Information processing apparatus, information processing system, information processing method, and computer program product

Abstract

The present invention relates to an information processing apparatus, an information processing system, an information processing method, and a computer program product. An object of the present invention is to enable production of a document containing information required by a user in an information processing system that produces a document based on input data. An information processing apparatus of the present invention produces a document based on input data, the information processing apparatus including: an acquisition section that acquires setting information related to production of the document; a document production section that produces the document using a large-scale language model based on the input data and the setting information; and an output section that outputs the document produced by the document production section.

Description

Technical Field

[0001] This disclosure relates to information processing apparatus, information processing system, information processing method, and computer program product. Background Technology

[0002] For example, there exist information processing systems for creating documents such as daily reports or meeting minutes. For instance, a document creation apparatus is known that acquires the action logs of a monitored terminal used in business operations, uses classification results determined from the logs to select an appropriate template, and generates prompts instructing a large-scale language model to create the document (see, for example, Patent Document 1).

[0003] [Existing Technical Documents] [Patent Documents] [Patent Document 1] Japanese Patent No. 7572760 Summary of the Invention In the technology disclosed in Patent Document 1, a template is selected from a pre-prepared prompt template via conditional branching. However, in this method, it is not possible to add, modify, or delete items defined in the template, so sometimes the generated document does not contain the information required by the user.

[0004] One embodiment of this disclosure addresses the aforementioned problems, enabling the creation of documents containing information required by the user within an information processing system that generates documents based on input data.

[0005] To address the aforementioned issues, one embodiment of this disclosure relates to an information processing apparatus that generates documents based on input data. The apparatus comprises: an acquisition unit for acquiring setting information related to the generation of the document; a document generation unit for generating the document using a large-scale language model based on the input data and the setting information; and an output unit for outputting the document generated by the document generation unit.

[0006] The effects of the present invention are explained below: According to one embodiment of this disclosure, in an information processing system that generates documents based on input data, it is possible to generate documents containing information required by the user. Attached Figure Description

[0007] Figure 1 This is a diagram illustrating an example of the system architecture of an information processing system involved in one implementation.

[0008] Figure 2 This is a diagram illustrating an example of the hardware structure of a computer in one embodiment.

[0009] Figure 3This is a diagram illustrating an example of the hardware structure of an information terminal involved in one implementation.

[0010] Figure 4 This is a diagram illustrating an example of the functional configuration of the information processing system involved in the first embodiment.

[0011] Figure 5 This is a diagram illustrating an example of user information involved in the first implementation.

[0012] Figures 6A-6C This is a diagram illustrating an example of information contained in the template involved in the first implementation.

[0013] Figure 7 Figure (1) shows an example of information reflecting personal settings involved in the first implementation.

[0014] Figure 8 Figure (2) shows an example of information reflecting personal settings involved in the first implementation.

[0015] Figure 9 This is a timing diagram (1) illustrating an example of the processing of the information processing system involved in the first embodiment.

[0016] Figure 10 This is a timing diagram (2) illustrating an example of the processing of the information processing system involved in the first implementation.

[0017] Figure 11 Figure (1) shows an example of a display screen involved in the first embodiment.

[0018] Figure 12 Figure (2) shows an example of a display screen involved in the first embodiment.

[0019] Figure 13 Figure (3) shows an example of a display screen involved in the first embodiment.

[0020] Figure 14 This is a diagram showing an example of the list of available templates involved in the first implementation.

[0021] Figure 15 This is a flowchart illustrating an example of a prompt generation process involved in one implementation.

[0022] Figure 16 Figure (1) is an example of a function that generates partial hints in an implementation.

[0023] Figure 17 Figure (2) is an example of a function that generates partial hints in an implementation.

[0024] Figure 18This is a diagram illustrating an example of an execution generation component prompt related to an implementation mode.

[0025] Figure 19 This is a diagram illustrating an example of the execution combination hints involved in an implementation mode.

[0026] Figure 20 This is a diagram illustrating an example of a generated prompt related to an implementation.

[0027] Figure 21 This is a timing diagram (1) illustrating an example of the processing of an information processing system involved in the second implementation.

[0028] Figure 22 This is a timing diagram (2) illustrating an example of the processing of the information processing system involved in the second implementation.

[0029] Figure 23 This is a diagram illustrating an example of the system architecture of the information processing system involved in the third implementation.

[0030] Figure 24 This is a diagram illustrating an example of the functional configuration of the information processing system involved in the third implementation.

[0031] Figure 25 This is a sequence diagram representing an example of the processing preceding the start of the dialogue involved in the third implementation.

[0032] Figure 26 This is a diagram illustrating an example of user information involved in the third implementation.

[0033] Figure 27 This is a sequence diagram illustrating an example of the processing of an information processing system involved in the third implementation.

[0034] Figure 28 Figure (1) shows an example of a display screen involved in the third embodiment.

[0035] Figure 29 Figure (2) shows an example of a display screen involved in the third embodiment.

[0036] Figure 30 This is a sequence diagram illustrating an example of the daily newspaper production process involved in the third implementation.

[0037] Figure 31 This is a diagram illustrating an example of a daily newsletter in the third implementation.

[0038] Figure 32 This is a sequence diagram illustrating an example of the daily report setting process involved in the third implementation.

[0039] Figure 33Figure (3) shows an example of a display screen involved in the third embodiment.

[0040] Figure 34 Figure (4) shows an example of a display screen involved in the third embodiment.

[0041] Figure 35 Figure (5) shows an example of a display screen involved in the third embodiment.

[0042] The symbols in the diagram are explained as follows: 1: Information Processing System 10: User Terminal 100: Information processing device 113: Large-scale language models 200: Computer 404: Acquisition Department 405: Prompt for Generation Department 406: Document Production Department 407: Output Section 409: Management Department 2300: Dialogue System 2412: Dialogue AI Control Department 2401: Dialogue Engine Specific Implementation Form Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0043] <System Structure> Figure 1 This is a diagram illustrating an example of the system architecture of an information processing system in one implementation. Figure 1 In the example, the information processing system 1 includes, for instance, a user terminal 10 and an information processing device 100 connected to a communication network N such as the Internet or a LAN (Local Area Network). Furthermore, the user terminal 10 and the information processing device 100 can communicate with external systems 110 (such as a voice recognition engine 111, a database 112, a large-scale language model 113, and a daily report system 114, etc.) via the communication network N.

[0044] User terminal 10 is an information terminal used by user 11, such as a PC (Personal Computer), tablet, smartphone, or wearable device. User terminal 10 can communicate with information processing device 100 via communication network N. In addition, user 11 can use the services provided by information processing device 100 through user terminal 10.

[0045] Information processing apparatus 100 is an information processing apparatus with a computer structure, or a system comprising multiple computers. Furthermore, the computers constituting information processing apparatus 100 are not limited to physical machines; for example, they could be virtual machines in the cloud. Information processing apparatus 100 provides a document creation service based on input data, for example, by executing a prescribed program using the computers within information processing apparatus 100.

[0046] For example, the information processing device 100 can create documents such as daily reports or meeting minutes based on input data such as sound data or text data derived from sound data (textualized text). Alternatively, the information processing device 100 can use documents such as daily reports or meeting minutes as input data to create documents that summarize the documents or documents that compile necessary information.

[0047] <Hardware Structure> (Computer hardware architecture) The information processing device 100 and the user terminal 10, for example, have Figure 2 The hardware structure of the computer 200 is shown. The information processing device 100 may also be composed of multiple computers 200.

[0048] Figure 2 This is a diagram illustrating an example of the hardware architecture of a computer in one embodiment. For example, such as... Figure 2 As shown, the computer 200 includes a CPU (Central Processing Unit) 201, ROM (Read-Only Memory) 202, RAM (Random Access Memory) 203, HD (Hard Disk Drive) 204, HDD (Hard Disk Drive) controller 205, display 206, external device connection I / F (interface) 207, network I / F 208, keyboard 209, indicator device 210, DVD-RW (Digital Multifunction Rewritable Optical Disc) drive 212, media I / F 214, bus 215, etc.

