Support system, support method, and support program

Abstract

To support a worker by reducing a burden of work record creation work.SOLUTION: An assistance system includes a sound collecting device, an terminal, and an assistance device, the sound collecting device includes a sound collection control unit, the terminal includes a voice-information acquiring unit and a voice-information converting unit, and the assistance device includes a text-information acquiring unit that acquires a work schedule of a worker and acquires text information from the terminal, a work-target identifying unit that identifies a work target of the worker, a performed-work identifying unit that identifies, as performed work, work corresponding to the work target among pieces of work included in the work schedule, a work-record control unit that creates a work record including the work target and stores the work record in association with the performed work, an unexecuted work extracting unit that extracts, as unexecuted work, work with which no work record is associated among pieces of work included in the work schedule, and a transmitting unit that transmits the unexecuted work to the terminal.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a support system, a support method, and a support program, and more particularly to a support system, a support method, and a support program suitable for supporting people engaged in caregiving. [Background technology]

[0002] The use of IT (Information Technology), ICT (Information and Communications Technology), and IoT (Internet of Things) has been promoted to improve productivity in the field of nursing care. Support systems such as nursing care record systems, nursing care billing systems, and user management systems are becoming widespread in nursing care settings. As an example of a support system, a nursing care support system has been proposed in which staff (workers) belonging to nursing care-related businesses share nursing care information required to provide services to users who are the targets of their work (see Patent Document 1).

[0003] Creating work records for the work performed in providing such services is an important task. Work records are records of the services provided by the worker to the user, the user's health status, progress of symptoms, and activity status.

[0004] If all work records are written by hand, it is said that the amount of work spent on creating work records by workers can reach 30% of the total work, making reducing the labor required for creating work records a major challenge. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2010-205263 Summary of the Invention [Problem to be solved by the invention]

[0006] Therefore, an object of the present invention is to provide a support system, a support method, and a support program that can support workers by reducing the burden of creating work records. [Means for solving the problem]

[0007] That is, the support system according to the first aspect is a support system that supports a worker engaged in assisting work, and includes a sound collection device, a terminal device, and the support device. The sound collection device includes a sound collection control unit that acquires voice information of the worker while he or she is working. The terminal device includes a voice information acquisition unit that acquires voice information from the sound collection device and a voice information conversion unit that converts the voice information into text information. The support device includes a work schedule acquisition unit that acquires the worker's work schedule for a predetermined period of time, a text information acquisition unit that acquires text information from the terminal device, a work object identification unit that identifies the worker's work object based on the text information, a performed work identification unit that identifies, from the work schedule, work that corresponds to the work object as performed work, a work record control unit that creates a work record including the work object and stores the work record in association with the performed work, an unperformed work extraction unit that extracts, from the work schedule, work that does not have an associated work record as unperformed work, and a transmission unit that transmits the unperformed work to the terminal device.

[0008] In a second aspect, in the support system according to the first aspect, the support device may further include a work type identification unit that identifies the type of work to be performed by the worker based on text information, and the completed work identification unit may identify, from among the work included in the work schedule, work that corresponds to the work target and work type as completed work, and the work record control unit may create a work record that includes the work target and work type and store the work record in association with the completed work.

[0009] In a third aspect, in the support system according to the first aspect, the voice information acquisition unit may acquire the date and time when the voice information was acquired as date and time information together with the voice information, the text information acquisition unit may acquire the date and time information together with the text information, and the work record control unit may store the date and time information in the work record.

[0010] In a fourth aspect, in the support system according to the first aspect, the terminal device may have the name of the worker using the terminal device registered therein, and the work record control unit may record the name of the worker using the terminal device that acquired the text information in the work record.

[0011] In a fifth aspect, in the assistance system according to the first aspect, the work object identification unit may estimate the work object corresponding to the text information by inputting the text information acquired by the text information acquisition unit into a learning model that has previously learned the correspondence between the text information and the work object.

[0012] In a sixth aspect, in the support system according to the fifth aspect, the work target is a person receiving assistance work, the learning model pre-learns the correspondence between the text information and at least one of the person's first name, surname, first name, and nickname, and the work target identification unit may input the text information acquired by the text information acquisition unit into the learning model, thereby estimating the first name, surname, first name, or nickname of the person who is the work target and corresponds to the text information.

[0013] In a seventh aspect, in the support system according to the second aspect, the work type identification unit may estimate the work type corresponding to the text information by inputting the text information acquired by the text information acquisition unit into a learning model that has previously learned the correspondence between text information and work types.

[0014] In an eighth aspect, in the assistance system according to the seventh aspect, the work types are identified by registered names pre-registered for each work type, the learning model pre-learns the correspondence between the text information and the registered names, and the work type identification unit inputs the text information acquired by the text information acquisition unit into the learning model, thereby estimating the registered name corresponding to the text information.

[0015] In a ninth aspect, in the support system according to the first aspect, the support system may further include a quantity identification unit that identifies the quantity and unit of quantity related to the worker's work by performing named entity extraction processing on the text information, and the work record control unit may store the quantity and unit of quantity related to the worker's work in the work record associated with the completed work.

[0016] In a tenth aspect, in the support system according to the ninth aspect, the work schedule includes a standard amount that the amount of work performed by the worker must meet for each task included in the work schedule, and the support system further includes a judgment unit that judges whether the amount of work performed by the worker identified by the quantity identification unit exceeds the standard amount of the completed task identified by the completed task identification unit, and the work record control unit includes the content determined by the judgment unit in the work record associated with the completed task, and the unperformed task extraction unit extracts the completed task as unperformed task when the judgment unit determines that the amount of work performed by the worker does not exceed the standard amount of the completed task.

[0017] The support method according to an eleventh aspect is a support method used in a support system that supports a worker engaged in assistance work, and that includes a sound collection device, a terminal device, and a support device, wherein the sound collection device includes a sound collection control unit that acquires voice information of the worker while he or she is working, and the terminal device includes a voice information acquisition unit that acquires voice information from the sound collection device and a voice information conversion unit that converts the voice information into text information, and the calculation unit of the support device executes the following steps: a work schedule acquisition step that acquires the worker's work schedule for a predetermined period of time; a text information acquisition step that acquires text information from the terminal device; a work object identification step that identifies the worker's work object based on the text information; a performed work identification step that identifies, from the work schedule, work that corresponds to the work object as performed work; a work record control step that creates a work record including the work object and stores the work record in association with the performed work; an unperformed work extraction step that extracts, from the work schedule, work that does not have an associated work record as unperformed work; and a transmission step that transmits the unperformed work to the terminal device.

[0018] The support program according to a twelfth aspect is a support program used in a support system that supports a worker engaged in assistance work and that includes a sound collection device, a terminal device, and the support device. The sound collection device includes a sound collection control unit that acquires voice information while the worker is working, and the terminal device includes a voice information acquisition unit that acquires voice information from the sound collection device and a voice information conversion unit that converts the voice information into text information. The calculation unit of the support device is characterized by realizing a work schedule acquisition function that acquires the worker's work schedule for a specified period of time, a text information acquisition function that acquires text information from the terminal device, a work object identification function that identifies the worker's work object based on the text information, a performed work identification function that identifies work included in the work schedule that corresponds to the work object as performed work, a work record control function that creates a work record including the work object and stores the work record in association with the performed work, an unperformed work extraction function that extracts work included in the work schedule that does not have a work record associated with it as unperformed work, and a transmission function that transmits the unperformed work to the terminal device. [Effects of the Invention]

[0019] The support system etc. according to the present invention is a support system that supports a worker engaged in assisting work, and includes a sound collection device, a terminal device, and a support device, wherein the sound collection device includes a sound collection control unit that acquires voice information of the worker while he or she is working, the terminal device includes a voice information acquisition unit that acquires voice information from the sound collection device and a voice information conversion unit that converts the voice information into text information, and the support device includes a work schedule acquisition unit that acquires the worker's work schedule for a predetermined period of time, a text information acquisition unit that acquires text information from the terminal device, a work object identification unit that identifies the worker's work object based on the text information, a performed work identification unit that identifies, from the work schedule, work that corresponds to the work object as performed work, a work record control unit that creates a work record including the work object and stores the work record in association with the performed work, an unperformed work extraction unit that extracts, from the work schedule, work that does not have an associated work record, as unperformed work, and a transmission unit that transmits the unperformed work to the terminal device, thereby reducing the burden of creating work records and supporting the worker. [Brief explanation of the drawings]

