for carrying out operations and what kind of work is done. Then summarize the following," and "The operational policies and approaches that the organization has for the departments that carry out operations," and "The business operations that the departments that carry out operations are responsible for" (Figure 9 (1-3)).

[0075] The first task allows the generation unit 134 to obtain "business content and organizational information" from an external information processing device or the like for the large-scale language model. The generation unit 134 also uses the obtained "business content and organizational information" as prior knowledge to perform pre-training, such as "making the model understand business policies (policies for carrying out business) and business content (what kind of work is being done)." Based on the first task, the generation unit 134 generates "operational policies and approaches" and "business content" to be included in the solution and talk script.

[0076] Furthermore, the second task includes the instruction "Understand the call volume analysis results and challenges" and the constraint "Obtain the call volume analysis results and challenges of company XX, and understand the challenges of company XX" (Figure 9 (1-4)).

[0077] The second task allows the generation unit 134 to obtain the "call volume analysis results for the target company (Company XX)" from the usage information DB 121 for the large-scale language model. The generation unit 134 can also obtain the "challenges of the target company (Company XX)" from the usage information DB 121 for the large-scale language model. Furthermore, the generation unit 134 can use the obtained "call volume analysis results and challenges" as prior knowledge to perform pre-training such as "making the model understand the challenges of Company XX".

[0078] Furthermore, the third task includes the instruction to "generate solutions and talk scripts" and the constraints to "create solutions tailored to the challenges of company XX and talk scripts for proposing those solutions to company XX. When doing so, create solutions and talk scripts that take into account the operational policies and approaches, business content, challenges, and call volume analysis results" (Figure 9 (1-5)).

[0079] For example, the third task allows the generation unit 134 to generate "solutions for the target company (Company XX)" based on Company XX's operational policies and approaches, business content, challenges, and call volume analysis results for the pre-trained large-scale language model. Furthermore, the third task allows the generation unit 134 to generate "talk scripts for the target company (Company XX)" based on Company XX's operational policies and approaches, business content, challenges, call volume analysis results, and the generated solutions for the pre-trained large-scale language model.

[0080] (Explanation of a solution and an example of a talk script) Here, an example of a solution and talk script generated by the generation device 100 will be explained using Figure 10. Figure 10 is a diagram showing an example of the output of a solution and talk script according to the embodiment. Figure 10 shows an example of a solution and talk script generated based on an example of prompts shown in Figure 9.

[0081] As described above, the generation unit 134 generates solutions and talk scripts that are tailored to the user's (XX company's) challenges. For example, as shown in Figure 10 (1), the generation unit 134 generates "solutions <1> (Figure 10 (1-1)) "Solution" <2> One or more solutions are generated, such as (Figure 10 (1-2)). A solution is information that shows methods for solving the extracted problem, etc., in the form of text, figures, symbols, etc.

[0082] Furthermore, as shown in (2) of Figure 10, the generation unit 134 generates a talk script that includes "Operating Policy and Approach" and "Proposal Flow." As shown in (2-1) of Figure 10, "Operating Policy and Approach" includes summary information such as the policies and initiatives to achieve the mission related to the organizational management of the target organization (Company XX).

[0083] The "proposal flow" is information that shows the flow of the conversation when proposing a solution to a user, and includes, for example, "1. Opening," "2. Problem identification (issue)," "3. Proposal of solution," "4. Emphasis on benefits," and "5. Sharing of implementation examples" (Figure 10 (2-2)).

[0084] The above "1. Opening" includes, for example, a brief overview of what will be proposed and icebreaker content before explaining the solution to the user. "2. Problem Statement (Issue)" includes, for example, a talk explaining what kind of issues the user is facing in the voice communication used in their business activities. "3. Proposal of Solution" includes, for example, a talk introducing means and solutions to solve the issues explained in "2. Problem Statement (Issue)". "4. Emphasis on Benefits" includes, for example, a talk explaining the benefits of the methods and solutions for solving the issues explained in "3. Proposal of Solution". "5. Sharing of Implementation Examples" includes, for example, a talk explaining actual implementation examples of the methods and solutions for solving the issues explained in "3. Proposal of Solution".

[0085] Please note that the solution shown in Figure 10(1) and the content of the talk script shown in Figure 10(2) are merely examples and are not limited to the above.

[0086] (An example of a solution and improvement to the talk script) Furthermore, the generation unit 134 can perform processes to improve the solutions and talk scripts generated by the processes described using Figures 9 and 10. Specifically, the collection unit 131 collects improvement information regarding the solutions and talk scripts generated by the generation unit 134 from the person in charge of proposing the solutions and talk scripts.