[0049] In addition, when the computer 200 is a user terminal 10, the computer 200 also includes a microphone 221, a speaker 222, a sound input / output I / F 223, a CMOS (Complementary Metal Oxide Semiconductor) sensor 224, and an image sensor I / F 225.

[0050] The CPU 201 controls the overall operation of the computer 200. The ROM 202 stores programs such as the IPL (Initial Program Loader) used to start the computer 200. The RAM 203 serves as the working area for the CPU 201. The HD 204 stores programs such as the operating system (OS), applications, device drivers, and various data. The HDD controller 205, for example, controls the reading of various data from the HD 204 and the writing of various data to the HD 210 according to the control of the CPU 201. The HD 204 and the HDD controller 205 are examples of storage devices.

[0051] The monitor 206 displays various information, such as cursors, menus, windows, characters, and images. Alternatively, the monitor 206 can also be located externally to the computer 200. The external device connection interface 207 is an interface for connecting various external devices to the computer 200. The network interface 208 is an interface for connecting the computer 200 to a communication network N to communicate with other devices.

[0052] Keyboard 209 is an input mechanism with multiple keys for inputting text, numbers, various indicators, etc. Indicator 210 is an input device that performs various command selection and execution, target selection, cursor movement, etc. Alternatively, keyboard 209 and indicator 210 can also be located externally to computer 200.

[0053] DVD-RW drive 212 controls the reading of various data from and writing of various data to DVD-RW 211, which is an example of a removable recording medium. DVD-RW 211 is not limited to DVD-RW and can also be other removable recording media. Media I / F 214 controls the reading of data from or writing (storing) of data to (such as flash memory) media 213. Bus 215 includes an address bus, a data bus, and various control signals for electrically connecting the above-mentioned components.

[0054] Microphone 221 is a built-in circuit that converts sound into electrical signals. Speaker 222 is a built-in circuit that converts electrical signals into physical vibrations to produce music, sound, and other sounds. Sound input / output I / F 223 is a circuit that processes the input and output of sound signals between microphone 221 and speaker 222 under the control of CPU 201.

[0055] CMOS sensor 224 is a built-in imaging device that, under the control of CPU 201, captures images of an object (e.g., its own image) to obtain image data. Computer 200 may have an imaging device such as a CCD (charge-coupled device) sensor instead of CMOS sensor 224. Image sensor I / F 225 is the circuitry that controls the driving of CMOS sensor 224.

[0056] (Hardware structure of information terminals) User terminal 10 may also have, for example, a Figure 3 The hardware structure of the information terminal 300 shown.

[0057] Figure 3 This is a diagram illustrating an example of the hardware structure of an information terminal involved in one implementation. Figure 3 In the example, the information terminal 300 includes a CPU 301, ROM 302, RAM 303, storage device 304, CMOS sensor 305, camera element I / F 306, acceleration / azimuth sensor 307, medium I / F 309, and GPS (Global Positioning System) receiver 310.

[0058] The CPU 301 controls the overall operation of the information terminal 300 by executing a prescribed program. The ROM 302 stores programs such as IPL for booting the CPU 301. The RAM 303 serves as the working area for the CPU 301. The storage device 304 is a large-capacity storage device that stores the OS, applications, and various data, such as an SSD (Solid State Drive) or flash ROM.

[0059] The CMOS sensor 305 is a built-in imaging device that, under the control of the CPU 301, captures images of the object (mainly its own image) to obtain image data. Alternatively, the user terminal 10 can also have a camera device such as a CCD sensor to replace the CMOS sensor 305. The image sensor I / F 306 is a circuit that controls the driving of the CMOS sensor 305. The accelerometer / azimuth sensor 307 is various sensors, such as an electromagnetic compass, rotary compass, and accelerometer for detecting geomagnetism. The medium I / F 309 controls the reading of data from or writing (storing) data to a medium (storage medium) 308, such as flash memory. The GPS receiver 310 receives GPS signals (positioning signals) from GPS satellites.

[0060] In addition, the information terminal 300 includes a long-range communication circuit 311, an antenna 311a of the long-range communication circuit 311, a CMOS sensor 312, an image sensor I / F 313, a microphone 314, a speaker 315, an audio input / output I / F 316, a display 317, an external device connection I / F 318, a short-range communication circuit 319, an antenna 319a of the short-range communication circuit 319, and a touch panel 320.

[0061] The long-distance communication circuit 311 is, for example, a circuit that communicates with other devices via a communication network N. The CMOS sensor 312 is a built-in imaging device that, under the control of the CPU 301, captures images of objects to obtain image data. The imaging element I / F 313 is a circuit that controls the driving of the CMOS sensor 312. The microphone 314 is a built-in circuit that converts sound into electrical signals. The speaker 315 is a built-in circuit that converts electrical signals into physical vibrations to produce music, sound, and other sounds. The sound input / output I / F 316 is a circuit that processes the input and output of sound wave signals between the microphone 314 and the speaker 315 according to the control of the CPU 301.

[0062] Display 317 is a type of display device such as liquid crystal or organic EL (electroluminescence) that displays images of the subject, various icons, etc. External device connection I / F 318 is an interface for connecting various external devices. Near-field communication circuit 319 includes circuitry for short-field wireless communication. Touch panel 320 is an input device that allows a user to operate the information terminal 300 by pressing the display 317. Additionally, the information terminal 300 includes a bus 321. Bus 321 includes components for electrical connections. Figure 3 The CPU301 and other components shown include the address bus, data bus, etc.

[0063] [First Implementation Form] <Functional Structure> Figure 4 This is a diagram illustrating an example of the functional configuration of the information processing system involved in the first embodiment.

[0064] (Functional structure of information processing device) The information processing device 100 executes a predetermined program stored in a storage medium by a computer 200 included in the information processing device 100, thereby achieving, for example... Figure 4 The functional structures are shown below. Figure 4In the example, the information processing device 100 includes a functional structure comprising a communication unit 401, a display control unit 402, a login processing unit 403, an acquisition unit 404, a prompt generation unit 405, a document creation unit 406, an output unit 407, a data processing unit 408, a management unit 409, and a storage unit 410. Furthermore, at least some of the aforementioned functional structures can also be implemented in hardware.

[0065] The communication unit 401 uses, for example, a network I / F 208 to connect the information processing device 100 to the communication network N and performs communication processing to communicate with the user terminal 10 or an external system 110.

[0066] The display control unit 402 performs display control processing to cause the user terminal 10 to display various display screens. For example, the display control unit 402 functions as a web server that causes the web browser 421 of the user terminal 10 to display web page content, which displays various display screens as described later.

[0067] Alternatively, the user terminal 10 may execute an application program (hereinafter referred to as "application") corresponding to the information processing system 1, which displays the various display screens described above. In this case, instead of the information processing device 100, the user terminal 10 includes a display control unit 402. Thus, the display control unit 402 may also be a device other than the information processing device 100 within the information processing system 1.

[0068] The login processing unit 403 authenticates the user 11 who has logged into the information processing device 100. If the authentication is successful, the login processing unit 403 obtains the user information of the user 11 and stores it in the storage unit 410 or the like for management. For example, if the user ID and password included in the login request from the user terminal 10 are registered in the user information 411 stored in the storage unit 410 or the database 112, the login processing unit 403 will determine that the user's authentication is successful. However, this is just one example; the login processing unit 403 may also use an external authentication system or the like to authenticate the user 11.

[0069] Figure 5 This is a diagram representing an example of user information in the first implementation. Figure 5 In the example, user information 411 includes "user_id", "pass_word", "name", "email", "role", and "department" as items. "user_id" is the user's identification information (user ID). "pass_word" is the password used to authenticate the user. Additionally, the password is an example of authentication information.