[0020] [Figure 1] FIG. 1 is a diagram showing an example of the system configuration of a support system including a sound collection device, a terminal device, and a support device according to this embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of the hardware configuration of the sound collection device according to this embodiment. [Figure 3] FIG. 3 is a block diagram showing an example of the hardware configuration of the terminal device according to this embodiment. [Figure 4] FIG. 4 is a block diagram showing an example of the hardware configuration of the support device according to this embodiment. [Figure 5] FIG. 5 is a block diagram showing an example of the functional configuration of the sound collection device according to this embodiment. [Figure 6] FIG. 6 is a block diagram showing an example of the functional configuration of a terminal device according to this embodiment. [Figure 7]FIG. 7 is a block diagram showing an example of the functional configuration of the support device according to this embodiment. [Figure 8] FIG. 8 is a diagram for explaining an example of a work schedule according to this embodiment. [Figure 9] FIG. 9 is a diagram for explaining an example of a display screen on the display of the terminal device according to this embodiment. [Figure 10] FIG. 10 is a diagram for explaining an example of a display screen of a display of a terminal device according to this embodiment. [Figure 11] FIG. 11 is a diagram for explaining an example of a display screen of a display of a terminal device according to this embodiment. [Figure 12] FIG. 12 is a flowchart of the firmware of the sound collection device according to this embodiment. [Figure 13] FIG. 13 is a flowchart of the drive program for the terminal device according to this embodiment. [Figure 14] FIG. 14 is a flowchart of the assistance program according to this embodiment. [Figure 15] FIG. 15 is a flowchart of a support program according to another embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0021] A support system 2 according to this embodiment will be described with reference to FIGS. The support system 2 includes a sound collection device 10, a terminal device 20, and a support device 30, and supports a worker 1 engaged in care work. The sound collection device 10, the terminal device 20, and the support device 30 each operate autonomously and cooperate with each other to realize the function of the support system 2. Assistance means providing assistance to a person with the activities necessary for daily life, regardless of whether or not it interferes with daily life. The support system 2 can be used not only in the fields of care and welfare, but also for the purpose of supporting people who work in jobs that involve caring for and supporting others. In the support system 2, the sound collection device 10, the terminal device 20, and the support device 30 are each depicted as one device for convenience in Fig. 1, but there may be multiple devices of each. That is, the support system 2 may include multiple sound collection devices 10, terminal devices 20, and support devices 30.

[0022] (Outline of the support device 30) The support device 30 is an information processing device represented by a server, a personal computer (hereinafter referred to as PC), a notebook PC, a tablet PC, a smartphone, etc., and is connected to an information communication network 40 and is connected to other devices such as a terminal device 20 so as to be able to perform two-way information communication. The support device 30 is connected via an information and communication network 40 to a sound collection device 10 and a terminal device 20 used at the site where the worker 1 is being assisted.

[0023] The support device 30 may be used in an on-premise operation mode or an off-premise operation mode. An on-premise operating mode refers to a mode in which hardware is installed within a company and the system is operated independently, i.e., the hardware of the support device 30 is installed within the company's facilities and the support device 30 is used by having the user operate it directly or by having the user operate it remotely via an information and communication network 40.

[0024] On the other hand, the off-premise operation mode refers to an operation mode in which the support device 30 is constructed using a system within hardware operated and managed by another company, and is operated, managed, etc., and made available to users via the information and communication network 40. An example of an off-premise operation mode is constructing the support device 30 using a cloud service provided by another company, such as AWS (registered trademark) from Amazon (registered trademark) or Azure (registered trademark) from Microsoft (registered trademark). The support device 30 may be managed and operated by the organization to which the worker 1 belongs, or the organization to which the worker 1 belongs may borrow the support device 30 operated by a person other than the organization to which the worker 1 belongs and allow the worker 1 to use it.

[0025] The sound collection device 10 is a headset, which is a headphone with a microphone, and is worn on the head of the worker 1 to collect the sound emitted by the worker 1 while working. Note that in addition to a headset, an intercom, an earphone microphone, a pin microphone, or the like may also be used as the sound collection device 10. The sound collection device 10 is connected to the terminal device 20 via wireless communication so as to be able to communicate bidirectionally. The wireless communication here may use a short-range wireless communication standard such as Bluetooth (registered trademark), Wi-Fi (registered trademark), or ZigBee (registered trademark). The sound collection device 10 and the terminal device 20 may be connected not only via wireless communication but also via wired communication. The sound collection device 10 is connected to the support device 30 via the terminal device 20 so as to be capable of two-way communication.

[0026] The terminal device 20 is an information processing device such as a PC, or a portable information communication terminal represented by a notebook PC, a tablet PC, a smartphone, or the like, and is connected to other devices such as the support device 30 via an information communication network 40 so as to be able to perform two-way information communication. The terminal device 20 is placed near the worker 1 by being carried by the worker 1, and serves as a relay point for communication of the sound collection device 10.

[0027] (Hardware configuration of sound collection device 10) The hardware configuration of the sound collection device 10 will be described with reference to Fig. 2. Fig. 2 is a block diagram showing an example of the hardware configuration of the sound collection device 10. The sound collection device 10 includes a communication interface 10a, a ROM (Read Only Memory) 10b, a RAM (Random Access Memory) 10c, a processing unit 10d, and an input / output interface 10e. The communication interface 10a has a function of transmitting and receiving data to and from the terminal device 20 based on short-range wireless communication standards such as Bluetooth, Wi-Fi, and ZigBee. The ROM 10b can be used as a recording device, and stores firmware required for the operation of the sound collection device 10, various applications, and various data used by the various applications. The RAM 10c is used to configure a main memory accessed by the calculation unit 10d, and may also be used to temporarily store audio information acquired by the sound collection device 10, which will be described later. The calculation unit 10d is used to execute the firmware of the sound collection device 10, and includes a central processing unit (CPU), a micro processing unit (MPU), a graphics processing unit (GPU), etc., and may be realized by a logic circuit (hardware) formed in an integrated circuit (IC (Integrated Circuit) chip, LSI (Large Scale Integration)), etc., or a dedicated circuit. The input / output interface 10e transmits and receives data to and from external devices and internal devices of the sound collection device 10. Examples of external devices and internal devices include a microphone 10f and a speaker 10g. The microphone 10f performs the sound collection function of the headset, and the speaker 10g performs the electroacoustic conversion function of converting an electrical signal in the headset into an acoustic signal.

[0028] (Hardware configuration of terminal device 20) The hardware configuration of the terminal device 20 will be described with reference to Fig. 3. Fig. 3 is a block diagram showing an example of the hardware configuration of the terminal device 20. The terminal device 20 includes a communication interface 20a, a ROM 20b, a RAM 20c, a storage unit 20d, a calculation unit 20e, an input / output interface 20f, and the like. The calculation unit 20e includes a central processing unit (CPU), an MPU, a GPU, etc., and may be realized by a logic circuit (hardware) formed by an integrated circuit (IC chip, LSI), etc., or a dedicated circuit. The communication interface 20a has a function of transmitting and receiving data to and from the sound collection device 10, and transmitting and receiving data to and from the support device 30 via the information communication network 40. The ROM 20b stores the basic bootstrap program of the terminal device 20 and basic operating instructions of the OS (basic system). The RAM 20c is used to configure the main memory that the calculation unit 20e accesses. The memory unit 20d is realized by an HDD (Hard Disk Drive), an SSD (Solid State Drive), online storage, etc., and stores the operating programs, OS (Operating System), various applications, and various data used by the applications required for the terminal device 20 to operate. The input / output interface 20f transmits and receives data to and from external devices and internal devices of the terminal device 20. The external devices and internal devices of the terminal device 20 include a microphone 20g, a camera 20h, and a touch screen 20i, which are used as input devices, and a display 20j and a speaker 20k, which are used as output devices.

[0029] (Hardware configuration of the support device 30) The hardware configuration of the support device 30 will be described with reference to Fig. 4. Fig. 4 is a block diagram showing an example of the hardware configuration of the support device 30. The support device 30 includes a communication interface 30a, a ROM 30b, a RAM 30c, a storage unit 30d, a calculation unit 30e, an input / output interface 30f, and the like. The calculation unit 30e includes a central processing unit (CPU), an MPU, a GPU, etc., and may be realized by a logic circuit (hardware) formed on an integrated circuit (IC chip, LSI), etc., or a dedicated circuit. The communication interface 30 a has a function of transmitting and receiving data to and from the terminal device 20 via the information communication network 40 . The ROM 30b stores the basic bootstrap program of the support device 30 and basic operating instructions of the OS (basic system). The RAM 30c is used to configure the main memory that is accessed by the calculation unit 30e. The memory unit 30d is realized by an HDD (hard disk drive), SSD (solid state drive), online storage, etc., and can be used as a recording device, and stores the assistance program described below, the OS (basic system) required for the assistance device 30 to operate, various applications, and various data used by the applications. Furthermore, the storage unit 30d may be constructed as a database, and the work record control unit 33f (described later) may store work records in the database storage unit 30d. Databases are suitable for organizing and managing large amounts of data, and make it possible to efficiently organize, manage, and analyze the vast amount of information in work records accumulated over many years. The input / output interface 30f transmits and receives data to and from an input device 31 and an output device 32, which are external devices of the support device 30. The input device 31 includes a keyboard 31a, a mouse 31b, a scanner 31c, and a microphone 31d. The output device 32 includes a monitor 32a, a speaker 32b, and a printer 32c.