[0087] Next, the generation unit 134 uses the improvement information for the solution and talk script to input prompts, expressed in natural language text, that instruct the generated solution and talk script to be improved, into the large-scale language model, thereby generating the improved solution and talk script.

[0088] For example, the generation unit 134 generates a summary such as "reduce the number of characters in the talk script" from conversation data (discussion information included in improvement information) stored in the talk script information DB 124, such as "The generated talk script has too many characters, so I want to reduce it." Then, the generation unit 134 inputs a prompt such as "Reduce the number of characters in the generated talk script and generate the talk script again" to the large-scale language model and generates a talk script with a reduced number of characters (i.e., an "improved" talk script).

[0089] (Output section 135) Now, let's return to Figure 2 and continue the explanation. The output unit 135 outputs the solutions and talk scripts generated by the generation unit 134 to the terminal device operated by the user via the communication unit 110, etc.

[0090] (Processing procedure by the generating device 100) From here, the processing procedure realized by the generation device 100 according to this embodiment will be explained with reference to Figure 11. Figure 11 is a flowchart showing the processing by the generation device 100 according to this embodiment.

[0091] The collection unit 131 collects usage information, including call volume data (S101). Next, the extraction unit 133 performs call volume analysis using the call volume data (S102). Next, the extraction unit 133 extracts issues based on the results of the call volume analysis (S103). Next, the generation unit 134 generates solutions to the issues and talk scripts based on the extracted issues and business activity information obtained by the large-scale language model (S104).

[0092] If improvements are made to the generated solution and talk script (Yes in S105), the collection unit 131 collects improvement information (S106). Next, the generation unit 134 generates the improved solution and talk script using the collected improvement information (S107). Then, the output unit 135 outputs the generated solution and talk script (S108). After that, the generation device 100 completes the process.

[0093] On the other hand, if the improvements to the generated solution and talk script are not implemented (No. in S105), the generation device 100 skips steps S106 and S107. Then, the output unit 135 outputs the generated solution and talk script (S108). After that, the generation device 100 terminates the process.

[0094] (effect) The effects of the generation device 100 according to this embodiment will now be explained. Conventionally, in order to address the decline in customer satisfaction when voice communication via telephone networks is used in business activities, analysis results of call status data for each unit of time are sometimes generated. However, merely outputting the analysis results makes it difficult to generate talk scripts, etc., to appropriately propose solutions, and there are challenges in appropriately realizing improvements in voice communication.

[0095] Therefore, the extraction unit 133 of the generation device 100 according to this embodiment extracts issues regarding the usage status of voice communication by users who use voice communication via the telephone network for their business activities. The generation unit 134 of the generation device 100 inputs prompts to a large-scale language model, which include commands to generate solutions corresponding to the issues and talk scripts in which the solutions are expressed based on a predetermined story, using the extracted issues and the acquired user business activity information, thereby generating solutions and talk scripts.

[0096] Through the process described above, the generation device 100 can automatically generate solutions and talk scripts, whereas conventionally, solutions and talk scripts were created manually based on analysis results using information on the usage status of voice communication, such as usage status information. Therefore, the generation device 100 according to this embodiment has the effect of appropriately realizing improvements to voice communication using the generated solutions and talk scripts.

[0097] Furthermore, the generation apparatus 100 according to this embodiment achieves predetermined effects by performing the processes described below.

[0098] The extraction unit 133 extracts issues from call volume data related to the use of voice communication in the user's business activities that meet predetermined conditions. The generation unit 134 inputs the issues extracted from the call volume data related to the use of voice communication in the user's business activities, the user's business activity information, and prompts including commands to generate solutions and talk scripts corresponding to those solutions into a large-scale language model to generate solutions and talk scripts.

[0099] Through the process described above, the generation device 100 can extract issues according to the user's usage of voice communication and generate solutions and talk scripts corresponding to the extracted issues. Therefore, the generation device 100 has the effect of being able to generate user-specific solutions more efficiently than before.

[0100] Furthermore, the generation device 100 can generate talk scripts to appropriately propose solutions to the user. Therefore, the generation device 100 not only generates and outputs solutions, but also has the effect of making it easier for the user to consider solutions to resolve issues related to the usage of voice communication.

[0101] The generation unit 134 uses prompts expressed in natural language text that include commands to cause a large-scale language model to acquire as prior knowledge a hypothetical problem associated with call volume data related to the use of voice communication in the user's business activities, and proposed solutions corresponding to the hypothetical problem, and commands to cause the large-scale language model to acquire as prior knowledge at least one of the following: information on the usage status of voice communication in the user's business activities, and information on the user's business activities (such as business content and organizational information), which have been acquired as information on the user's business activities.

[0102] The generation device 100 can use the prompts described above to cause the large-scale language model to acquire information for generating solutions and talk scripts from an external information processing device or the like. As a result, the generation device 100 can efficiently generate solutions and talk scripts by eliminating the need for manual acquisition of information for generating solutions and talk scripts.

