Order assistance device, method for order assistance using the same, and order assistance program executed by the same

Abstract

To provide an order assistance technology capable of maintaining or further improving customer satisfaction while achieving labor savings.SOLUTION: An order assistance device (2) assists order operations in a restaurant. The device includes: identification means (22) for identifying a customer terminal (1); digital human display means (25) for displaying and operating, on the customer terminal (1), a digital human image created on the basis of a real person; conversation means (26) for performing a conversation through natural language processing using an artificial voice created on the basis of the voice of the person via the customer terminal (1); order extraction means (26) for extracting order details from the content of the conversation; and order means (27) for transmitting order information to a kitchen terminal (4).SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to ordering assistance technology in restaurants. [Background technology]

[0002] As of 2024, there is a serious labor shortage, making it difficult to secure workers. This is particularly evident in the food and beverage industry, such as restaurants. Even if part-time workers can be hired, there are large individual differences in customer service skills. It also takes time to train them in customer service skills.

[0003] Therefore, efforts are being made to reduce the number of staff members as much as possible. For example, self-order processing is being considered (see, for example, Patent Document 1). This will enable optimal labor allocation, such as allowing staff to concentrate their efforts on cooking itself. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2020-135846 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the specifications of self-ordering systems vary from restaurant to restaurant, and customers who are not familiar with operating the ordering terminals may find it confusing. Furthermore, customers who enjoy eating out do not just want to satisfy their appetite, but also want to enjoy a casual conversation with the waiter. In particular, the higher the order unit price, the more important satisfaction beyond just appetite becomes. Therefore, there is a risk that pursuing efficiency too much will result in a decline in customer satisfaction.

[0006] The present invention is intended to solve the above-mentioned problems, and aims to provide an order assistance technology that can maintain or even improve customer satisfaction while reducing the labor required. [Means for solving the problem]

[0007] The present invention, which solves the above-mentioned problems, is a device for assisting restaurant ordering. The ordering assistance device includes an identification means for identifying a customer terminal, a digital human display means for displaying and operating a digital human image created based on a real person on the customer terminal, a conversation means for conducting a conversation via the customer terminal using natural language processing with an artificial voice created based on the voice of the real person, and an order extraction means for extracting the order details from the content of the conversation.

[0008] With this device, a virtual store clerk, a digital human, assists customers with ordering as part of a conversation. Because the digital human is based on a real person, it is expected to have a natural conversation. As a result, it is well suited to assisting customers with ordering.

[0009] In this way, the device of the present invention can maintain or even improve customer satisfaction while reducing the number of workers required.

[0010] In the above invention, the ordering assistance device is preferably capable of communicating with a person registration server that stores digital human images and artificial voices of multiple real people, and is equipped with a waiter hiring means for hiring one or more virtual waiters, and a remuneration calculation means for calculating a remuneration based on the order performance of the hired virtual waiter.

[0011] This allows real people to register as virtual servers and receive rewards.

[0012] In the above invention, preferably, the order assistance device is capable of communicating with a person registration server that stores digital human images and artificial voices of multiple real people, and is equipped with a waiter hiring means for hiring two or more virtual waiters, a waiter selection means for selecting one of the hired virtual waiters, and a remuneration calculation means for calculating a remuneration based on the order performance of the selected virtual waiter.

[0013] This allows you to hire multiple people in the same way as hiring real part-time workers. It also gives customers a choice of waiters. Each virtual waiter has a different salary. As a result, it is possible to estimate the potential demand for real people.

[0014] In the above invention, preferably, the remuneration calculation means evaluates the order record based on the number of conversations from the start of an order to the confirmation of the order and the time from the start of an order to the confirmation of the order.

[0015] This will allow for a short but effective conversation, which will likely encourage customers to place orders.

[0016] In the above invention, preferably, the remuneration calculation means evaluates the order performance based on the order amount.

[0017] This is expected to have the effect of encouraging more orders.

[0018] In the above invention, preferably, the real person is a waiter who can serve in the store, and the system comprises waiter enumeration means for enumerating waiters who are currently available to serve as virtual waiters, waiter selection means for selecting one of the listed virtual waiters, and waiter preparation command means for commanding the waiter corresponding to the selected virtual waiter to prepare to serve.