[0030] (Functional configuration of sound collection device 10) Next, the functional configuration of the sound collection device 10 will be described with reference to Fig. 5. Fig. 5 is a block diagram showing an example of the functional configuration of the sound collection device 10. The sound collection device 10 loads the firmware stored in the ROM 10b into a main memory configured with the RAM 10c, etc. The calculation unit 10d accesses the main memory into which the firmware has been loaded and executes the firmware. By executing the firmware, the sound collection device 10 has a calculation unit 10d equipped with functional units such as a sound collection control unit 11a, a storage control unit 11b, and a communication unit 11c.

[0031] The sound collection control unit 11a acquires voice information of the worker 1 while he or she is working. The sound collection control unit 11a converts the sound collected by the microphone 10f of the sound collection device 10 into a digital signal and acquires it as sound information. The voice information refers to information obtained by converting the voice uttered by the worker 1 while working into a digital signal. The voice uttered by the worker 1 while working refers to a conversation between the worker 1 and the user who is the target of the work, the worker 1 talking to himself while working, a conversation with someone nearby the worker 1, the voice of the worker 1 talking on the phone, etc. It is to be noted that the voice uttered by the worker 1 while working is not limited to the voice uttered by the worker 1, but may also be the voice uttered by the worker 1 when the worker 1 is not working, or before or after work.

[0032] The storage control unit 11b stores the acquired voice information in the RAM 10c. The storage control unit 11b may store the audio information acquired by the sound collection control unit 11a and before being transmitted to the terminal device 20 in the RAM 10c for temporary storage.

[0033] The communication unit 11c transmits the voice information stored in the RAM 10c to the terminal device 20 via the communication interface 10a. The sound collection control unit 11a, the storage control unit 11b, and the communication unit 11c are repeatedly executed at predetermined time intervals, and the sound information acquired by the sound collection device 10 is transmitted to the terminal device 20 in real time (instantaneously).

[0034] (Functional configuration of terminal device 20) Next, the functional configuration of the terminal device 20 will be described with reference to Fig. 6. Fig. 6 is a block diagram showing an example of the functional configuration of the terminal device 20. The terminal device 20 loads the operating program stored in the ROM 20b or the storage unit 20d into a main memory configured with the RAM 20c, etc. The calculation unit 20e accesses the main memory into which the operating program has been loaded and executes the operating program. By executing the driving program of the terminal device 20, the calculation unit 20e is provided with functional units such as a voice information acquisition unit 21a, a voice information conversion unit 21b, a memory control unit 21c, a display control unit 21d, and a receiving unit 21e.

[0035] The voice information acquisition unit 21 a acquires voice information from the sound collection device 10 . The audio information acquisition unit 21a acquires audio information acquired by the sound collection device 10, and stores the audio information in the RAM 20c or the storage unit 20d. The voice information acquisition unit 21a is repeatedly executed at predetermined time intervals to acquire voice information transmitted from the sound collection device 10 and store it in the RAM 20c or the storage unit 20d.

[0036] The voice information conversion unit 21b converts voice information into text information. The voice information conversion unit 21b converts the voice information stored in the RAM 20c or the storage unit 20d into text information using voice recognition processing. The voice information conversion unit 21b checks at predetermined time intervals whether or not there is voice information in RAM 20c or memory unit 20d that has not yet been converted to text information, and is executed if there is voice information in RAM 20c or memory unit 20d that has not yet been converted to text information.

[0037] The speech recognition process may employ, for example, deep neural network (DNN)-hidden Markov model (HMM) speech recognition, end-to-end speech recognition, or the like. The voice recognition process removes noise contained in the voice information as a preprocessing step for converting the voice information into text information. Noise refers to white noise, environmental sounds, and voices of people other than the worker 1 wearing the sound collection device 10. There are various existing techniques for removing noise, such as a method for removing noise by extracting or removing a predetermined frequency band using a filter circuit, or a method for removing noise by using a learning model that has previously learned the voice information of the worker 1 by machine learning.

[0038] The storage control unit 21c stores the text information converted by the voice information conversion unit 21b in the storage unit 20d in association with the original voice information. The storage control unit 21c may store the text information together with the original audio information in the storage unit 20d, or alternatively, the original audio information may not be stored in the storage unit 20d, and only the text information associated with the original audio information may be stored in the storage unit 20d.

[0039] The display control unit 21d controls the display of various pieces of information acquired by the terminal device 20 on the display 20j. Examples of various information include text information obtained by the voice information conversion unit 21b, information received by the receiving unit 21e from the support device 30, such as the work schedule of the worker 1 within a specified period, and unperformed work extracted by the unperformed work extraction unit 33g.

[0040] The receiving unit 21e receives the unperformed work transmitted by the transmitting unit 33h of the support device 30. The receiving unit 21e receives not only the unperformed tasks but also various information transmitted by the transmitting unit 33h of the support device 30. The information received by the receiving unit 21e is stored in the RAM 20c or the memory unit 20d of the terminal device 20.

[0041] (Functional configuration of the support device 30) Next, the functional configuration of the support device 30 will be described with reference to Fig. 7. Fig. 7 is a block diagram showing an example of the functional configuration of the support device 30. The assistance device 30 loads an assistance program (described later) stored in the ROM 30b or the storage unit 30d into a main memory configured with the RAM 30c, etc. The calculation unit 30e accesses the main memory into which the assistance program has been loaded and executes the assistance program. By executing the assistance program, the assistance device 30 has functional units such as a work schedule acquisition unit 33a, a text information acquisition unit 33b, a work target identification unit 33c, a work type identification unit 33d, a completed work identification unit 33e, a work record control unit 33f, an unperformed work extraction unit 33g, a transmission unit 33h, a quantity identification unit 33i, and a judgment unit 33j in the calculation unit 30e.

[0042] The work schedule acquisition unit 33a acquires the work schedule of the worker 1 for a predetermined period. The work schedule may be a work schedule for one worker 1, or may include work schedules for other workers 1. The work schedule acquisition unit 33a acquires a work schedule that has been created in advance. The work schedule for worker 1 within a specified period is a timeline of the work that worker 1 plans to perform within a specified period, and is a scheduled work schedule. Specifically, the work schedule for worker 1 within a specified period is displayed by the time of worker 1's visit to the home of the user, who is the target of the work in worker 1's daily visiting service. Note that the work is not limited to visiting service, but may also be services within a nursing facility as part of day care service and residential service. Visiting services involve visiting users (those requiring nursing care and assistance) living at home and providing support for daily life, nursing care, rehabilitation, etc. Day care services involve users (those requiring nursing care and assistance) who live at home coming to a facility to receive assistance with eating and toileting, rehabilitation, and health care. Inpatient services refer to services in which users (those requiring nursing care or assistance) are admitted to a facility for a certain period of time and provided with assistance with eating and toileting, rehabilitation, nursing care, etc.

[0043] The work schedule for a specified period for day care services and residential services is determined by the planned stay at the facility for the users who are the target of the work (see Table 1 below). The table shown in Table 1 may be used as the work schedule for workers. Workers engaged in day care services and residential services confirm whether each user is at the facility as scheduled, and record the actual period of stay at the facility for each user in the work record. Table 1 shows the schedule for a specific day (YY / ZZ / 20XX) and displays the planned stay time and assistance type for each user. For example, user W is scheduled to stay from 9:00 to 16:00, and the assistance type is "independent." The type of assistance refers to the level of assistance required by the user, and can include independence, supervision, partial assistance, and full assistance. Independence means that the user requires almost no assistance except when requested by the worker. Monitoring means that the worker needs to speak to the user as appropriate, but little assistance is required unless the user requests it from the worker. Partial assistance means that the user needs assistance from a worker to perform a specific action. Full assistance means that the worker needs to assist the user with all of their actions. Workers engaged in day care services and residential care services may record, as part of their work records, the actual length of stay for each user, as well as whether the user bathed, whether they urinated, whether they took medication, or how much they ate during their stay.