[0070] "name" indicates the user's name, etc. "email" is the user's email address. "role" indicates the user's role (responsibilities, function). "department" indicates the user's organization (department, etc.).

[0071] The acquisition unit 404 performs acquisition processing to acquire setting information related to document creation. For example, the acquisition unit 404 acquires setting information from a setting screen that is input to the display control unit 402 and displayed on the user terminal 10 to set setting information. This setting information includes, for example, information such as the type of document, the purpose of the document, the type of input data, the number of characters to be output, information that should be included in the document, and information that should not be included in the document.

[0072] In addition, the acquisition unit 404 acquires input data from the user terminal 10, etc., and stores the acquired input data in the storage unit 410 or the database 112, etc.

[0073] The prompt generation unit 405 performs prompt generation processing based on the input data and setting information to generate prompts instructing the large-scale language model 113 to create a document. Here, the input data may be, for example, audio data, text data converted from audio data (textualized text), or document data. Furthermore, the setting information is the document creation-related setting information acquired by the acquisition unit 404.

[0074] Preferably, the prompt generation unit 405 generates a prompt using a template that includes information about the output items, setting information, etc. For example, the prompt generation unit 405 generates at least a portion of a prompt based on the template, which instructs the large-scale language model 113 to create a document, and generates a prompt that includes information about the output items and at least a portion of the generated prompt. Specific methods for generating prompts will be described later.

[0075] The document creation unit 406 performs document creation processing using the large-scale language model 113 based on input data and setting information. For example, the document creation unit 406 uses prompts generated by the prompt generation unit 405 to instruct the large-scale language model 113 to create a document. As another example, the document creation unit 406 can instruct the large-scale language model 113 to create a document based on input data and setting information using pre-made prompts, thereby creating a document in a prescribed form.

[0076] The output unit 407 performs output processing on the document created by the output document creation unit 406. For example, the output unit 407 outputs the created document to the display screen shown on the user terminal 10 via the display control unit 402. Alternatively, the output unit 407 can also send the created document to a designated destination via email. Furthermore, the output unit 407 can also store (output) the created document to the output data 414 in the storage unit 410 or database 112, etc.

[0077] The data processing unit 408 performs data processing on the acquired data. For example, the data processing unit 408 modifies the data acquired by the acquisition unit 404 into a form suitable for processing.

[0078] The management unit 409 performs management processing on templates that contain information about the output items of the document creation process. Preferably, the template contains setting information for document creation. The management unit 409 manages the templates, for example, by storing them in a template 413 in a storage unit 410 or a database 112.

[0079] Figures 6A-6C This diagram illustrates an example of information contained in a template involved in the first embodiment. Template 413 may contain, for example, attribute information 413a, setting information 413b, and output item information 413c.

[0080] Attribute information 413a retains information related to the template's attributes. In Figures 6A-6C In the example, attribute information 413a includes information such as "No.", "template_name", "available_role", "available_department", "input_data_type", and "purpose" as items. "No." is the template's serial number. "template_name" is information indicating the template's name, etc.

[0081] "available_role" indicates the roles (responsibilities, functions) that are allowed to use the template. If multiple roles exist, "available_role" maintains multiple roles in list form. Here, the role (responsibilities, functions) corresponds to... Figure 5 The "role" in user information 411 is described below. "available_department" indicates the unit (department) to which the template is allowed. If multiple units belong to this template, "available_department" will be maintained in a list format. Here, the unit (department) corresponds to "department" in user information 411.

[0082] "input_data_type" indicates the type of input data. "purpose" indicates the purpose of the template.

[0083] Setting information 413b retains the template's settings. In Figures 6A-6C In the example, setting information 413b includes information such as "type", "user_id", "template_no", "template_name", "input_data_type", "purpose", "important_information", "expressions_to_avoid", "min_length", and "max_length" as items.

[0084] The "type" setting indicates whether it's the default template ("default") or a personal template ("individual"). "user_id" is the user ID that identifies the user when "type" is "individual". "template_no" is the template's serial number. "template_name" is the template's name. "input_data_type" indicates the type of input data. "purpose" indicates the template's purpose.

[0085] "important_information" represents information that should be included in the document being created. If multiple instances of this information exist, "important_information" will be presented as a list.

[0086] "expressyons_to_avoid" represents information that should not be included in the document being created. If multiple instances of this information exist, "expressyons_to_avoid" maintains the information as a list. "min_length" represents the minimum number of characters required to be the target when creating the document. "max_length" represents the maximum number of characters required to be the target when creating the document.

[0087] The output item information 413c maintains information related to the template's output items. The output item information 413c includes, for example, information such as "target_template_no", "item_no", "user_id", "type", "target_role", "target_department", "item_name", "data_type", "description", "details", and "condition" as items.

[0088] "target_template_no" represents the "template_no" information of the corresponding template in attribute information 413a and setting information 413b. "item_no" is the serial number within the range of "template_no" that becomes the target. "user_id" retains the user ID if the template corresponding to "template_no" reflects personal settings. "type" is set to "default" if the template corresponding to "template_no" is the default template, and set to "individual" if it reflects personal settings.

[0089] "target_role" indicates the role that can use the template corresponding to "template_no". "target_department" indicates the department to which the template corresponding to "template_no" belongs. "item_name" indicates the name of the template corresponding to "template_no". "data_type" indicates the data type of the template corresponding to "template_no". "description" indicates the description of the template corresponding to "template_no".

[0090] "details" contains detailed information about the template corresponding to "template_no". "condition" contains information about the conditions of the template corresponding to "template_no".

[0091] Figure 7 This represents an example of setting information 413b that reflects personal settings. Additionally, Figure 8 This is an example of information 413c that reflects the output items set by an individual.

[0092] The storage unit 410 may be implemented, for example, by a storage device such as an HDD 204 and an HDD controller 205 included in the information processing apparatus 100, and a program executed by the CPU 201. The storage unit 410 may store various information and data, such as user information 411, input data 412, templates 413, and output data 414. Alternatively, the information processing apparatus 100 may store at least a portion of the user information 411, input data 412, templates 413, and output data 414 in the database 112 of the external system 110, instead of the storage unit 410.

[0093] (Functional structure of user terminals) As an example, such as Figure 4 As shown, user terminal 10 has the functionality of a web browser 421 and utilizes the services provided by information processing device 100. Alternatively, user terminal 10 can also cooperate with information processing device 100 and utilize the services provided by information processing system 1 by executing an application corresponding to information processing system 1.

[0094] (External system) External system 110 includes, for example, a voice recognition engine 111, a database 112, a large-scale language model 113, and a daily report system 114.

[0095] The voice recognition engine 111 performs voice recognition processing, converting voice data into text data (textualized text). In this embodiment, existing voice recognition engines or voice recognition services are used directly without any processing. The database 112 is, for example, a database that stores various data such as user information 411, input data 412, templates 413, and output data 414.

[0096] Large-scale language model 113 is a language model constructed using large amounts of text data and deep learning techniques. Large-scale language model 113 can perform various natural language processing tasks, such as text reading, understanding, generation, translation, summarization, and response. In this embodiment, for example, it is assumed that an existing large-scale language model 113 is used directly without modification.

[0097] Daily report system 114 is, for example, a system for managing daily reports. Daily report system 114 is, for example, an example of a system for managing documents such as daily reports or meeting minutes. Information processing system 1 may also not use daily report system 114, etc., but store and manage documents such as daily reports or meeting minutes in, for example, in database 112 or storage unit 410.

[0098] <Processing Flow> Next, the processing flow of the information processing method involved in the first embodiment will be described. Figure 9 and Figure 10 This diagram illustrates an example of the processing of an information processing system according to a first embodiment. This processing, for example, represents a system with... Figure 4 This is an example of the processing performed by the information processing system 1, which has a functional structure described herein. Here, we assume that the user is a user who has registered with the information processing system 1, and the following explanation will be provided.