[0019] This allows a customer to select a virtual store clerk and receive service from a real store clerk who corresponds to the virtual store clerk.

[0020] In the above invention, it is preferable to have a compatibility judgment means for judging the compatibility between the selected virtual employee and the customer based on the ordering history of the selected virtual employee, and the serving preparation command means considers whether to command the employee corresponding to the selected virtual employee to prepare for serving if the compatibility is poor. As a result, if the customer and the virtual store clerk are compatible, the customer can receive service from a real store clerk who is responsive to the virtual store clerk.

[0021] In the above invention, preferably, the compatibility determination means evaluates the order record based on the number of conversations from the start of the order to the order confirmation and the time from the start of the order to the order confirmation.

[0022] This will allow for a short but effective conversation, which will likely encourage customers to place orders.

[0023] In the above invention, preferably, the compatibility determining means evaluates the order record based on the order amount.

[0024] This is expected to have the effect of encouraging more orders.

[0025] The present invention, which solves the above-mentioned problems, is a method for assisting restaurant ordering using an information processing device. It identifies a customer terminal, displays and operates a digital human image based on a real person on the customer terminal, converses with the customer through the customer terminal using an artificial voice created based on the person's voice using natural language processing, and extracts the order details from the content of the conversation.

[0026] The program of the present invention, which solves the above-mentioned problems, assists ordering in a restaurant by causing an information processing device to perform the following processes: identifying a customer terminal; displaying and operating a digital human image created based on a real person on the customer terminal; conducting a conversation via the customer terminal using natural language processing with an artificial voice created based on the person's voice; and extracting the order details from the content of the conversation. [Effects of the Invention]

[0027] The order assistance technology of the present invention can maintain or even improve customer satisfaction while reducing manpower requirements. [Brief explanation of the drawings]

[0028] [Figure 1] Overall system configuration diagram (basic embodiment) [Figure 2] Server configuration diagram [Figure 3] Processing flow chart (basic embodiment) [Figure 4] Conceptual diagram of accompanying processing (first embodiment) [Figure 5] Conceptual diagram of accompanying processing (first embodiment) [Figure 6] Functional block diagram (first embodiment) [Figure 7] Sequence diagram (first embodiment) [Figure 8] Accompanying processing flow chart (first embodiment) [Figure 9] Conceptual diagram of accompanying processing (second embodiment) [Figure 10] Conceptual diagram of accompanying processing (second embodiment) [Figure 11] Functional block diagram (second embodiment) [Figure 12] Sequence diagram (second embodiment) [Figure 13] Accompanying processing flow chart (second embodiment) DETAILED DESCRIPTION OF THE INVENTION

[0029] ~Basic implementation form~ Figure 1 shows the overall configuration of the system of the present application. The system assists customers in ordering at restaurants and other food service establishments. While store staff may assist as needed, the system is characterized in that customers' orders are generally placed by a digital human acting as a virtual store staff member, rather than by a real store staff member. In particular, the digital human in this application is created based on a real person.

[0030] In this application, virtual store clerk, virtual waiter, and digital human are roughly similar concepts, but a virtual store clerk is the opposite concept to an actual store clerk, and a virtual waiter is the opposite concept to an actual waiter. Store clerk and waiter are also roughly similar concepts, but store clerk includes waiters. A digital human is an image that is reproduced by a computer to look exactly like a human, and the image is controlled by artificial intelligence and can converse like a human.

[0031] The system of the present invention comprises a customer terminal 1, an order assistance server 2, a personnel management server 3, and a kitchen terminal 4. For ease of explanation, the order assistance server 2 and the personnel management server 3 are shown as separate components, but they may also be configured as a single component. Conversely, the functions of the order assistance server 2 and the personnel management server 3 may be located on separate servers connected via a network.

[0032] The customer terminal 1 is assumed to be a personal terminal such as a smartphone owned by a customer who uses a store. This does not exclude tablet computers, personal computers, etc. There are multiple customers, and each customer has their own customer terminal.