[0044] [Table 1]

[0045] The support system 2 according to this embodiment will be described using a visiting service schedule 41 shown in FIG. 8 as an example of a work schedule for a predetermined period. An example of a work schedule for a predetermined period will be described with reference to Fig. 8. Fig. 8 is a diagram for explaining an example of a work schedule according to this embodiment. In Fig. 8, a schedule 41 for a visiting service among home nursing care services is taken as an example of the work schedule for a predetermined period. Work by workers engaged in the visiting service is performed in accordance with schedule 41, which defines the types of work to be provided to users who are the targets of the work. The schedule 41 shows the work schedule for each worker on a specific day (20XX, YY month, ZZ day). The schedule 41 includes a worker column 41a, a work content column 41b, and a date display area 41c. The worker column 41a displays the name of the worker (which may be a first name, last name, first name, nickname, or identification code, etc.). In Fig. 8, for convenience, the multiple workers 1 are displayed as worker A, worker B, worker C, and worker D. The work content column 41b shows, as a Gantt chart, the planned time for each worker 1 to stay at the user's home during the visiting service. The date display area 41c displays the date of the scheduled work indicated in the work content column 41b. The work schedule acquisition unit 33a acquires the following work schedule by acquiring a schedule 41 as the work schedule of the worker 1 within a predetermined period. Visiting services are divided into physical care and daily living assistance, with physical care divided into 30-minute time slots, while daily living assistance is divided into time slots of less than 45 minutes or more than 45 minutes. The specific work content of the visiting service can be displayed on the schedule using the unit code for the visiting care service.

[0046] Physical care refers to care services that mainly involve physical contact, and specifically includes assistance with excretion, eating, bathing, wiping, grooming, changing clothes, changing positions, moving and transferring, getting up and going to bed, and taking medication. In addition to physical care, there are other services such as those to support users' independence and prevent their condition from worsening, as well as other services that utilize specialized knowledge and skills. Assisted living refers to services that support the daily life of users other than physical care, specifically care that involves performing household chores on behalf of elderly people, etc. Specific examples of assistance with daily living include cleaning, laundry, bed making, cooking, shopping, and picking up prescription medications.

[0047] The work schedule 42 for Worker A on ZZ day, YY month, 20XX is shown below. First task 42a: 10:15~10:45. Body 2 work at AA's house. *Physical 2 is a unit code and refers to physical care provided for 30 minutes or more but less than 1 hour. 2nd work 42b: 11:30~12:00. Body 1 work at BB's house. *Physical 1 is a unit code and refers to physical care provided for 20 minutes or more but less than 30 minutes. 3rd task 42c: 12:45~14:45. Physical 2 and lifestyle 2 tasks at CC's house. *Physical 2, Lifestyle 2 is a unit code and refers to the provision of physical care for 1.5 hours to less than 2 hours followed by life support for 45 minutes to less than 70 minutes.

[0048] Work schedule 43 for Worker B on ZZ / YY / 20XX is shown below. First task 43a: 8:30-9:00. Body 2 work at DD's house. 2nd task 43b: 9:30~11:15. Physical 2, Lifestyle 1 task at EE's house. *Physical 2, Lifestyle 1 is a unit code, and refers to providing physical care for 1.5 hours or more but less than 2 hours, followed by daily life assistance for 20 minutes or more but less than 45 minutes.

[0049] Work schedule 44 for Worker C on ZZ day, YY month, 20XX is shown below. First task 44a: 12:00~12:30. Working on body 2 at FF's house. 2nd work 44b: 13:15~13:45. Body 1 work at GG's house. 3rd task 44c: 14:30~16:30. Physical 2 and lifestyle 2 tasks at HH's house.

[0050] Work schedule 45 for Worker D on ZZ / YY / 20XX is shown below. First task 45a: 11:00-11:30. Body 2 work at Mr. II's house. 2nd work 45b: 12:00~13:45. Work on body 2 and life 1 at JJ's house.

[0051] In addition, in all cases of home visit services, day care services, and residential care services, vital signs and physical measurements of the users who are the subject of the work are measured as part of daily record management during the care service.

[0052] The transmission unit 33h transmits the work schedule acquired by the work schedule acquisition unit 33a to the terminal device 20 used by the worker 1. The terminal device 20 used by the worker 1 receives the work schedule through the receiving unit 21e, and the work schedule is stored in the storage unit 20d by the storage control unit 21c. The display control unit 21d of the terminal device 20 displays the work schedule stored in the storage unit 20d on the display 20j.

[0053] The work schedule displayed on the display 20j will be described with reference to Fig. 9. Fig. 9 is a diagram for explaining an example of a display screen of the display 20j of the terminal device 20. In Fig. 9, the terminal device 20 used by worker A is shown as an example of the terminal device 20 of worker 1. The terminal device 20 in FIG. 9 displays a display screen 50 showing the work schedule for worker A for the day. The display screen 50 has a title 50a, a display area 51 for the first task, a scheduled time 51a for the first task, a display area 52 for the second task, a scheduled time 52a for the second task, a display area 53 for the third task, and a scheduled time 53a for the third task. The title 50a displays a name assigned in relation to the display content of the display screen 50. The title 50a shown in FIG. 9 is "Daily visit schedule for worker A." The display area 51 for the first job displays a unit code indicating the destination and work content of the first job of worker A. The display area 51 for the first job shown in FIG. 9 displays AA's house (body 2). The scheduled time 51a for the first task displays the scheduled time (10:15 to 10:45) for engaging in the first task. The display area 52 for the second job displays a unit code indicating the destination and work content of the second job of worker A. The display area 52 for the second job shown in FIG. 9 displays BB's house (body 1). The second work scheduled time 52a displays the scheduled time (11:30 to 12:00) for engaging in the second work. The display area 53 for the third task displays a unit code indicating the destination and task content of the third task of worker A. The display area 53 for the third task shown in FIG. 9 displays CC's house (body 2, life 2). The scheduled time 53a for the third work displays the scheduled time (12:45 to 14:45) for engaging in the third work.

[0054] In the display areas 51 to 53 of each task, a message board icon 50b, a camera icon 50c, and an execution icon 50d are displayed. The bulletin board icon 50b is an icon for displaying the bulletin board for each task, and a bulletin board (not shown) is displayed by touching it. The "●" displayed on the right side of the bulletin board icon 50b is a badge indicating that there are unread messages. The bulletin board displays messages from the service provider manager, who is the worker's superior, and other workers in a timeline display format. Messages can include not only text information, but also photos, video messages, audio messages, links, stamps, and the like, which can be shared with others. The camera icon 50c is touched to take a photograph by activating the camera 20h built into the terminal device 20. The photograph taken by the camera 20h can be attached to a bulletin board, e-mail, or the like.

[0055] The execution icon 50d indicates whether each task has been executed or not. A check mark is displayed on the execution icon 50d of a task that has been identified as a completed task by a completed task identification unit 33e (to be described later). On the other hand, the execution icon 50d of a task extracted as an unexecuted task by the unexecuted task extraction unit 33g (described later) is displayed with a "cross mark." The display modes of the bulletin board icon 50b, the camera icon 50c, and the implementation icon 50d are not limited to those described above, and various display modes can be used, and can be changed by the user of the terminal device 20.

[0056] At the bottom of the display screen of the terminal device 20, a My Records button 50e, an All Records button 50f, a Contact by Voice button 50g, and a My Page button 50h are displayed. The "My Record" button 50e is a button that is touched to display the user's own work schedule. The record all button 50f is a button that is touched to display the work schedules of all workers 1, that is, the worker's own work schedule and the work schedules of other workers. The voice contact button 50g is a touch button that activates a call function and allows a user to make a voice call to another person. The My Page button 50h is a button that is touched to display My Page, which allows the user to check and change the settings of the terminal device 20 that the user uses.

[0057] The text information acquisition unit 33b acquires text information from the terminal device 20. The text information acquisition unit 33b acquires text information converted from the voice information by the voice information conversion unit 21b of the terminal device 20. The text information acquisition unit 33b stores the acquired text information in the storage unit 30d. When the text information is transmitted from the terminal device 20 together with the original voice information, the text information acquisition unit 33b stores the acquired text information together with the original voice information in the storage unit 30d.