[0103] The generation unit 134 uses information regarding improvements to the solution and talk script to input prompts, expressed in natural language text, that instruct the generation of the solution and talk script to improve it, into a large-scale language model to generate the improved solution and improved talk script.

[0104] Through the process described above, the generation device 100 is able to improve the solutions and talk scripts output to the person in charge so that they become more appropriate solutions and talk scripts. By generating improved solutions and talk scripts, the generation device 100 has the effect of being able to propose more appropriate solutions and talk scripts to the user.

[0105] The collection unit 131 collects information regarding discussions on improvements made to the generated solutions and talk scripts, and stores it as information on improvements to the solutions and talk scripts.

[0106] Through the process described above, the generation device 100 automatically collects information related to improvement discussions between personnel, such as chats, emails, voice calls, and video conferences, which are used to improve solutions and talk scripts, and makes this information available for use in improving solutions and talk scripts. As a result, the generation device 100 has the effect of enabling more efficient improvement of solutions and talk scripts.

[0107] <Variation> The following describes modifications that can be implemented by the generation apparatus 100 according to this embodiment.

[0108] (Data, etc.) The voice communication usage status, usage status information, business activity information, call volume analysis (results), business information, organizational information, solutions, talk scripts, names of the functional parts of the generation device 100, steps, processes, names of steps or processes, etc., used in the description of the above embodiment are merely examples and can be changed at will.

[0109] For example, while it was explained that the usage information DB121 stores information related to each of the following items in a table format, such as "No," which is information that identifies individual data included in user information, "identification information," "telephone number," "call volume data," "call volume analysis results," and "issue," the items and types of information stored are not limited to what has been explained. While it was explained that it stores information related to the user's business activities (business activity information), including information related to the user's organization (organization information) and information related to the user's work content (work content), the items and types of information stored are not limited to what has been explained. While it was explained that the solution candidate DB123 stores information related to each of the following items in a table format, such as "No," which is information that identifies individual data included in the solution candidate, "hypothetical issue," and "proposed solution," the items and types of information stored are not limited to what has been explained. While it was explained that the Talk Script Information DB124 stores information related to each item in the Talk Script Information, such as "No." (which identifies individual data items included in the Talk Script Information), "User Name," "Talk Script," "Improvement Information," and "Improvement Talk Script," in a table format, the items and types of information stored are not limited to those described.

[0110] (Regarding the timing of pre-training large-scale language models) In this embodiment, it has been explained that the generation unit 134 causes the large-scale language model to acquire prior knowledge when it performs the generation process of solutions and talk scripts, but the timing of learning is not limited to this. For example, the generation unit 134 can acquire prior knowledge in advance and perform pre-learning before performing the generation process of solutions and talk scripts.

[0111] (Regarding the use of generative models) In this embodiment, the model (large-scale language model) used by the generation device 100 is described as being stored in the generation model DB 125 of the storage unit 120, but this is not limited to this. For example, the generation device 100 can access an external information processing device (server, etc.) and use a predetermined model.

[0112] (Flowcharts, etc.) In flowcharts, each step may be rearranged as long as it does not create inconsistencies, and some steps may be omitted. Furthermore, conjunctions such as "next," "continue," "in addition," "at this time," and "on this occasion" in flowchart descriptions do not limit the order or timing of the processes in the flowchart.

[0113] <Hardware Configuration> Each component of the illustrated device is a functional concept and does not necessarily have to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions. Furthermore, each processing function performed by each device can be implemented, all or any part of it, by a CPU and the program that is analyzed and executed by that CPU, or by hardware using wired logic.

[0114] Furthermore, among the processes described in this embodiment, all or part of those described as being performed automatically can be performed manually using known methods. In addition, the processing procedures, control procedures, specific names, and information including various data and parameters shown in the drawings can be arbitrarily changed unless otherwise specified.

[0115] <Program> In one embodiment, the various devices constituting the generation device 100 can be implemented by installing the generation program as packaged software or online software on a desired computer. For example, by having the above generation program executed on an information processing device, the various devices constituting the generation device 100 can be made to function. The information processing device referred to here includes desktop or notebook personal computers. In addition, the information processing device also includes mobile communication terminals such as smartphones and mobile phones, and slate terminals such as PDAs (Personal Digital Assistants).

[0116] Figure 12 shows an example of a computer that implements the generation device 100 according to the embodiment. The computer 1000 has, for example, memory 1010 and CPU 1020. The computer 1000 also has a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

[0117] Memory 1010 includes ROM (Read Only Memory) 1011 and RAM 1012. ROM 1011 stores, for example, a boot program such as BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1090. The disk drive interface 1040 is connected to the disk drive 1100. For example, a removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to, for example, a mouse 1110 and a keyboard 1120. The video adapter 1060 is connected to, for example, a display 1130.