[0033] For example, when the customer terminal 1 reads a two-dimensional barcode attached to each seat in the store, the customer terminal 1 and the order assistance server 2 can communicate with each other via a network line.

[0034] Alternatively, it may be a dedicated terminal provided at each seat in the restaurant. Each dedicated terminal can communicate with the order assistance server 2. The order assistance server 2 identifies each dedicated terminal.

[0035] The order assistance server 2 and the personnel management server 3 can communicate with each other. The order assistance server 2 can also communicate with the kitchen terminal 4. The communication may be wireless or wired. The order assistance server 2 is the core component of the system of the present invention, and will be described later.

[0036] The human resource management server 3 manages digital humans. In this application, the digital humans are created based on real people.

[0037] The physical characteristics of a real person are registered as data in advance via a camera or other device, and the data is then reproduced as computer graphics.

[0038] Furthermore, the actual voice of a real person is registered as data via a microphone or the like, and the data is reproduced as an artificial voice.

[0039] The facial expressions and voice of the digital human are linked, allowing the person to act as a digital human in the system. Specifically, the person can assist with ordering as a virtual store clerk.

[0040] The kitchen terminal 4 is a tablet installed in the kitchen or a smartphone owned by the chef. A large monitor may be installed separately to display order information.

[0041] 2 is a hardware configuration diagram of the order assistance server 2. The order assistance server 2 is an information processing device 100.

[0042] The information processing device 100 includes a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, a non-volatile memory 104, and an I / O 105, and each component is connected to each other via a bus 106.

[0043] The CPU 101 executes each function based on a program. The ROM 102 stores the program. The RAM 103 is used as a temporary work area for the CPU 101. The non-volatile memory 104 is an example of a storage device that maintains stored information even if the power supply is cut off. For example, it is a semiconductor memory or a hard disk. The non-volatile memory 104 does not need to be built-in, and may be removable and portable. The I / O 105 is an input / output interface.

[0044] To the I / O 105, for example, a communication unit 107, an input unit 108, a display unit 109, and the like are connected.

[0045] The communication unit 107 is connected to a communication line and has a communication protocol for communicating data with the outside. The input unit 108 is a touch panel, keyboard, mouse, etc., and is used to input operation instructions, etc. The display unit 109 is, for example, a liquid crystal display, an organic EL display, a projector, etc., and displays the results of information processing, etc., and may be displayed on the screen of another terminal via the communication line.

[0046] The order assistance device 2 includes an identification unit 22 that identifies the customer terminal 1, a digital human display unit 25 that displays and operates a digital human image created based on an actual person on the customer terminal 1, a conversation unit 26 that converses via the customer terminal 1 using natural language processing with an artificial voice created based on the person's voice, an order extraction unit 26 that extracts the order content from the content of the conversation, and an order unit 27 that transmits order information to the kitchen terminal 4.

[0047] The functional components 21-30 of the order assisting server 2 will be described later in the description of the first and second embodiments.

[0048] 3 is a processing flow diagram of the order assistance server 2 and the human resource management server 3. Steps S1 to S3 are the processing of the human resource management server 3, and steps S4 to S9 are the processing of the order assistance server 2. For ease of explanation, they are summarized in one diagram.

[0049] First, the physical characteristics and voice of a real person are acquired and converted into data (Step S1). Based on this data, a digital human image is created (Step S2), and an artificial voice is created (Step S3), which is then stored and managed.

[0050] For simplicity, the image diagram in FIG. 1 shows only one digital human, but the human resource management server 3 manages digital humans for many people.

[0051] The order assistance server 2 receives the identification information from the customer terminal 1 and identifies the customer terminal 1 (step S4).

[0052] On the other hand, a digital human managed by the human resource management server 3 is selected and sent to the customer terminal 1 (step S5). The digital human is displayed on the customer terminal 1.

[0053] The customer terminal 1 and the order assistance server 2 can communicate with each other. The order assistance server 2 converses with the customer using natural language processing (step S6). The customer's comments are converted into text and input, and new sentences are output based on the input using an artificial intelligence algorithm. An artificial voice managed by the human resources management server 3 is used to output the sentences. The conversation is based on voice, but this does not exclude text input.