[0099] In step S901, user 11 logs in to user terminal 10 and inputs settings. For example, user 11 logs in to user terminal 10 by accessing settings displayed on user terminal 10, such as... Figure 11 On the login screen 1100, enter the user ID and password, etc. 1101, and select the "Login" button 1102 to log in to the information processing system 1.

[0100] In addition, user 11 pairs Figure 12 The input settings screen 1200 shows the input data settings bar 1201 and the document creation purpose settings bar 1202. Select the "Start Input" button 1203 to set the input.

[0101] Regarding the input data settings, you can specify audio data, text data, or daily report data from an external daily report system 114. Daily report system 114 is just one example; it's not limited to daily reports and could also be a system that manages other document data. Furthermore, when selecting daily report data from daily report system 114, you can select one or multiple data points.

[0102] User 11 uploads input data when selecting audio or text data from the input data settings bar 1201. Additionally, user 11 enters a search query when selecting a daily report from the input data settings bar 1201. Furthermore, user 11 enters the purpose of the document creation.

[0103] In step S902, the user terminal 10 sends the entered user ID and password to the information processing device 100. Therefore, in step S903, the login processing unit 403 of the information processing device 100 checks whether the user ID and password are registered in the user information 411 stored in the database 112 (or storage unit 410). Here, since user 11 is already registered in the information processing system 1, the information processing system 1 executes the processing after step S904. If the user ID and password are not registered in the user information 411, the login processing unit 403 sends a message indicating authentication failure to the user terminal 10 and suspends the execution of the processing after step S904.

[0104] In step S904, the user terminal 10 sends the input data (in this case, voice data) set in the input setting screen 1200 and the purpose of document creation, etc., to the information processing device 100. As a result, the acquisition unit 404 of the information processing device 100 acquires the voice data and the purpose of document creation sent by the user terminal 10.

[0105] In step S905, the acquisition unit 404 of the information processing device 100 sends the acquired sound data to the voice recognition engine 111. Furthermore, in step S906, the acquisition unit 404 obtains the text data converted into text by the voice recognition engine 111. In step S904, since the acquisition unit 404 has already obtained the converted text data, the processing steps S905 and S906 are omitted.

[0106] In step S907, the management unit 409 of the information processing device 100 obtains a template from the template 413 of the database 112 (or storage unit 410). For example, the management unit 409 obtains a template such as... Figures 6A-6C The attribute information 413a, setting information 413b, and output item information 413c shown are as follows.

[0107] In step S908, the display control unit 402, based on user information, input data, and document creation purpose information acquired by the acquisition unit 404, and the template acquired by the management unit 409, causes the user terminal 10 to display the output setting screen. For example, the display control unit 402 causes the user terminal 10 to display... Figure 13 The output settings screen shown is 1300.

[0108] exist Figure 13 In the example, the display control unit 402 displays a setting bar 1301 for the type of input data, a setting bar 1302 for the purpose of document creation, a selection bar 1303 for output items, an output item selection bar 1304 for templates, and a setting bar 1305 for additional items on the output setting screen 1300. Additionally, the display control unit 402 displays a setting bar 1306 for the output word count standard, a setting bar 1307 for particularly important information, and a setting bar 1308 for displays that should be avoided on the output setting screen 1300.

[0109] Preferably, the display control unit 402 is based on, for example, Figures 6A-6C The attribute information 413a, setting information 413b, and output item information 413c shown are displayed in the output setting screen 1300, showing the default setting values, etc.

[0110] For example, the display control unit 402 can classify the content of the input data by inputting the input data into the large-scale language model 113 or a machine learning-based classification model, and then display the default setting value in the input data type setting column 1301. Alternatively, the display control unit 402 can also display the selectable input data types in the input data type setting column 1301, allowing the user 11 to select the type of input data.

[0111] Additionally, the display control unit 402 inputs the document creation purpose entered by the user 11 into the large-scale language model 113 and modifies it into a form suitable for reflecting prompts used in document creation. As a result, the default setting value is displayed in the document creation purpose setting field 1302. Alternatively, the display control unit 402 may also display selectable document creation purposes in the document creation purpose setting field 1302, allowing the user 11 to select the document creation purpose.

[0112] Furthermore, the display control unit 402 can selectively display candidate templates for document creation in the output item selection bar 1303. For example, the display control unit 402 refers to... Figure 5 User information 411 and Figures 6A-6C Template 413, as described in the text, displays the type of input data, the purpose of document creation, and the content of the template that the user can select. Preferably, the management unit 409 of the information processing device 100 is based on, as... Figures 6A-6C The attribute information 413a, setting information 413b, and output item information 413c shown are as follows: Figure 14 As shown, here is a list of 1400 templates that user 11 can use to create and manage.

[0113] Preferably, the display control unit 402 displays the output items of the currently selected template in the output items 1304 of the template. Preferably, the display control unit 402 displays the setting bar 1305 for adding items, and accepts the user 11's addition of output items.

[0114] Additionally, the display control unit 402 displays information associated with the template selected by default in the output item selection bar 1303 in settings such as the output character count reference setting bar 1306, the setting bar for particularly important information 1307, and the setting bar for desired display. For example, the display control unit 402 will display information such as... Figures 6A-6C The default values ​​set in attribute information 413a, setting information 413b, and output item information 413c are shown in the setting bar 1306 for the output character count baseline, the setting bar 1307 for particularly important information, and the setting bar 1308 for behaviors that should be avoided. 1309 indicates the "Settings Button".

[0115] In step S909, user 11 selects a template in the selection bar 1303 of the output items on the output setting screen 1300. Then, in step S910, user terminal 10 sends the selected template information to information processing device 100.

[0116] Here, if user 11 does not change the output items and settings of the template (using the default template directly), information processing system 1 executes... Figure 9 Processing without changing the output items and settings (920).

[0117] For example, in step S921, as Figure 13 As shown, the display control unit 402 of the information processing device 100 displays the default settings of the selected template on the output setting screen 1300.

[0118] In step S922, the management unit 409 of the information processing device 100 sets the template selected by the user 11 as the template for document output.

[0119] On the other hand, if user 11 changes the output items or settings of the template, information processing system 1 executes the change. Figure 10 Processing 1000 in cases of output items or setting information.

[0120] For example, when using the default template, in step S1000, the display control unit 402 of the information processing device 100 displays the default output items and setting information of the template selected by the user 11 on the output setting screen 1300.

[0121] In step S1001, the display control unit 402 of the information processing device 100 causes the output setting screen 1300 to display the setting bar 1305 for additional items.

[0122] In step S1002, when user 11 changes (adds, modifies, or deletes, etc.) the output item or setting information, in step S1003, the acquisition unit 404 of the information processing device 100 acquires the set information.

[0123] In step S1004, the data processing unit 408 of the information processing device 100 preprocesses the setting information. For example, the data processing unit 408 modifies the setting information into a form suitable for being reflected in the prompts for document creation. As a specific example, the data processing unit 408 uses a large-scale language model 113 to modify the setting information into a regular expression.

[0124] In step S1005, the management unit 409 of the information processing device 100 sets the preprocessed information as a user-specific template in template 413.

[0125] In step S1006, the management unit 409 of the information processing device 100 sets the user-specific template set in step S1005 as the template for document output. For example, such as Figure 7 , 8 As shown, the management unit 409 registers user settings information in the setting information 413b and the output item information 413c.

[0126] For example, such as Figure 7 As shown, in setting information 413b, management unit 409 sets "individual" in the "type" field of the user-defined template. Additionally, management unit 409 sets the user ID of the user in the "user_id" field of the user-defined template. Furthermore, management unit 409 sets the template's serial number in the "template_no" field of the user-defined template, according to each user. For example, if a user is allowed to add up to 10 templates, "template_no" might be a value of 90-100 for each user. Furthermore, management unit 409 sets other fields in setting information 413b, such as... Figure 10 The information set in steps S1001 to S1004.