[0058] The work object specifying unit 33c specifies the work object of the worker 1 based on the text information. The work object specifying unit 33c specifies the work object of the worker 1 based on the text information stored in the storage unit 30d by the text information acquiring unit 33b. The work object identification unit 33c estimates the work object corresponding to the text information by inputting the text information acquired by the text information acquisition unit 33b into a learning model that has previously learned the correspondence between text information and work objects. The work target is a person receiving assistance work, and the learning model pre-learns the correspondence between the text information and at least one of the person's first name, surname, first name, and nickname. The work target identification unit 33c inputs the text information acquired by the text information acquisition unit 33b into the learning model, thereby estimating the first name, surname, first name, or nickname of the work target person corresponding to the text information. In detail, the learning model learns in advance the correspondence between specific words (keywords) contained in the text information and the work object. The specific words (keywords) are a person's first name, last name, first name, and nickname, and are associated with identification information assigned to each work object so that the person can be uniquely identified. The identification information refers to a work object ID, such as a registration number for identifying an individual, and is assigned to each worker. The work object identification unit 33c determines whether or not a specific word (keyword) is included in the text information, and if the specific word (keyword) is included, identifies a work object associated with the specific word (keyword).

[0059] The method by which the work object specifying unit 33c estimates the person who is the work object can also use existing technology. A specific example of the method by which the work object specifying unit 33c estimates the person who is the work object will be described below. The work target identification unit 33c extracts named entities (Named Entities) contained in the text information by performing Named Entity Recognition (NER), a natural language processing technique, on the text information stored in the memory unit 30d, and classifies them into predefined named entity categories such as proper nouns (personal names, organization names, place names, etc.), dates, time expressions, quantities, amounts, and percentages. Next, the work target identification unit 33c estimates the person contained in the text information by inputting the named entities classified as proper nouns into a learning model that has previously machine-learned the registered names (first name, surname, given name, and nickname) of the person to be worked on. The names (first name, last name, first name, and nickname) of the person to be worked on are registered in advance as shown in Table 2 below. Table 2 is used as training data for the learning model. Instead of using named entity recognition (NER), named entities contained in text information may be extracted by analyzing dependency relationships in natural language processing. Dependency relationships represent the modifier or subordinate relationship between words or phrases in a sentence. Examples of dependency relationships include subject-predicate relationships, adverbial modifier relationships, and oblique complement relationships.

[0060] [Table 2]

[0061] Table 2 shows examples of registered names (first name, last name, first name, and nickname) of the person who will be the work target. The item names in each column of Table 2 are "Work target ID," "First name," "Last name," "First name," and "Nickname." The work target ID is unique identification information assigned to each person who will receive the assistance work, who is the work target. "First name" is information indicating the first name of the person who will be the work target. "Last name" is information indicating the last name of the person who will be the work target. "First name" is information indicating the first name of the person who will be the work target. "Nickname" is information indicating the nickname of the person who will be the work target. For example, the registered names (first name, surname, given name, and nickname) in the first row of Table 2 are: "Work Target ID" p001, "First Name" AA, "Surname" Aaa, "First Name" Abc, and "Nickname" A.

[0062] For example, when a unique entity extracted from text information and classified as a proper noun includes nickname A, the work object identification unit 33c estimates that the person included in the text information has work object ID "p001" by inputting the person's registered name (first name, surname, given name, and nickname) shown in Table 2 into a learning model that has been machine-learned in advance.

[0063] The work target specification unit 33c may extract honorifics (for example, san, sama, and chan) included in the text information by executing named entity extraction (NER) processing on the text information stored in the storage unit 30d. In this case, it is estimated that the part of the text information immediately before the honorific is the registered name (given name, surname, given name, and nickname). Furthermore, the work object identification unit 33c may use a large language model (LLM) instead of a learning model that learns in advance the correspondence between specific words (keywords) included in the text information and the work objects. In this case, the large language model learns the correspondence between a person's first name, last name, first name, and nickname and the identification information of the work object during the fine-tuning stage, making it possible to extract the work object from the text information.

[0064] The work type identification unit 33d identifies the work type of the worker 1 based on the text information. The task type identification unit 33d estimates the task type corresponding to the text information by inputting the text information acquired by the text information acquisition unit 33b into a learning model that has previously learned the correspondence between text information and task types. The work types are identified by registered names that are pre-registered for each work type, and the learning model pre-learns the correspondence between text information and registered names. The work type identification unit 33d inputs the text information acquired by the text information acquisition unit 33b into the learning model, thereby estimating the registered name corresponding to the text information. Specifically, the learning model pre-learns the correspondence between specific words (keywords) contained in the text information and task types. The specific words (keywords) are the type keywords shown in Table 3, and are associated with the registered names assigned to each task type so that the task type can be uniquely identified. The work type identification unit 33d determines whether or not a specific word (keyword) is included in the text information, and if the specific word (keyword) is included, identifies the work type associated with the specific word (keyword).

[0065] Existing technology can be used as a method by which the task type identifying unit 33d estimates the task type of the worker 1. A specific example of a method by which the task type identifying unit 33d estimates the task type of the worker 1 will be described below. The work type identification unit 33d performs named entity extraction processing (NER) on the text information stored in the memory unit 30d to extract named entities (Named Entities) contained in the text information and classify them into predefined named entity categories such as proper nouns (personal names, organization names, place names, etc.), dates, time expressions, quantities, amounts, and percentages. Next, the work type identification unit 33d estimates the work type of worker 1 contained in the text information by inputting the named expressions classified as proper nouns into a learning model that has previously been machine-learned to use type keywords that correspond to registered names that have been pre-registered as work types of worker 1.

[0066] The registered names corresponding to the work types of worker 1 are registered in advance as shown in Table 3 below. Examples of registered names of work types include urination assistance, eating assistance, bathing assistance, medication assistance, vital sign measurement, and physical measurement. The category keywords corresponding to the above registered names are registered in advance as shown in Table 3 below. For example, the category keywords corresponding to the registered name "urinary assistance" in the first row of Table 3 are "urination" and "pee." For example, if the named entities extracted from the text information and classified as proper nouns include "body temperature," the work type identification unit 33d estimates that the work type of worker 1 included in the text information is "vital sign measurement" by inputting type keywords that correspond to the registered names shown in Table 3 into a learning model that has been trained by machine learning in advance.

[0067] [Table 3]

[0068] As with the task target identification unit 33c, the task type identification unit 33d may use a large-scale language model instead of a learning model that pre-learns the correspondence between specific words (keywords) included in the text information and task types. In this case, the large-scale language model learns the correspondence between the type keywords shown in Table 3 and the registered names assigned to the task types during the fine-tuning stage, making it possible to extract task types from text information.

[0069] 10 and 11, a display screen and an input screen for the work record of the worker 1 on the terminal device 20 will be described. Fig. 10 and Fig. 11 are diagrams for explaining an example of the display screen of the display of the terminal device 20. In FIG. 10, the display screen for daily record management will be explained as an example. Daily record management refers to recording and managing the content of the assistance provided, as well as the condition and reaction of the person receiving the assistance, each time assistance is performed, and the content of this record can vary widely. Specifically, daily record management records the date, time, and location of the assistance provided, the specific content of the assistance provided, the health condition, facial expressions, conversations, and other information passed on to other workers. Referring to FIG. 10, the display screen for daily record management will be described using the work record of the first work 42a of worker A as an example.

[0070] The first task 42a of worker A includes a first task, a second task, and a third task, where the first task is measuring vital signs, the second task is physical examination, and the third task is regular medication. Regular medication refers to taking medication at a set time every day. The daily record management display screen 60 includes a title 60a, a display setting button 60b, a date and time display area 60c, a service type display area 60d, a first implementation content display area 61, a second implementation content display area 62, and a third implementation content display area 63. The title 60a displays the assigned name related to the display content of the display screen 60. The title 60a shown in FIG. 10 is "List of Achievements." The "List of Achievements" in the title 60a refers to a list display of the tasks performed by worker A as the first task 42a. The display setting button 60b is a button that is touched when changing the display format of the display screen 60. The location and date / time display area 60c displays the location and date / time at which the first task 42a was performed. The location and date / time display area 60c shown in Figure 10 displays "20XX year YY month ZZ day AA's house 10:15-10:45". The service type display area 60d displays the type of service for the first task 42a of worker A. The service type display area 60d shown in Fig. 10 displays "visiting service." In addition to visiting service, the types of services include day service and residential service.