[0054] In addition, high-level commercial services are available in the fields of natural language processing and text generation AI, and external services may be used as appropriate.

[0055] In this application, the content of conversations with the digital human is not strictly limited, but it is important to train the digital human to avoid going off on a tangent and to lead to ordering food, etc.

[0056] The order assistance server 2 extracts the order details from the conversation (step S7) and sends the order information to the kitchen terminal 4 (step S8). The chef starts cooking according to the order information displayed on the kitchen terminal 4.

[0057] The order assistance server 2 may perform further accompanying processing based on the content of the conversation leading up to the order (step S9). Specific examples of accompanying processing will be described later in the explanations of the first and second embodiments.

[0058] Incidentally, in recent years, image generation AI technology has made remarkable advances, making it possible to create digital humans that do not exist at all. However, typical digital humans face issues related to the uncanny valley. Appetite is particularly important in restaurants and other food service establishments, where the uncanny valley, caused by natural human emotions, becomes particularly pronounced. For example, experiencing the uncanny valley can lead to a loss of appetite.

[0059] In this application, the digital human is based on a real person, which alleviates the problem of the uncanny valley, making it possible to use the digital human to assist customers in ordering at restaurants and other food service establishments.

[0060] - Overview of the first embodiment - It is empirically known that when a beautiful actress (not excluding men) introduces a product in a television advertisement, it attracts customers and leads to sales promotion. However, there are few positions available for television advertisements, and it is rare for an aspiring actress to be immediately cast in a television advertisement. Furthermore, it is almost impossible for an aspiring actress to be immediately cast in a television advertisement.

[0061] On the other hand, the physical characteristics and voice that form the basis of a digital human are typical personal information, and if they are released uncontrolled on the Internet, they could be misused, making information management difficult.

[0062] 4 and 5 are conceptual diagrams that supplement the understanding of the first embodiment.

[0063] To address the above issue, we assume that aspiring actresses (or actors) who are confident in their appearance, or aspiring actresses (or actors), register their data (steps S1-S3). The digital human administrator is a kind of talent bank, contracting with many aspiring actresses and managing a talent management server. Restaurant owners select candidates who fit the atmosphere of their establishments, much like hiring part-time workers. In the example, multiple choices are possible.

[0064] Digital humans who contribute to assisting with orders are paid according to their contributions, just like real part-time workers. Budding actresses and aspiring actresses can receive compensation while waiting for a big job opportunity. Also, having a large number of choices can serve as a basis for appealing to people in the entertainment industry that there is a high potential demand for them as actresses. While there are many benefits, there are no disadvantages, such as the time constraints of real part-time workers.

[0065] By limiting the use of digital humans to assist with ordering under a strict usage agreement, personal information can be managed reliably, and there is no risk of it being spread uncontrollably across the Internet and misused.

[0066] Although the application field of the digital human of the present invention is not limited to virtual waiters, the first embodiment uses virtual waiters as an example of a limited usage contract.

[0067] In recent years, awareness of human rights has increased, and people tend to avoid using appearance as a criterion for judgment. On the other hand, if customers can enjoy a conversation with a partner who is a good match, it is likely to lead to more orders.

[0068] In this example, multiple waiters can be selected. In reality, a virtual waiter is selected, so excessive concern about human rights is unnecessary.

[0069] The first embodiment is limited to ordering assistance by a virtual waiter, and does not specifically limit the actual waiter. A robot waiter may also be considered.

[0070] ~Details of the first embodiment~ 6 is a functional block diagram according to the first embodiment. The system of the present invention is the same as the basic embodiment in that it comprises a customer terminal 1, an order assistance server 2, a personnel management server 3, and a kitchen terminal 4.

[0071] As with the basic embodiment, the order assistance server 2 has a terminal identification unit 22, a display control unit 25, a conversation control unit 26, and an order unit 27. The order assistance server 2 in the first embodiment further has a waiter hiring unit 21, a waiter selection unit 24, and a remuneration calculation unit 28.