[0127] In addition, such as Figure 8 As shown, in the output project information 413c, the management unit 409 sets "individual" in the "type" field of the user-defined template. Additionally, the management unit 409 sets the user ID of the user in the "user_id" field of the user-defined template. Furthermore, in the output project information 413c, the management unit 409 leaves the "target_role" and "target_department" fields of the user-defined template blank.

[0128] Execution remains unchanged Figure 9 The processing of output items and setting information, or changes, 920. Figure 10 After processing the output items or setting information in case of step S1000, the information processing system 1 executes the processing after step S1011.

[0129] In step S1011, the prompt generation unit 405 of the information processing device 100 generates a prompt based on a set template, which instructs the large-scale language model 113 to generate a document. For example, the prompt generation unit 405 performs the prompt generation process described later.

[0130] In step S1012, the document creation unit 406 of the information processing device 100 uses the prompt generated by the prompt generation unit 405 to instruct the large-scale language model 113 to create a document. In step S1013, the document creation unit 406 obtains the document generated by the large-scale language model 113.

[0131] In step S1014, the output unit 407 of the information processing device 100 uses the display control unit 402 to cause the user terminal 10 to display the document created by the document creation unit 406. This process is an example of the process by which the output unit 407 outputs the document created by the document creation unit 406.

[0132] In step S1015, the output unit 407 of the information processing device 100 stores the document created by the document creation unit 406 in the database 112 or the output data 414 of the storage unit 410. This process is another example of the process by which the output unit 407 outputs the document created by the document creation unit 406.

[0133] pass Figure 9 , Figure 10 In the information processing system 1 that creates documents based on input data, it is easy to create documents containing the information required by the user.

[0134] like Figure 14 As shown, the management unit 409 of the information processing device 100 can also manage each user separately by establishing a correspondence between the available template list 1400 and the user information 411. Furthermore, the display control unit 402 of the information processing device 100 can, based on the available template list 1400, cause the user terminal 10 to display... Figure 13 The output item selection bar 1303 and the template output item 1304, etc.

[0135] In addition, in the above description, the attribute information 413a, setting information 413b, output item information 413c, and list of available templates 1400 are recorded in tabular form, but this information can also be prepared in other forms such as JSON (JavaScript Object Notation: a lightweight data exchange format based on the JavaScript language).

[0136] (Hint generation and processing) Figure 15 This is a flowchart illustrating an example of a prompt generation process involved in an implementation. This process is represented, for example, in... Figure 10 The following is an example of the process in which the prompt generation unit 405 generates a prompt based on a set template in step S1011, etc.

[0137] In step S1501, the generation unit 405 is prompted to retrieve data from the template 413 stored in the storage unit 410 or the database 112. For example, such as Figures 6A-6C As shown, the prompt generation unit 405 obtains the data of the set template from attribute information 413a, setting information 413b, output item information 413c, etc.

[0138] In step S1502, the prompt generation unit 405 generates a partial prompt based on the acquired data. For example, the prompt generation unit 405 divides the text set in the prompt into four parts. These four parts include, for example, "parts related to basic rules and guidelines", "parts related to input data", "parts related to output items", and "parts related to detailed rules".

[0139] For example, prompt generation unit 405 uses, such as Figure 16 The function 1600, as shown, generates "the part related to basic rules and guidelines." Similarly, the prompt generation unit 405 prepares, as... Figure 17 The function 1700 shown generates the four prompt sections mentioned above. Figure 18 Example 1800 represents the case where function 1700 is executed.

[0140] Here, return to Figure 15 Continuing with the flowchart, in step S1503, the generation unit 405 combines the parts generated in step S1502 based on rules. Figure 19 Example 1900 shows the combination of the four generated parts.

[0141] In step S1504, the generation unit 405 prompts the generation unit to perform post-processing on the data obtained by combining the four parts as needed, to generate, for example... Figure 20 The prompt shown is 2000.

[0142] Figure 20 This is an example of a generated prompt related to an embodiment. The prompt 2000 is generated by the prompt generation unit 405. Figure 15 An example of a prompt generated during the processing of a large-scale language model 113 to instruct the creation of a document.

[0143] In prompt 2000, the prompt generation unit 405 sets the information of output item 2001 based on the "description" data of output item information 413c. Furthermore, the prompt generation unit 405 sets the description 2002 about the output item based on the "details" data of output item information 413c. Additionally, when "conditions" are set in the output item information 413c, the prompt generation unit 405 adds the "conditions" as conditions 2003 for outputting that item to the description 2002 about the output item.

[0144] Additionally, in prompt 2000, prompt generation unit 405 sets document creation information 2004 based on the "purpose" data in setting information 413b. Furthermore, prompt generation unit 405 sets input data information 2005 based on the "input_data_type" data in setting information 413b. Also, prompt generation unit 405 sets the output character count information 2006 based on the "min_length" and "max_length" data in setting information 413b. Furthermore, prompt generation unit 405 sets important information 2008 based on the "important_information" data in setting information 413b. Similarly, prompt generation unit 405 sets information on desired expressions 2007 based on the "expression_to_avoid" data in setting information 413b. Additionally, prompt 2000 includes input data 2009.

[0145] As described above, the prompt generation unit 405 generates a prompt based on a template 413 including output item information 413c and setting information 413b, which instructs the large-scale language model 113 to create a document.

[0146] [Second Implementation Form] Figure 21 and Figure 22 This is a sequence diagram illustrating an example of processing in the information processing system according to the second embodiment. This processing represents an example of processing when user 11 is viewing a daily newspaper. The functional configuration of the information processing system 1 according to the second embodiment can be... Figure 4 The information processing system 1 described in the first embodiment has the same functional configuration. Furthermore, the basic processing content is the same as in the first embodiment; therefore, detailed descriptions of the processing identical to that in the first embodiment are omitted here.

[0147] In step S2101, user 11 logs in to user terminal 10 and inputs settings.

[0148] In step S2102, the user terminal 10 sends the entered user ID and password to the information processing device 100. Therefore, in step S2103, the login processing unit 403 of the information processing device 100 confirms whether the user ID and password are registered in the user information 411 stored in the database 112 (or storage unit 410). Here, since user 11 has already registered in the information processing system 1, the information processing system 1 executes the processing after step S2104.

[0149] In step S2104, the management unit 409 of the information processing device 100 obtains a template from the template 413 of the database 112 (or storage unit 410). For example, the management unit 409 obtains a template such as... Figures 6A-6C The attribute information 413a, setting information 413b, and output item information 413c shown are as follows.

[0150] In step S2105, the display control unit 402, based on user information, input data, and document creation purpose information acquired by the acquisition unit 404, and the template acquired by the management unit 409, causes the user terminal 10 to display the output setting screen. For example, the display control unit 402 causes the user terminal 10 to display... Figure 13 The output settings screen shown is 1300.

[0151] In step S2106, user 11 selects "Daily Summary" in the output item selection bar 1303 of the output setting screen 1300 and enters a search query. Then, in step S2107, user terminal 10 sends the entered search query to information processing device 100.

[0152] In step S2108, the data processing unit 408 of the information processing apparatus 100 performs preprocessing as needed, setting the search query received from the user terminal 10 into search conditions of a predetermined form. This processing can be performed by the data processing unit 408, or it can be performed by the large-scale language model 113 by the data processing unit 408.

[0153] In step S2109, the acquisition unit 404 of the information processing device 100 searches the daily report system 114 to acquire the daily reports that are the target. Additionally, in step S2110, the acquisition unit 404 searches the database 112 (or storage unit 410) to acquire the daily reports and documents that are the target.

[0154] In step S2111, the display control unit 402 of the information processing device 100 displays, for example, candidate daily reports and documents to be targeted on the output setting screen 1300.

[0155] Next, in Figure 22In step S2112, if user 11 selects a daily report or document to be the object from the candidates of daily reports and documents to be the object, then information processing system 1 performs the processing after step S2113.

[0156] In step S2113, the user terminal 10 sends the information of the daily report and document selected by the user 11 to the information processing device 100.