[0118] The hard disk drive 1090 stores, for example, an OS (Operating System) 1091, an application program 1092, a program module 1093, and program data 1094. That is, the programs that define the various processes of the various devices constituting the generation device 100 are implemented as program modules 1093 in which executable code for a computer is written. The program modules 1093 are stored, for example, in the hard disk drive 1090. For example, a program module 1093 for performing processes similar to the functional configuration of the various devices constituting the generation device 100 is stored in the hard disk drive 1090. Note that the hard disk drive 1090 may be replaced by an SSD (Solid State Drive).

[0119] Furthermore, the configuration data used in the processing of the embodiment described above is stored as program data 1094 in, for example, memory 1010 or hard disk drive 1090. The CPU 1020 then reads the program module 1093 and program data 1094 stored in memory 1010 or hard disk drive 1090 into RAM 1012 as needed and executes the processing of the embodiment described above.

[0120] Furthermore, the program module 1093 and program data 1094 are not limited to being stored in the hard disk drive 1090; for example, they may be stored in a removable storage medium and read by the CPU 1020 via a disk drive 1100 or the like. Alternatively, the program module 1093 and program data 1094 may be stored in another computer connected via a network (LAN, WAN (Wide Area Network), etc.). The program module 1093 and program data 1094 may then be read from the other computer by the CPU 1020 via a network interface 1070.

[0121] <Other> Although this embodiment has been described above, this embodiment is not limited by the description and drawings that constitute part of the disclosure. That is, all other embodiments, examples, and operational techniques made by those skilled in the art based on this embodiment are included in the scope of this embodiment. [Explanation of Symbols]

[0122] 100 generator 110 Communications Department 120 Storage section 121 Usage Information Database 122 Business activity information DB 123 Solution candidate DB 124 Talk Script Information Database 125 Generative Model DB 130 Control Unit 131 Collection Department 132 Reception Department 133 Extraction part 134 Generation part 135 Output section

Claims

1. An extraction unit that identifies issues regarding the usage of voice communications by users who utilize voice communications via the telephone network for business activities, A generation unit that inputs prompts into a large-scale language model, including commands to generate a solution corresponding to the problem and a talk script in which the solution is expressed based on a predetermined story, using the extracted problem and the acquired information on the user's business activities, to generate the solution and the talk script. A generating apparatus characterized by having the following features.

2. The extraction unit is From the data relating to call volume in the business activities of the aforementioned user, the aforementioned issues are extracted from data that meets predetermined conditions. The generating unit is The problem extracted from call volume data related to the use of voice communication in the user's business activities, information about the user's business activities, and the prompt including a command to generate the solution and the corresponding talk script are input to a large-scale language model to generate the solution and the talk script. The generating apparatus according to feature 1.

3. The generating unit is A command to cause the large-scale language model to acquire, as prior knowledge, a hypothetical problem associated with call volume data related to the use of voice communication in the user's business activities, and proposed solutions to the hypothetical problem; A command to cause the large-scale language model to acquire, as prior knowledge, at least one of the following: information regarding the use of voice communication in the user's business activities, and information regarding the user's business activities, which has been acquired as information regarding the user's business activities. This uses prompts expressed in natural language text. The generating apparatus according to feature 2.

4. The generating unit is Using information regarding improvements to the solutions and talk scripts, prompts expressing the generated instructions for improving the solutions and talk scripts in natural language text are input to the large-scale language model. Generate improved solutions and improved talk scripts. The generating apparatus according to feature 2 or 3.

5. The system further includes a collection unit that collects information regarding discussions on improvements made to the generated solutions and talk scripts, and stores this information as information on improvements to the solutions and talk scripts. The generating apparatus according to feature 4.

6. A generation method to be executed by a generation device, An extraction process to identify issues regarding the usage of voice communications by users who utilize voice communications via the telephone network for business activities, A generation step of inputting prompts into a large-scale language model, which include commands to generate a solution corresponding to the problem and a talk script in which the solution is expressed based on a predetermined story, using the extracted problem and the acquired information on the user's business activities, to generate the solution and the talk script, A method for generating a product, characterized by including the following:

7. An extraction step to identify issues regarding the usage of voice communications by users who utilize voice communications via the telephone network for business activities, A generation step involves inputting prompts into a large-scale language model, which include commands to generate a solution corresponding to the problem and a talk script in which the solution is expressed based on a predetermined story, using the extracted problem and the acquired information on the user's business activities, thereby generating the solution and the talk script. A generation program that causes a computer to execute something.

Images