[0072] The human resource management server 3 includes a human resource management unit 31 and a remuneration management unit 32 .

[0073] Fig. 7 is a sequence diagram according to the first embodiment. Fig. 8 is a flow diagram showing detailed processing of the accompanying process (step S9 above). The functional configurations of the order assistance server 2 and the human resource management server 3 will be explained using the sequence diagram and flow diagram.

[0074] A plurality of digital humans are registered in the human resource management section 31 of the human resource management server 3 in association with real people.

[0075] A contract between a restaurant owner and a digital human administrator allows the restaurant owner to select a digital human. After the contract is concluded, the waiter recruitment unit 21 of the order assistance server 2 can access or download data related to the digital human.

[0076] The terminal identification unit 22 of the order assistance server 2 identifies the customer terminal 1 (step S4). The waiter selection unit 24 transmits waiter options to the customer terminal 1 and receives a waiter selection command from the customer terminal 1.

[0077] Based on a waiter selection command from the customer terminal 1, the display control unit 25 obtains data relating to the digital human corresponding to the selected waiter from the human resources management server 3, displays the digital human (step S5), and controls the digital human's facial expressions, etc.

[0078] The conversation control unit 26 acquires data relating to the artificial voice corresponding to the selected waiter from the personnel management server 3, and controls the conversation with the customer using the artificial voice (step S6).

[0079] The order unit 27 of the order assistance server 2 extracts the order details from the contents of the conversation (step S7), and transmits the order information to the kitchen terminal 4 (step S8).

[0080] The remuneration calculation unit 28 of the order assistance server 2 calculates the remuneration based on the order record. Various methods for calculating the order record are conceivable.

[0081] One possible method would be to provide a fixed amount of compensation each time a waiter is selected.

[0082] Furthermore, if conversation has the effect of encouraging customers to order, it is expected that the order amount will also increase. A method of evaluating the order record based on the order amount can be considered (step S11).

[0083] The present invention is characterized by extracting order details from conversations. If the conversation is too short, the conversation will not be enjoyable and the present invention will not be able to demonstrate its features. In other words, the more conversation there is between the start of an order and the order confirmation, the more enjoyable and lively the conversation will be. On the other hand, if the conversation continues for a long time, the order will not be decided and store turnover will slow down. In other words, it is preferable to have a short time until the order is confirmed.

[0084] Therefore, a method of evaluating the order record using an index based on the number of conversations from the start of an order until the order is confirmed and the time from the start of an order until the order is confirmed (step S12) is conceivable.

[0085] A method of evaluating the order performance may be considered in which these order performance evaluations are appropriately combined.

[0086] The reward is calculated so that the higher the order performance rating, the higher the reward (step S13).

[0087] In the conversation control of this invention, the AI ​​uses a learning model. Order records are used as a score, and the higher the score, the better the conversation. The learning model is then updated.

[0088] The remuneration calculation unit 28 of the order assistance server 2 transmits the remuneration information to the remuneration management unit 32 of the human resource management server 3 (step S14).

[0089] The remuneration management unit 32 of the human resource management server 3 accumulates the remuneration amount for a predetermined period (for example, weekly), and transfers the remuneration, excluding management costs, to the account of the target digital human.

[0090] Furthermore, the talent management unit 31 of the talent management server 3 manages the order performance scores of the digital humans as evaluations of the subjects themselves. For example, by presenting talent management information to people in the entertainment industry, they can use it as a reference for discovering new actresses or for casting movie roles.

[0091] ~Modification of the first embodiment~ A restaurant owner selects candidates who are suitable for the atmosphere of his or her establishment, much like hiring a part-time worker. In the first embodiment, multiple candidates can be selected, but a single candidate can also be selected. In this case, the waiter selection unit 24 is not required.

[0092] For ease of explanation, in the first embodiment, it is assumed that a digital human manager, who is a human resources bank, dispatches a digital human to one store, but digital humans can be duplicated, and the same digital human can be dispatched to multiple stores.