[0157] In step S2114, the acquisition unit 404 of the information processing device 100 acquires the selected daily newspaper from the daily newspaper system 114. Furthermore, in step S2115, the acquisition unit 404 acquires the selected daily newspaper and document from the database 112 (or storage unit 410).

[0158] In step S2116, the acquisition unit 404 organizes the acquired daily reports and documents into text data to create input data.

[0159] When user 11 is reading the daily report, the information processing device 100 can obtain input data through the above processing.

[0160] In step S2120, if user 11 selects a template, then information processing system 1 performs document creation processing in step S2121 in the same manner as in the first embodiment. For example, information processing system 1 executes... Figure 9 , Figure 10 The processing after step S910.

[0161] According to the second embodiment, multiple documents can be used as input data, such as documents that can easily generate daily summaries.

[0162] [Third Implementation Form] In the third embodiment, an example of creating documents using a voice-conversational AI (Artificial Intelligence) agent (hereinafter referred to as conversational AI) will be explained.

[0163] <System Structure> Figure 23 This is a diagram illustrating an example of the system configuration of the information processing system according to the third embodiment. The information processing system 1 according to the third embodiment is... Figure 1 The implementation described herein involves the structure of the information processing system 1 plus the dialogue system 2300.

[0164] exist Figure 23 In the middle, user terminal 10 displays dialogue AI 2302 (virtual human) provided by dialogue system 2300, and user 11 can have a conversation with dialogue AI 2302. Additionally, in Figure 23In the example, user 11 brings dialogue AI 2302 to the table during a discussion with customer 2301.

[0165] Dialogue AI2302 is a dialogue system with a virtual human interface that can recognize the speech of speakers (e.g., user 11 and customer 2301) through voice, and perform voice-based responses, text-based responses, and display of associated images.

[0166] <Functional Structure> Figure 24 This is a diagram illustrating an example of the functional configuration of the information processing system involved in the third implementation. Figure 24 In the example, the dialogue system 2300 has an input receiving unit 2411 and a dialogue AI control unit 2412. Besides, according to reference... Figure 4 In addition to the functional configuration of the information processing apparatus 100 in the first embodiment described, the information processing apparatus 100 also includes a dialogue engine 2401. However, Figure 24 The functional configuration of the information processing system 1 shown is an example. The information processing system 1 only requires any device within it to have an input receiving unit 2411, a dialogue AI control unit 2412, and a dialogue engine 2401.

[0167] The input receiving unit 2411 is implemented, for example, by a program executed by the computer 200 included in the dialogue system 2300. The input receiving unit 2411 receives speech data (voice data, or text data obtained by texturing the speech content) and user 11's operations from the user terminal 10, and sends the received data to the information processing device 100.

[0168] The dialogue AI control unit 2412 is implemented, for example, by a program executed by the computer 200 included in the dialogue system 2300, and controls the dialogue AI 2302 according to the response data sent from the information processing device 100. For example, the dialogue AI control unit 2412 converts the response data (text data) sent from the dialogue engine 2401 into voice data, sends it to the user terminal 10, displays the dialogue AI 2302, and thereby controls the dialogue AI 2302 displayed on the user terminal 10.

[0169] The dialogue engine 2401 is implemented, for example, by a program executed by the computer 200 included in the information processing device 100, to generate response data to speech data sent from the dialogue system 2300 by the user 11. Furthermore, the dialogue engine 2401 controls the information processing device 100 based on the user 11's operations or instructions given through speech, etc.

[0170] <Processing Flow> Next, the processing flow of the information processing method according to the third embodiment will be described.

[0171] (Preparation before the dialogue begins) Figure 25 This is a sequence diagram illustrating an example of the processing prior to the start of a dialogue according to the third implementation. Before initiating a dialogue with customer 2301, user 11 uses user terminal 10 to perform... Figure 25 The processing. Additionally, execution. Figure 25 The user terminal 10 that processes the data can be the same terminal as the user terminal 10 that conducts the dialogue, or it can be a different terminal.

[0172] In step S2501, the user terminal 10 sends login information (e.g., user ID and password) to the information processing device 100 based on the login operation of the user 11.

[0173] In step S2502, the login processing unit 403 of the information processing device 100 confirms the user information 411 and determines whether the login information is registered in the user information 411.

[0174] In step S2503, if the login information is registered in the user information 411, the login processing unit 403 of the information processing device 100 sends an authentication OK message to the user terminal 10. Conversely, if the login information is not registered in the user information 411, the login processing unit 403 suspends the process. Figure 25 The processing.

[0175] In step S2504, the user terminal 10 sends a change instruction for the daily report information to the information processing device 100 based on the user 11's change operation on the daily report information.

[0176] In step S2505, the display control unit 402 of the information processing device 100 obtains the template number corresponding to the user 11 from the user information 411 stored in the database 112 or the storage unit 410.

[0177] Figure 26 This diagram illustrates an example of user information in the third implementation. It shows an example of user information 2600 for user 11. (Example...) Figure 26 As shown, the user information 2600 in the third embodiment includes a template number 2601 for the daily report corresponding to user 11. In step S2505, the display control unit 402 obtains the template number 2601.

[0178] In step S2506, the display control unit 402 of the information processing device 100 obtains information on the output items and setting information from the template corresponding to the obtained template number.

[0179] In step S2507, the display control unit 402 of the information processing device 100 displays the current daily report settings on the user terminal 10 based on the information of the obtained output items and setting information.

[0180] In step S2508, the user terminal 10 sends the changes made by user 11 to the information processing device 100 according to the changes made by user 11 to the settings of the daily report.

[0181] In step S2509, the management unit 409 of the information processing device 100 updates the template 413 stored in the database 112 or storage unit 410 based on the changes received from the user terminal 10.

[0182] In step S2510, the display control unit 402 of the information processing device 100 causes the user terminal 10 to display a message indicating that the update is complete.

[0183] pass Figure 25 In terms of processing, for example, user 11 can set the output items and settings of the daily report generated by information processing device 100 before consulting with customer 2301.

[0184] (Processing by the information processing system) Figure 27 This is a sequence diagram illustrating an example of processing in an information processing system according to a third embodiment. Here, as... Figure 23 As shown, user 11 displays dialogue AI 2302 on user terminal 10 and negotiates with customer 2301.

[0185] In step S2701, user terminal 10 sends a meeting record creation request to dialogue system 2300 according to user 11's meeting record creation instructions. For example, user 11 can... Figure 28 The AI ​​2302 shown in the display screen 2800 uses voice commands to instruct on the creation of meeting minutes.

[0186] In step S2702, the input receiving unit 2411 of the dialogue system 2300 sends the request for the creation of meeting minutes received from the user terminal 10, for example, to the information processing device 100.

[0187] In step S2703, the dialogue engine 2401 of the information processing device 100 generates a response to the request to create meeting minutes from the dialogue system 2300. For example, the dialogue engine 2401 inputs a prompt including the voice text of user 11 into the large-scale language model 113 to generate a response.

[0188] In step S2704, the dialogue engine 2401 sends the generated response text to the dialogue system 2300.

[0189] In step S2705, the dialogue AI control unit 2412 of the dialogue system 2300 converts the response text received from the information processing device 100 into voice and describes the dialogue AI 2302, thereby enabling the dialogue AI 2302 displayed on the user terminal 10 to respond.

[0190] In step S2706, the prompt generation unit 405 of the information processing device 100 retrieves the dialogue history data between the user 11 and the customer 2301 from the database 112 or the storage unit 410. This dialogue history data is, for example, data stored in the database 112 or the storage unit 410 by the dialogue engine 2401 from the speech data received by the dialogue system 2300 and the response data generated by the dialogue engine 2401.

[0191] In step S2707, the prompt generation unit 405 of the information processing device 100 divides the acquired dialogue history data into predetermined sizes.

[0192] Next, the information processing device 100 repeatedly executes steps S2708 to S2710 for each segmented dialogue history data.