[0071] The first implementation content display area 61, the second implementation content display area 62, and the third implementation content display area 63 are areas where the specific content performed as the first task 42a is displayed. The first implementation content display area 61 includes an operation type 61a, an operation time 61b, and a record display field 61c. The second implementation content display area 62 includes an operation type 62a, an operation time 62b, and a record display field 62c. The third implementation content display area 63 includes an operation type 63a, an operation time 63b, and a record display field 63c. For example, the display area 61 of the first implementation content shown in FIG. 10 is displayed as follows: Work Type 61a: Vital Sign Measurement Implementation time 61b: Implementation 10:15-10:25 Record display section 61c: Body temperature 36.4℃, blood pressure 110 / 80mmHg

[0072] Next, with reference to FIG. 11, the voice input for vital sign measurement in daily record management will be described using the work record of the first work 42a of worker A as an example. As described above, the voice input of worker A is carried out by converting the voice information collected by the sound collection device 10 into text information by the terminal device 20, and the work object and type of work are identified based on this text information. The display screen 70 displays a vital sign measurement recording / editing screen, and includes a back button 70a, a title 70b, an implementation time input area 71, a start time input field 71a, an end time input field 71b, a body temperature input area 72, a body temperature input field 72a, a blood pressure input area 73, an upper blood pressure input field 73a, a lower blood pressure input field 73b, and a memo input area 74. The back button 70a is a button that is touched when the user wishes to return to the previous display screen. The title 70b displays the name assigned in relation to the display content of the display screen 70. The title 70b shown in FIG. 11 is "Vital Sign Measurement Recording and Editing." A start time input field 71a and an end time input field 71b are displayed in the implementation time input area 71. The start time input field 71a and end time input field 71b shown in Fig. 11 display the start time (10:15) and end time (10:25) of the implementation of the first implementation content of the first task 42a. A body temperature input field 72a is displayed in the body temperature input region 72. The body temperature input field 72a shown in Fig. 11 displays the body temperature measured in the first task 42a, 36.4 (°C). An upper blood pressure input field 73a and a lower blood pressure input field 73b are displayed in the blood pressure input area 73. The blood pressures measured in the first task 42a, 110 (mmHg) and 80 (mmHg), are displayed in the upper blood pressure input field 73a and the lower blood pressure input field 73b shown in Fig. 11 . The memo input area 74 is where worker A's memorandum and handover information to other workers are input.

[0073] A method for identifying the amount contained in text information will be described below. Quantity refers, for example, to degrees (°C) and millimeters (mmHg) in vital sign measurements, and kilograms (kg) and centimeters (cm) in body measurements. The quantity specifying unit 33i specifies the quantity and the unit of the quantity related to the work of the worker 1 by performing named entity extraction processing on the text information.

[0074] The amount specifying unit 33i can specify the amount using existing technology. A specific example of the method used by the amount specifying unit 33i to specify the amount will be described below. The quantity identification unit 33i performs named entity extraction (NER) processing on the text information stored in the memory unit 30d to extract named entities (Named Entities) contained in the text information and classify them into predefined named entity categories such as proper nouns (personal names, organization names, place names, etc.), dates, time expressions, quantities, amounts, and percentages.

[0075] Next, the quantity specifying unit 33i acquires a quantity keyword corresponding to the work type specified by the work type specifying unit 33d. The quantity keywords corresponding to the work type are acquired based on pre-acquired Table 3. For example, if the work type is vital sign measurement, the quantity keywords are "degrees (°C)" and "millimeters (mmHg)," and if the work type is physical measurement, the quantity keywords are "kilograms (kg)" and "centimeters (cm)."

[0076] Next, the quantity specification unit 33i compares the acquired quantity keywords with named entities classified into the quantitative named entity classification among the named entities included in the text information. The quantity specification unit 33i checks whether there is a quantity keyword among the named entities classified into the quantitative named entity classification, and if there is, specifies that the quantity keyword is a quantity related to vital sign measurement. The quantity specifying unit 33i specifies a quantity related to vital sign measurement based on the specified quantity keyword. Examples of quantities related to vital sign measurement are "36.4 degrees (°C)" and "110 mmHg and 80 mmHg" specified by the quantity specifying unit 33i. When the quantity keyword specified by the quantity specifying unit 33i is a unit such as "degrees (°C)", a numerical value including the unit is specified as the quantity related to vital sign measurement. In addition, if the quantity keyword identified by the quantity identification unit 33i is a keyword that represents the quantity itself, such as ``a lot,'' ``a little,'' ``all,'' or ``half,'' the quantity keyword itself may be identified as the quantity related to the work type identified by the work type identification unit 33d.

[0077] The completed work identifying unit 33e identifies, from among the works included in the work schedule, a work that corresponds to the work target and work type as a completed work. The completed work identification unit 33e identifies, from among the work included in the work schedule, work that corresponds to the work object identified by the work object identification unit 33c and the work type identified by the work type identification unit 33d as completed work. The completed work identified by the completed work identifying unit 33e is transmitted by the transmitting unit 33h to the terminal device 20. The terminal device 20 displays a check mark on the completed icon 50d of the work identified as the completed work in the daily work schedule of worker A shown in FIG.

[0078] The work record control unit 33f creates a work record including the work target and the work type, and stores the work record in association with the completed work. The work record control unit 33f creates a work record including the work object identified by the work object identification unit 33c and the work type identified by the work type identification unit 33d, and stores the work record in association with the completed work identified by the completed work identification unit 33e. The work record control unit 33f stores the work record in the storage unit 30d.

[0079] The voice information acquisition unit 21a of the terminal device 20 acquires the date and time when the voice information was acquired as date and time information along with the voice information, the text information acquisition unit 33b of the support device 30 acquires the date and time information along with the text information, and the work record control unit 33f stores the date and time information in the work record. The terminal device 20 has the name of the worker 1 who uses the terminal device 20 registered, and the work record control unit 33f records the name of the worker 1 who uses the terminal device 20 that obtained the text information in the work record.

[0080] The unperformed work extraction unit 33g extracts, from among the works included in the work schedule, works that are not associated with a work record as unperformed works. The unperformed work extraction unit 33g extracts as unperformed work, from among the work included in the work schedule acquired by the work schedule acquisition unit 33a, work whose work record has not been stored in the memory unit 30d by the work record control unit 33f, or work that has not been identified as performed work by the performed work identification unit 33e.

[0081] The transmission unit 33h transmits the unperformed work to the terminal device 20. The transmission unit 33h transmits to the terminal device 20 the unperformed tasks extracted by the unperformed task extraction unit 33g. The terminal device 20 displays a "cross mark" on the execution icon 50d of the work identified as an unexecuted work in the daily work schedule of worker A shown in FIG.

[0082] The determination unit 33j determines whether the amount of work performed by the worker 1 identified by the amount identification unit 33i exceeds the reference amount of completed work identified by the completed work identification unit 33e. The reference amount of completed work may be determined in advance, and the work schedule acquisition unit 33a may acquire the work schedule and also acquire the reference amount corresponding to the work included in the work schedule. The work schedule includes a reference amount that the amount of work performed by worker 1 must satisfy for each work included in the work schedule. The work record control unit 33f includes the content determined by the determination unit 33j in the work record associated with the completed work, and the unperformed work extraction unit 33g extracts the completed work as unperformed work when the determination unit 33j determines that the amount of work related to worker 1 does not exceed the standard amount of completed work.

[0083] In addition, work schedules for day care services and residential services are determined based on the facility stay schedule of the users who are the subject of the work. In this case, the completed work identification unit 33e identifies, from among the works included in the work schedule, the work that corresponds to the work target. Specifically, from among the works included in the work schedule, the completed work identification unit 33e identifies, as completed work, the work that corresponds to the user who is the work target identified by the work target identification unit 33c. The work record control unit 33f then creates a work record including the work target and stores the work record in association with the completed work. Specifically, the work record control unit 33f creates a work record including the user who is the work target identified by the work target identification unit 33c, and stores the work record in association with the completed work identified by the completed work identification unit 33e. Tasks for which no associated task record exists in day care services and inpatient services indicate that the user who is the subject of the task is not present at the scheduled time of stay. The transmitting unit 33h transmits to the terminal device 20 the name of the user who will not be present during the scheduled stay time, the work target ID, and the like. The terminal device 20 displays on the display 20j (not shown) the names of users who have been identified as the unperformed tasks in the worker's daily work schedule.

[0084] (About the firmware of the sound collection device 10) The firmware of the sound collection device 10 will be described with reference to Fig. 12. Fig. 12 is a flowchart of the firmware of the sound collection device 10. The firmware of the sound collection device 10 includes a sound collection control step S11a, a storage control step S11b, and a communication step S11c. The sound collection device 10 executes the firmware using the calculation unit 10d, thereby realizing a sound collection control function, a storage control function, a communication function, and the like. These functions are executed in the order shown in the flowchart of Fig. 12, but the order can be changed as needed. Note that each function overlaps with the description of the various functional units of the sound collection device 10 described above, so detailed description thereof will be omitted.

[0085] The sound collection control function acquires voice information of the worker 1 while he or she is working (S11a: sound collection control step). The storage control function stores the acquired audio information in the RAM 10c (S11b: storage control step). The communication function transmits the voice information stored in the RAM 10c to the terminal device 20 via the communication interface 10a (S11c: communication step).