[0093] The more opportunities there are for use, the greater the potential demand for aspiring actresses and others will be able to earn more in a short period of time, as well as gain exposure similar to that of television advertisements, thereby increasing their name recognition.

[0094] That is, in the first embodiment, a single store employs multiple virtual waiters, and a customer selects one virtual waiter. When a customer makes a selection, healthy competition occurs between the virtual waiters.

[0095] In contrast, in the first embodiment variant, one virtual waiter is employed at multiple stores. Healthy competition occurs among the virtual waiters when they are hired. In any case, aspiring actresses and others with high potential demand can receive high compensation and many opportunities.

[0096] Of course, multiple virtual waiters may be employed at multiple stores.

[0097] - Overview of the second embodiment - As described in the first embodiment, if customers can enjoy conversation with someone who is a good match for them, it is highly likely that this will lead to increased orders. On the other hand, in recent years, awareness of human rights has increased, and people tend to avoid using appearance as a criterion for judgment. Some customers are hesitant to directly request a waiter. Also, even if they think they have a good match at first glance, it may be a mistake.

[0098] 9 and 10 are conceptual diagrams that help understand the second embodiment.

[0099] To address the above issues, the digital humans and artificial voices are created based on the staff members who can serve in the restaurant. When a restaurant owner enters into a labor contract with a staff member, they create a digital human and artificial voice corresponding to that staff member (steps S1-S3) and enter into a service agreement. The creation of the digital humans and artificial voices for each staff member can also be outsourced.

[0100] In the second embodiment, the actual staff may perform other duties besides serving, so the term "staff" is used. The term "staff" is used as a concept that includes serving.

[0101] The ordering assistance is performed through conversation by a virtual digital human, while the actual serving is performed by a real staff member. Furthermore, the compatibility between the customer and staff member is judged based on the content of the conversation, and if they are not compatible, a different staff member may be considered to serve the customer.

[0102] This prevents mismatches between customers and store clerks, and prevents both customers and store clerks from feeling uncomfortable, thereby ensuring that more orders are made.

[0103] ~Details of the second embodiment~ Figure 11 is a functional block diagram of the second embodiment. The system of the present invention is the same as the basic embodiment in that it comprises a customer terminal 1, an order assistance server 2, a personnel management server 3, and a kitchen terminal 4. The second embodiment further comprises waiter terminals 5 and 6.

[0104] As with the basic embodiment, the order assistance server 2 has a terminal identification unit 22, a display control unit 25, a conversation control unit 26, and an ordering unit 27. The order assistance server 2 in the second embodiment further has a staff shift management unit 23, a waiter selection unit 24, a compatibility determination unit 29, and a waiter preparation command unit 30.

[0105] The human resource management server 3 includes a human resource management unit 31 .

[0106] Fig. 12 is a sequence diagram according to the second embodiment. Fig. 13 is a flow diagram showing detailed processing of the accompanying process (step S9 above). The functional configurations of the order assistance server 2 and the human resource management server 3 will be explained using the sequence diagram and flow diagram.

[0107] A plurality of digital humans are registered in the personnel management section 31 of the personnel management server 3 in association with real store clerks.

[0108] In the second embodiment, the digital humans are store clerks in the same store, so the order assistance server 2 and the human resource management server 3 do not necessarily have to be separate servers.

[0109] The terminal identification unit 22 of the order assistance server 2 identifies the customer terminal 1 (step S4).

[0110] The staff shift management unit 23 of the order assistance server 2 checks the staff shift schedule for the day and lists the staff available to serve based on the time the customer terminal was identified. In other words, staff who are not in the store that day are excluded. Staff who are already serving or preparing to serve are also excluded. Waiter options are created from the listed staff.

[0111] The waiter selection unit 24 transmits waiter options to the customer terminal 1 and receives waiter selection commands from the customer terminal 1.

[0112] Based on the waiter selection command from the customer terminal 1, the display control unit 25 obtains data relating to the digital human corresponding to the selected virtual waiter from the personnel management server 3, displays the digital human (step S5), and controls the digital human's facial expressions, etc.