[0193] In step S2708, the prompt generation unit 405 of the information processing device 100 generates a prompt based on the segmented dialogue history data, which instructs the large-scale language model 113 to generate meeting minutes.

[0194] In step S2709, the document creation unit 406 of the information processing device 100 inputs the prompt generated by the prompt generation unit 405 into the large-scale language model 113, instructing it to generate meeting minutes. Furthermore, in step S2710, the document creation unit 406 obtains the meeting minutes generated by the large-scale language model 113.

[0195] After performing steps S2708 to S2710 on the segmented dialogue history data, the information processing device 100 performs processing after step S2711.

[0196] In step S2711, the document creation unit 406 of the information processing device 100 summarizes multiple meeting minutes generated by the large-scale language model 113 to generate a meeting record, which is then stored in the database 112 or the storage unit 410, etc. The document creation unit 406 can cause the large-scale language model 113 to perform processing for summarizing multiple meeting minutes.

[0197] In step S2712, the dialogue engine 2401 of the information processing device 100 generates a response. Furthermore, in step S2713, the dialogue engine 2401 sends the generated response and the meeting minutes prepared by the document production unit 406 to the dialogue system 2300.

[0198] In step S2714, the dialogue system 2300 causes the user terminal 10 to display the conference recording and respond to the user 11. For example, the dialogue system 2300 causes the user terminal 10 to display... Figure 29 The display screen shown is 2900. In Figure 29 In the example, the meeting minutes 2901 generated by the information processing device 100 are displayed on the display screen 2900, and the dialogue AI 2302 speaks to indicate that the meeting minutes have been completed.

[0199] In step S2715, the information processing system 1, for example, performs... Figure 30 The daily report generation process is shown below.

[0200] (Daily report generation process) Figure 30 This is a sequence diagram illustrating an example of the daily report generation process in the third implementation. For example, information processing system 1 repeatedly performs this process at predetermined intervals.

[0201] In step S3001, the document creation unit 406 of the information processing device 100 obtains the current date and time. Additionally, in step S3002, the document creation unit 406 obtains meeting minutes and daily report data from the database 112 or storage unit 410, etc., within the most recent specified period.

[0202] In step S3003, the document production department 406 obtains the dialogue IDs for meetings with minutes but no daily reports. Here, the dialogue ID is, for example, identification information for dialogues such as business negotiations.

[0203] In the case of meeting minutes but no daily report dialogue IDs, the document production department 406 performs the processing after step S3011 for each obtained dialogue ID.

[0204] In step S3011, the document creation unit 406 obtains user information corresponding to the dialogue ID that is the object of processing from the database 112 or the storage unit 410, etc.

[0205] In step S3012, the document production unit 406 obtains the template number from the acquired user information.

[0206] In step S3013, the document creation unit 406 obtains a template corresponding to the obtained template number from the database 112 or the storage unit 410, etc.

[0207] In step S3014, the prompt generation unit 405 generates a prompt to remove the dialogue history in the prompt instructing the large-scale language model 113 to create a log based on the information of the output items included in the acquired template and the setting information.

[0208] In step S3015, the prompt generation unit 405 retrieves the dialogue history data corresponding to the dialogue ID that is being processed from the database 112, the storage unit 410, etc.

[0209] In step S3016, the prompt generation unit 405 divides the acquired dialogue history data into predetermined sizes.

[0210] Next, the information processing device 100 repeatedly executes steps S3017 to S3019 for each segmented dialogue history data.

[0211] In step S3017, the prompt generation unit 405 of the information processing device 100 sets the segmented dialogue history data to a prompt that removes the dialogue history, and generates a prompt to instruct the large-scale language model 113 to generate a daily report.

[0212] In step S3018, the document production unit 406 of the information processing device 100 inputs the prompt generated by the prompt generation unit 405 into the large-scale language model 113, instructing it to generate a daily report. Furthermore, in step S3019, the document production unit 406 obtains the daily report generated by the large-scale language model 113.

[0213] After performing steps S3017 to S3019 on the segmented dialogue history data, the information processing device 100 performs processing after step S3020.

[0214] In step S3020, the document creation unit 406 of the information processing device 100 summarizes multiple daily reports generated by the large-scale language model 113 to generate a daily report, and generates an email containing the generated daily report. For example, the document creation unit 406 generates an email containing the generated daily report. Figure 31 The email 3100 shown contains daily report 3101.

[0215] In step S3021, the output unit 407 of the information processing device 100 sends the email 3100 created by the document creation unit 406 to a designated destination (e.g., the email address of user 11).

[0216] (Daily News Settings) Figure 32 This is a sequence diagram illustrating an example of the daily report setting process in the third implementation. The diagram shows an example of how user 11 changes the daily report settings through a dialogue with AI 2302.

[0217] In step S3201, user terminal 10 sends a daily report setting start request to dialogue system 2300 according to user 11's daily report setting start instruction. For example, user 11 can communicate via voice. Figure 33The dialog AI2302 displayed on the screen 3300, as shown, indicates the start of the daily report settings.

[0218] In step S3202, the dialogue engine 2401 of the information processing device 100 obtains the daily setting information corresponding to user 11 from the database 112 or storage unit 410, for example, using the management unit 409, according to the start instruction of the daily setting of user 11.

[0219] In step S3203, the dialogue engine 2401 instructs the large-scale language model 113 to generate a response. Furthermore, in step S3204, the dialogue engine 2401 acquires the response generated by the large-scale language model.

[0220] In step S3205, the dialogue engine 2401 of the information processing device 100 sends the generated response and the acquired current daily report settings to the dialogue system 2300. As a result, the dialogue system 2300, for example, causes the user terminal 10 to display... Figure 34 The display screen shown is 3400.

[0221] exist Figure 34 In the example, when the current settings 3401 of the daily report are displayed on screen 3400, the dialogue AI 2302 makes a designated speech to urge changes to the content. Here, the user 11 can also instruct the dialogue AI 2302 to change the content via voice.

[0222] In step S3206, the dialogue system 2300 converts the user 11's indication of changes into text data and sends it to the information processing device 100.

[0223] In step S3207, the dialogue engine 2401 of the information processing device 100 updates the template settings of the daily report corresponding to user 11 according to the instructions of the user 11 regarding the changes, for example, by using the management unit 409.

[0224] In step S3208, the dialogue engine 2401 of the information processing device 100 instructs the large-scale language model 113 to generate a response. Additionally, in step S3209, the dialogue engine 2401 acquires the response generated by the large-scale language model 113.

[0225] In step S3210, the dialogue engine 2401 of the information processing device 100 sends, for example, the generated response and the modified daily report settings to the dialogue system 2300. Consequently, the dialogue system 2300, for example, causes the user terminal 10 to display... Figure 35 The display screen shown is 3500.

[0226] exist Figure 35In the example, when the changed daily report settings 3501 are displayed on screen 3500, the dialogue AI 2302, as an example, prompts the user to confirm the changes. Conversely, when user 11 expresses gratitude, in step S3211, the dialogue system 2300 converts the user's gratitude into text data, for example, and sends it to the information processing device 100.

[0227] In steps S3212 to S3214, the dialogue engine 2401 of the information processing device 100 generates a thank-you message to the user and sends it to the dialogue system 2300. The dialogue system 2300 then instructs the dialogue AI 2302 to respond to the user 11's thanks (e.g., "You're welcome. Please let me know if you need anything.").

[0228] According to the various embodiments of this disclosure, in the information processing system 1 for creating documents based on input data, documents containing information required by the user can be created.

[0229] The above embodiments illustrate the use of information processing system 1 to create documents such as meeting minutes or daily reports during negotiations. However, these are merely examples; information processing system 1 according to this embodiment can be applied to various fields, such as nursing, education, or training. For instance, information processing system 1 according to this embodiment can create various documents such as nursing logs, daily course reports, class reports, training reports, childcare reports, or departmental activity reports.