[0086] (Regarding the driving program of the terminal device 20) The drive program of the terminal device 20 will be described with reference to Fig. 13. Fig. 13 is a flowchart of the drive program of the sound collection device 10. The driving program of the sound collection device 10 includes an audio information acquisition step S21a, an audio information conversion step S21b, a storage control step S21c, a display control step S21d, and a reception step S21e. The terminal device 20 executes the driving program by the calculation unit 20e, thereby realizing functions such as an audio information acquisition function, an audio information conversion function, a storage control function, a display control function, and a reception function. These functions are executed in the order shown in the flowchart of Fig. 13, but the order can be changed as needed. Note that each function overlaps with the description of the various functional units of the terminal device 20 described above, and therefore detailed description thereof will be omitted.

[0087] The voice information acquisition function acquires voice information from the sound collection device 10 (S21a: voice information acquisition step). The voice information conversion function converts voice information into text information (S21b: voice information conversion step). The storage control function stores the text information converted by the voice information conversion function in association with the original voice information in the storage unit 20d (S21c: storage control step). The display control function controls the display of various pieces of information acquired by the terminal device 20 on the display 20j (S21d: display control step). The receiving function receives the unperformed work transmitted by the transmitting function of the support device 30 (to be described later) (S21e: receiving step).

[0088] (About support methods and support programs) Next, an assistance program according to one embodiment of the present invention will be described together with an assistance method with reference to Fig. 14. Fig. 14 is a flowchart of the assistance program according to this embodiment. The assistance method is executed by the calculation unit 30e of the assistance device 30 based on the assistance program. The assistance program includes a work schedule acquisition step S33a, a text information acquisition step S33b, a work object identification step S33c, a completed work identification step S33e, a work record control step S33f, an uncompleted work extraction step S33g, a transmission step S33h, a quantity identification step S33i, and a judgment step S33j. The assistance program causes the calculation unit 30e of the assistance device 30 to perform functions such as a work schedule acquisition function, a text information acquisition function, a work target identification function, a completed work identification function, a work record control function, an uncompleted work extraction function, a transmission function, a quantity identification function, and a judgment function. These functions are executed in the order shown in the flowchart of Fig. 14, but the order can also be changed as appropriate. Note that each function overlaps with the description of the various functional units of the assistance device 30 described above, and therefore detailed description thereof will be omitted.

[0089] The work schedule acquisition function acquires the work schedule of the worker 1 within a predetermined period (S33a: work schedule acquisition step).

[0090] The text information acquisition function acquires text information from the terminal device 20 (S33b: text information acquisition step).

[0091] The work object identification function identifies the work object of the worker 1 based on the text information (S33c: work object identification step).

[0092] The completed work identification function identifies, from among the works included in the work schedule, the work that corresponds to the work target, as completed work (S33e: completed work identification step).

[0093] The work record control function creates a work record including the work target, and stores the work record in association with the completed work (S33f: work record control step).

[0094] The unperformed work extraction function extracts, from among the works included in the work schedule, works that do not have associated work records as unperformed works (S33g: unperformed work extraction step).

[0095] The transmission function transmits the unperformed work to the terminal device 20 (S33h: transmission step).

[0096] The quantity identification function identifies the quantity and the unit of quantity related to the work of the worker 1 by performing named entity extraction processing on the text information (S33i: quantity identification step).

[0097] The determination function determines whether the amount of work related to the worker 1 identified by the amount identifying unit 33i exceeds the reference amount of completed work identified by the completed work identifying unit 33e (S33j: determination step).

[0098] (Other embodiments of the support method and support program) Next, an assistance program according to another embodiment of the present invention will be described together with an assistance method according to another embodiment with reference to Fig. 15. The flowchart of the assistance program according to another embodiment shown in Fig. 15 differs from the flowchart of the assistance program shown in Fig. 14 in that a task type identification step S33d has been added. Accordingly, the contents of the completed task identification step S33e and the work record control step S33f of the assistance program according to another embodiment shown in Fig. 15 also differ from those of the completed task identification step S33e and the work record control step S33f of the assistance program shown in Fig. 14. The assistance method according to the other embodiment is executed by the calculation unit 30e of the assistance device 30 based on the assistance program according to the other embodiment shown in FIG. The assistance program according to another embodiment shown in FIG. 15 includes a work schedule acquisition step S33a, a text information acquisition step S33b, a work target identification step S33c, a work type identification step S33d, a completed work identification step S33e, a work record control step S33f, an unperformed work extraction step S33g, a transmission step S33h, a quantity identification step S33i, and a judgment step S33j.

[0099] 15 causes the calculation unit 30e of the assistance device 30 to perform functions such as a work schedule acquisition function, a text information acquisition function, a work target identification function, a work type identification function, a completed work identification function, a work record control function, an uncompleted work extraction function, a transmission function, a quantity identification function, and a judgment function. These functions are executed in the order shown in the flowchart of FIG. 15, but the order can also be changed as appropriate. The support method and support program according to another embodiment shown in FIG. 15 will be described below, focusing only on the differences from the support method and support program shown in FIG. Furthermore, since each function overlaps with the description of the various functional units of the support device 30 described above, detailed description thereof will be omitted.

[0100] The work type identification function identifies the work type of the worker 1 based on the text information (S33d: work type identification step).

[0101] The completed work identification function identifies, from among the works included in the work schedule, works that correspond to the work target and work type as completed works (S33e: completed work identification step).

[0102] The work record control function creates a work record including the work target and the work type, and stores the work record in association with the completed work (S33f: work record control step).

[0103] According to the support system 2 of the above-described embodiment, a work record can be generated and stored based on the voice uttered by the worker 1, so that the worker 1 can create a work record while working, thereby reducing the burden on the worker 1.

[0104] Furthermore, according to the support system 2 of the above-described embodiment, the unperformed work extracted by the unperformed work extraction unit 33g is transmitted to the terminal device 20 and displayed, thereby informing the worker 1 that there is unperformed work, and preventing the worker 1 from forgetting to perform the scheduled work.

[0105] Furthermore, according to the support system 2 of the above-described embodiment, the completed work identified by the completed work identification unit 33e is sent to the terminal device 20 and displayed, thereby making it possible to inform the worker 1 of the completed work and to inform the worker 1 of the progress of the scheduled work.

[0106] Furthermore, according to the support system 2 of the above-described embodiment, the work object identification unit 33c can identify the user who is the work object by the user's first name, surname, first name, and nickname, which are registered as specific words (keywords), so that the work object can be identified in the worker's natural conversation.

[0107] The present invention is not limited to the assistance system 2, assistance method, and assistance program according to the above-described embodiments, and can be embodied in various other modified examples or application examples without departing from the gist of the present invention as set forth in the claims. Also, although the term "data" is used in the above-described embodiments, the term "data" can be replaced with "information," and the term "information" can be replaced with "data." [Explanation of symbols]

[0108] 1. Worker 2. Support System 10 Sound collection device 10a Communication Interface 10b ROM 10c RAM 10d Arithmetic unit 10e Input / Output Interface 10f Microphone 10g speaker 11a Sound collection control section 11b Memory control unit 11c Communication Department 20 Terminal equipment 20a Communication Interface 20b ROM 20c RAM 20d storage section 20e Calculation unit 20f Input / Output Interface 20g microphone 20h Camera 20i Touchscreen 20j display 20k speaker 21a Voice information acquisition unit 21b Voice information conversion unit 21c Memory control unit 21d Display control unit 21e Receiver 30 Support equipment 30a Communication Interface 30b ROM 30c RAM 30d storage section 30e Arithmetic unit 30f Input / Output Interface 31 Input Devices 31a keyboard 31b Mouse 31c scanner 31d Mike 32 Output Devices 32a Monitor 32b speaker 32c printer 33a Work Schedule Acquisition Section 33b Text information acquisition section 33c Work Object Identification Section 33d Work Type Identification Section 33e Completed Work Identification Section 33f Work record control section 33g Unperformed work extraction section 33h Transmitter 33i Quantity identification part 33j Judgment section 40 Information and Communication Networks 41 Schedule 41a Worker row 41b Work content column 41c Date display area 42 Work schedule for worker A 42a First Work 42b Second Work 42c Third Work 43 Worker B's work schedule 43a First Work 43b 2nd work 44 Worker C's work schedule 44a First Work 44b Second Work 44c 3rd work 45 Work schedule for worker D 45a First Work 45b Second Work 50 Worker A's daily work schedule display screen 50a Title 50b bulletin board icon 50c camera icon 50d implementation icon 50e My Record Button 50f All Recording Button 50g Voice contact button 50h My Page button 51 Display area for first operation 51a Scheduled time for first operation 52 Display area for second operation 52a Scheduled time for second operation 53 Third operation display area 53a Scheduled time for the third operation 60 Daily Record Management Display Screen 60a Title 60b Display settings button 60c Location and date display area 60d Service type display area 61 Display area for first implementation details 61a Work Type 61b Implementation time 61c Record display column 62 Display area for second implementation details 62a Work Type 62b Implementation time 62c Record display column 63 Third implementation content display area 63a Work Type 63b Implementation time 63c Record display column 70 Vital Sign Measurement Recording / Editing Screen 70a Back button (return to daily record management display screen 60) 70b Title 71 Implementation time input field 71a Disclosure time input field 71b End time input field 72 Body Temperature Input Area 72a Physical strength input field 73 Blood Pressure Input Area 73a Upper blood pressure input field 73b Blood pressure input field below 74 Memo input area