[0113] The conversation control unit 26 acquires data relating to the artificial voice corresponding to the selected virtual store clerk from the personnel management server 3, and controls the conversation with the customer using the artificial voice (step S6).

[0114] The order unit 27 of the order assistance server 2 extracts the order details from the contents of the conversation (step S7), and transmits the order information to the kitchen terminal 4 (step S8).

[0115] The compatibility determination unit 29 of the order assistance server 2 determines the compatibility between the customer and the store clerk based on the order history. Various methods can be considered for determining the order history.

[0116] A method is conceivable in which the fact that a waiter has been selected can be counted as one of the order records.

[0117] Furthermore, if conversation has the effect of encouraging customers to order, it is expected that the order amount will also increase. A method of evaluating the order record based on the order amount can be considered (step S21).

[0118] The present invention is characterized by extracting order details from conversations. If the conversation is too short, the conversation will not be enjoyable and the present invention will not be able to demonstrate its features. In other words, the more conversation there is between the start of an order and the order confirmation, the more enjoyable and lively the conversation will be. On the other hand, if the conversation continues for a long time, the order will not be decided and store turnover will slow down. In other words, it is preferable to have a short time until the order is confirmed.

[0119] Therefore, a method of evaluating the order record using an index based on the number of conversations from the start of an order until the order is confirmed and the time from the start of an order until the order is confirmed (step S22) is conceivable.

[0120] A method of evaluating the order performance may be considered in which these order performance evaluations are appropriately combined.

[0121] The higher the order record rating, the better the compatibility is determined. For example, a threshold is set, and if the rating is equal to or higher than the threshold, the compatibility is determined to be good. If the rating is less than the threshold, the compatibility is determined to be bad (step S23).

[0122] If the compatibility is good, the waiter preparation command unit 30 commands the waiter terminal 5 of the waiter corresponding to the selected virtual waiter to prepare for serving (step S24). The waiter corresponding to the selected virtual waiter will then actually serve the customer.

[0123] Since the customer has experienced a conversation with the selected virtual store clerk, the customer can receive service from a real store clerk without feeling uncomfortable.

[0124] If the compatibility is poor, the waiter preparation command unit 30 considers whether to issue a waiter preparation command to the waiter terminal 5 of the waiter corresponding to the selected virtual waiter. For example, it sends a compatibility review notice to the customer terminal 1, receives the review result from the customer terminal 1, and judges the review result (step S25). The compatibility review notice states whether the customer was satisfied with the conversation with the virtual waiter.

[0125] If the result of the review is favorable, the waiter preparation command unit 30 commands the waiter terminal 5 of the waiter corresponding to the selected virtual waiter to prepare for serving (step S24).

[0126] Even if it is determined that the two are not a good match, if the results of the review are positive, the customer can receive service from a real store clerk without feeling uncomfortable.

[0127] If the result of the review is unfavorable, the waiter preparation command unit 30 commands the waiter terminals 6 of the waiter corresponding to the selected virtual waiter and another waiter to prepare for serving (step S26). The other waiter will then actually serve the customer.

[0128] Among the staff members listed by the staff shift management unit 23 of the order assistance server 2, the staff member with the highest order record is selected as the other staff member.

[0129] If a mismatch between a customer and a store clerk is deemed to be incompatible and the consideration results are negative, this can prevent mismatches between customers and store clerks in advance.

[0130] ~Modification of the second embodiment~ In the second embodiment, a compatibility judgment is made (step S23), but instead of making a compatibility judgment, the serving preparation command unit 30 may command the serving terminal 5 of the waiter corresponding to the selected virtual waiter to prepare for serving (step S24) after the order is completed. In this case, the compatibility judgment unit 29 is not required.

[0131] In the second embodiment, if the compatibility judgment (step S23) is poor, a compatibility review (step S25) is carried out, but it is also possible to not review the compatibility and to instruct the serving terminals 6 of the staff member corresponding to the selected virtual staff member and another staff member to prepare for serving (step S26).

[0132] As in the first embodiment, the order record may be used as a score, and the higher the score, the better the conversation, so that the learning model may be re-learned and updated.