[0230] <Supplement> Each function of the above-described embodiments can be implemented by one or more processing circuits. Here, "processing circuit" in this specification includes processors programmed to perform functions via software, such as processors mounted via electronic circuits, ASICs (Application Specific Integrated Circuits), DSPs (digital signal processors), FPGAs (field-programmable gate arrays), existing circuit modules, and other devices designed to perform the functions described above.

[0231] Furthermore, the apparatus group described in the examples represents only one of a plurality of computing environments used to implement embodiments of the present disclosure. In some embodiments, the information processing apparatus 100 includes a plurality of computing devices, such as a server cluster. The plurality of computing devices are configured to communicate with each other via any type of communication link (including networks, shared memory, etc.) to perform the processing of the present disclosure.

[0232] Furthermore, the functional configuration of the information processing device 100 can be integrated into one device or divided into multiple devices. In addition, at least a portion of the functional structures of the information processing device 100 can also be provided by the user terminal 10, the external system 110, or the dialogue system 2300, etc.

[0233] <Postscript> This specification discloses information processing apparatus, information processing system, information processing method, program, computer-readable storage medium, and computer program product.

[0234] (Postscript 1) An information processing device for creating documents based on input data, characterized in that the information processing device comprises: The acquisition unit acquires setting information related to the creation of the document; The document production department, based on the input data and the specified information, uses a large-scale language model to produce the document; and The output section outputs the documents created by the document creation section.

[0235] (Postscript 2) The information processing apparatus according to Appendix 1 is characterized in that: The system further includes a prompt generation unit that generates prompts based on the input data and the setting information, the prompts instructing the large-scale language model to produce the document; The document creation department uses the prompt to instruct the large-scale language model to create the document.

[0236] (Note 3) The information processing apparatus according to Appendix 2 is characterized in that: It further includes a management department that manages templates containing information about the output items of the document; The prompt generation department uses the template to generate the prompt.

[0237] (Postscript 4) The information processing apparatus according to Appendix 3 is characterized in that: The template includes the setting information; The acquisition unit acquires the setting information input to the setting screen for setting the setting information or the setting information stored in the database.

[0238] (Note 5) The information processing apparatus according to Appendix 3 or 4 is characterized in that: The prompt generation unit generates at least a portion of the prompt based on the template; The prompt generation unit generates the prompt, which includes information about the output item and at least a portion of the generated prompt.

[0239] (Note 6) The information processing apparatus according to Appendix 3 or 4 is characterized in that: The prompt includes: The input data; Information indicating the output item; Information indicating the attributes of the large-scale language model; and Information indicating the rules for creating the prompt.

[0240] (Note 7) The information processing apparatus according to any one of Appendices 1 to 5 is characterized in that: The setting information includes one or more of the following: the type of document, the purpose of the document, the type of input data, the number of output characters, the information that the document should contain, and the information that the document should not contain.

[0241] (Postscript 8) The information processing apparatus according to any one of Appendices 1 to 7 is characterized in that: The input data includes sound data, or text data in which the sound data is textualized. The documents include daily reports or meeting minutes.

[0242] (Note 9) The information processing apparatus according to any one of Appendices 1 to 8 is characterized in that: The input data includes multiple document data; The document includes a summary of the data from the multiple documents.

[0243] (Postscript 10) An information processing system for creating documents based on input data, characterized in that the information processing system comprises: The acquisition unit acquires setting information related to the creation of the document; The document production department, based on the input data and the specified information, uses a large-scale language model to produce the document; and The output section outputs the documents created by the document creation section.

[0244] (Postscript 11) The information processing system according to Appendix 10 is characterized in that: It further includes a conversational AI control unit, which controls the conversational AI that engages in conversations with users; The acquisition unit acquires the setting information set by the user using the dialogue AI.

[0245] (Postscript 12) An information processing method, characterized in that a computer creating a document based on input data performs the following processing: The process involves acquiring and processing settings related to the creation of the document. Document creation processing, based on the input data and the defined information, uses a large-scale language model to create the document; and Output processing: Outputs the document created in the document creation process.

[0246] (Postscript 13) A program that causes a computer creating a document based on input data to perform the following processing: The process involves acquiring and processing settings related to the creation of the document. Document creation processing, based on the input data and the defined information, uses a large-scale language model to create the document; and Output processing: Outputs the document created in the document creation process.

[0247] (Postscript 14) A computer-readable storage medium having stored thereon computer program instructions that cause a computer creating a document based on input data to perform the following processes: The process involves acquiring and processing settings related to the creation of the document. Document creation processing, based on the input data and the defined information, uses a large-scale language model to create the document; and Output processing: Outputs the document created in the document creation process.

[0248] (Postscript 15) A computer program product having stored computer program instructions that cause a computer creating a document based on input data to perform the following processes: The process involves acquiring and processing settings related to the creation of the document. Document creation processing, based on the input data and the defined information, uses a large-scale language model to create the document; and Output processing: Outputs the document created in the document creation process.

[0249] The embodiments of the present invention have been described above, but the present invention is not limited to the specific embodiments described above. Various modifications and applications can be made within the scope of the spirit of the present invention as described in the claims.

Claims

1. An information processing device for creating documents based on input data, characterized in that, The information processing device includes: The acquisition unit acquires setting information related to the creation of the document; The document production department, based on the input data and the specified information, uses a large-scale language model to produce the document; and The output section outputs the documents created by the document creation section.

2. The information processing device according to claim 1, characterized in that... : The system further includes a prompt generation unit that generates prompts based on the input data and the setting information, the prompts instructing the large-scale language model to produce the document; The document creation department uses the prompt to instruct the large-scale language model to create the document.

3. The information processing device according to claim 2, characterized in that... : It further includes a management department that manages templates containing information about the output items of the document; The prompt generation department uses the template to generate the prompt.

4. The information processing device according to claim 3, characterized in that... : The template includes the setting information; The acquisition unit acquires the setting information input to the setting screen for setting the setting information or the setting information stored in the database.

5. The information processing apparatus according to claim 3 or 4, characterized in that... : The prompt generation unit generates at least a portion of the prompt based on the template; The prompt generation unit generates the prompt, which includes information about the output item and at least a portion of the generated prompt.

6. The information processing apparatus according to claim 3 or 4, characterized in that... : The prompt includes: The input data; Information indicating the output item; Information indicating the attributes of the large-scale language model; and Information indicating the rules for creating the prompt.

7. The information processing apparatus according to any one of claims 1 to 4, characterized in that... : The setting information includes one or more of the following: the type of document, the purpose of the document, the type of input data, the number of output characters, the information that the document should contain, and the information that the document should not contain.

8. The information processing apparatus according to any one of claims 1 to 4, characterized in that... : The input data includes sound data, or text data that is textualized from the sound data; The documents include daily reports or meeting minutes.

9. The information processing apparatus according to any one of claims 1 to 4, characterized in that... : The input data includes multiple document data; The document includes a summary of the data from the multiple documents.

10. An information processing system for creating documents based on input data, characterized in that, The information processing system includes: The acquisition unit acquires setting information related to the creation of the document; The document production department, based on the input data and the specified information, uses a large-scale language model to produce the document; and The output section outputs the documents created by the document creation section.

11. The information processing system according to claim 10, characterized in that... : It further includes a conversational AI control unit to control the conversational AI that engages in conversations with the user; The acquisition unit acquires the setting information set by the user using the dialogue AI.

12. An information processing method, characterized in that, A computer that creates a document based on input data performs the following processing: The process involves acquiring and processing settings related to the creation of the document. Document creation processing, based on the input data and the defined information, uses a large-scale language model to create the document; and Output processing: Outputs the document created in the document creation process.

13. A computer program product having computer program instructions stored thereon, characterized in that, The computer program instructions cause the computer that creates the document based on the input data to perform the following processes: The process involves acquiring and processing settings related to the creation of the document. Document creation processing, based on the input data and the defined information, uses a large-scale language model to create the document; and Output processing: Outputs the document created in the document creation process.

Images