Claims

1. A support system that includes a sound collection device, a terminal device, and a support device and supports a worker engaged in care work, The sound collection device is a sound collection control unit for acquiring voice information of the worker while he or she is working, The terminal device registers the name of the worker who uses the terminal device, a voice information acquisition unit that acquires, from the sound collection device, a date and time when the voice information was acquired as date and time information together with the voice information; a voice information conversion unit that converts the voice information into text information, The support device includes: a schedule showing the worker's work schedule for a predetermined period, a worker column showing the worker's name as the work schedule, and the planned time of stay at the user's home receiving assistance work from the worker as a Gantt chart for each task, and a work content column showing the planned stay time and type of assistance for each user; and a work schedule acquisition unit acquiring a standard amount that the amount of work related to the worker for each task included in the work schedule must meet; a text information acquisition unit that acquires the date and time information together with the text information from the terminal device; a work object identification unit that executes named entity extraction processing using natural language processing technology on the text information to extract named entities contained in the text information, classifies the entities into predefined named entity categories of proper nouns, dates, time expressions, quantities, amounts, and percentages, and identifies the work object of the worker; a task type identification unit that identifies a task type of the worker based on the text information; a completed work identification unit that identifies, from among the works included in the work schedule, a work that corresponds to the work target and the work type as a completed work; a work record control unit that records the name of the worker who uses the terminal device that acquired the text information in the work record, creates a work record that includes the work object, and stores the date and time information in the work record in association with the completed work; an unperformed work extraction unit that extracts, from among the works included in the work schedule, works that are not associated with the work record as unperformed works; a transmission unit that transmits the unperformed work to the terminal device; a quantity specifying unit that specifies a quantity related to the work of the worker and a unit of the quantity by performing a named entity extraction process on the text information; a determination unit that determines whether the amount of work related to the worker identified by the amount identification unit exceeds a reference amount of completed work identified by the completed work identification unit, The work record control unit stores the amount and the unit of the amount related to the work of the worker in the work record associated with the completed work, and includes the content determined by the determination unit in the work record associated with the completed work, The unperformed work extraction unit extracts the performed work as an unperformed work when the determination unit determines that the amount of work performed by the worker does not exceed a reference amount of the performed work. A support system characterized by:

2. The assistance system according to claim 1, characterized in that the work object identification unit estimates the work object corresponding to the text information by inputting the text information acquired by the text information acquisition unit into a learning model that has previously learned a correspondence between the text information and the work object.

3. the work target is a person receiving the assistance work, the learning model learns in advance a correspondence relationship between text information and at least one of the person's first name, last name, first name, and nickname; The assistance system according to claim 2, characterized in that the work object identification unit estimates the first name, surname, given name, or nickname of the person who is the work object corresponding to the text information by inputting the text information acquired by the text information acquisition unit into the learning model.

4. The assistance system according to claim 1, characterized in that the task type identification unit estimates the task type corresponding to the text information by inputting the text information acquired by the text information acquisition unit into a learning model that has previously learned the correspondence between the text information and the task type.

5. The work types are identified by registered names that are registered in advance for each of the work types, The learning model learns in advance the correspondence between text information and the registered name, The assistance system according to claim 4, characterized in that the task type identification unit estimates the registered name corresponding to the text information by inputting the text information acquired by the text information acquisition unit into the learning model.

6. A support method used in a support system that supports a worker engaged in care work, the support system including a sound collection device, a terminal device, and a support device, comprising: The sound collection device is a sound collection control unit for acquiring voice information of the worker while he or she is working, The terminal device registers the name of the worker who uses the terminal device, a voice information acquisition unit that acquires, from the sound collection device, a date and time when the voice information was acquired as date and time information together with the voice information; a voice information conversion unit that converts the voice information into text information, The calculation unit of the support device a schedule showing the worker's work schedule for a predetermined period, a worker column showing the worker's name as the work schedule, and the planned time of stay at the user's home receiving assistance work from the worker as a Gantt chart for each task, and a work content column showing the planned stay time and type of assistance for each user; and a work schedule acquisition step for acquiring a standard amount that the worker's work quantity for each task included in the work schedule must meet; a text information acquisition step of acquiring the date and time information together with the text information from the terminal device; a work object identification step of extracting named entities contained in the text information by executing a named entity extraction process using natural language processing technology on the text information, and classifying the named entities into predefined named entity categories of proper nouns, dates, time expressions, quantities, amounts, and percentages, thereby identifying the work object of the worker; a task type identification step of identifying a task type of the worker based on the text information; a completed work identification step of identifying, from among the works included in the work schedule, a work that corresponds to the work target and the work type as a completed work; a work record control step of recording the name of the worker who uses the terminal device that acquired the text information in the work record, creating a work record including the work object, and storing the date and time information in the work record in association with the completed work; an unperformed work extraction step of extracting, from among the works included in the work schedule, works to which the work record is not associated as unperformed works; a transmission step of transmitting the unperformed work to the terminal device; a quantity identification step of identifying a quantity related to the work of the worker and a unit of the quantity by performing named entity extraction processing on the text information; a determination step of determining whether or not the amount of work related to the worker identified in the amount identification step exceeds a reference amount of completed work identified in the completed work identification step; The work record control step includes storing the amount and the unit of the amount related to the work of the worker in the work record associated with the completed work, and includes the content determined by the determination unit in the work record associated with the completed work, The unperformed work extraction step extracts the performed work as an unperformed work when the determination step determines that the amount of work performed by the worker does not exceed a reference amount of the performed work. A support method characterized by:

7. An assistance program used in an assistance system that includes a sound collection device, a terminal device, and an assistance device and that assists a worker engaged in care work, The sound collection device is a sound collection control unit for acquiring voice information of the worker while he or she is working, The terminal device registers the name of the worker who uses the terminal device, a voice information acquisition unit that acquires, from the sound collection device, a date and time when the voice information was acquired as date and time information together with the voice information; a voice information conversion unit that converts the voice information into text information, The calculation unit of the support device a schedule showing the worker's work schedule for a predetermined period, a worker column showing the worker's name as the work schedule, and the expected time of stay at the user's home receiving assistance work from the worker as a Gantt chart for each task, and a work content column showing the expected stay time and type of assistance for each user; and a work schedule acquisition function for acquiring a standard amount that the worker's work volume for each task included in the work schedule must meet; a text information acquisition function for acquiring the date and time information together with the text information from the terminal device; a work target identification function that executes named entity extraction processing using natural language processing technology on the text information to extract named entities contained in the text information, classifies them into predefined named entity classifications of proper nouns, dates, time expressions, quantities, amounts, and percentages, and identifies the work target of the worker; a task type identification function for identifying a task type of the worker based on the text information; a completed work identification function that identifies, from among the works included in the work schedule, a work that corresponds to the work target and the work type as a completed work; a work record control function that records the name of the worker who uses the terminal device that acquired the text information in the work record, creates a work record that includes the work object, and stores the date and time information in the work record in association with the completed work; an unperformed work extraction function that extracts, from among the works included in the work schedule, works that are not associated with the work records as unperformed works; a transmission function for transmitting the unperformed work to the terminal device; a quantity identification function that identifies a quantity related to the work of the worker and a unit of the quantity by performing named entity extraction processing on the text information; a determination function for determining whether or not the amount of work related to the worker identified by the amount identification function exceeds a reference amount of completed work identified by the completed work identification function; The work record control function stores the amount and unit of the amount related to the work of the worker in the work record associated with the completed work, and includes the content determined by the determination unit in the work record associated with the completed work, The unperformed work extraction function extracts the performed work as an unperformed work when the determination function determines that the amount of work related to the worker does not exceed the reference amount of the performed work. A support program characterized by:

Images