[0133] In the first embodiment, the human resource management server 3 manages the order performance score as an evaluation of the subject himself / herself, but in the second embodiment, the order performance score may also be managed as an evaluation of the store clerk. As in the first embodiment, the human resource management information may be used for other purposes within the scope of the contract.

[0134] The remuneration calculation of the first embodiment may be applied to the remuneration calculation of the store clerk in the second embodiment. That is, the second embodiment has a remuneration calculation unit 28. [Explanation of symbols]

[0135] 1. Customer terminal 2 Order assistance server 3. Human Resources Management Server 4 Kitchen terminal 5 serving terminals 6 Waiter terminal (separate staff) 21 Waiter Recruitment Department 22 Terminal Identification Unit 23 Staff Shift Management Department 24 Waiter Selection Department 25 Display control unit 26 Conversation control unit 27 Order Department 28 Compensation Calculation Department 29 Compatibility Judgment Section 30 Serving Preparation Command 31 Human Resources Management Department 32 Compensation Management Department 100 Information processing device 101 CPU 102 ROM 103 RAM 104 Non-volatile memory 105 Input / Output Interface Bus 106 107 Communication Unit 108 input units 109 Display Unit

Claims

1. A device that assists ordering in a restaurant, Identification means for identifying a customer terminal; a digital human display means for displaying and operating a digital human image created based on a real person on the customer terminal; a conversation means for conducting a conversation through the customer terminal by natural language processing using an artificial voice created based on the voice of the person; an order extraction means for extracting order details from the content of the conversation; An order assistance device comprising:

2. a waiter hiring means capable of communicating with a person registration server that stores digital human images and artificial voices relating to a plurality of real people, and that hires one or more virtual waiters; a reward calculation means for calculating a reward based on the order record of the employed virtual waiter; 2. The ordering assistance device according to claim 1, further comprising:

3. a waiter hiring means capable of communicating with a person registration server that stores digital human images and artificial voices relating to a plurality of real people, and that hires two or more virtual waiters; a waiter selection means for selecting one of the employed virtual waiters; a reward calculation means for calculating a reward based on the order record of the selected virtual waiter; 2. The ordering assistance device according to claim 1, further comprising:

4. The remuneration calculation means evaluates the order performance based on the number of conversations from the start of the order to the order confirmation and the time from the start of the order to the order confirmation.

4. The order assisting device according to claim 2 or 3.

5. The remuneration calculation means evaluates the order performance based on the order amount.

4. The order assisting device according to claim 2, further comprising:

6. The real person is a waiter available in the store, a staff member enumeration means for enumerating staff members currently available to serve as virtual staff members; a salesperson selection means for selecting one of the enumerated virtual salespersons; a waiter preparation command means for commanding the waiter corresponding to the selected virtual waiter to prepare for serving; 2. The ordering assistance device according to claim 1, further comprising:

7. a compatibility determination means for determining compatibility between the selected virtual store clerk and the customer based on the order history of the selected virtual store clerk; Equipped with The waiter preparation command means, when the compatibility is poor, considers whether to command the waiter corresponding to the selected virtual waiter to prepare for serving.

7. The order assisting device according to claim 6.

8. The compatibility determination means evaluates the order record based on the number of conversations from the start of the order to the order confirmation and the time from the start of the order to the order confirmation.

8. The order assisting device according to claim 7, further comprising:

9. The compatibility determination means evaluates the order record based on the order amount.

8. The order assisting device according to claim 7, further comprising:

10. A method for assisting ordering at a restaurant using an information processing device, comprising: Identify the customer device, Displaying and operating a digital human image based on a real person on the customer terminal; a conversation is conducted through the customer terminal using natural language processing with an artificial voice created based on the voice of the person; Extract order details from the conversation An order assistance method comprising:

11. To assist ordering in restaurants, an information processing device is A process for identifying a customer terminal; displaying and operating a digital human image based on a real person on the customer terminal; a process of having a conversation through the customer terminal using an artificial voice created based on the person's voice through natural language processing; A process of extracting order details from the content of the conversation; An order assistance program characterized by executing the above.

Images