JP2018116427A - Electronic order-taking system and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a user-friendly electronic order-taking system.SOLUTION: In an electronic order-taking system, when a user terminal logs on a matching site via a network (16), multiple agent images are displayed as a list on an agent displaying unit 32 of a screen 30 of the user terminal. When a user drags any of the agent images to a drag region 38 of an actuating unit 34, a selected agent is enlarged in size, and is displayed so as to face an avatar 36 that is the agent of the user. The agent is caused to utter in accordance with a scenario, and selection choice buttons 42a and 42b in accordance with such an utterance are displayed in accordance with the scenario. The user touches any of the selection choice buttons to keep having a conversation with the agent. The user can freely select the agent from the agent displaying unit.SELECTED DRAWING: Figure 3

Description

  The present invention relates to an electronic listening system and program, and more particularly to an electronic listening system and program that interacts through, for example, a touch display.

  For example, an electronic commerce type listening support device as disclosed in Patent Document 1 is known. The system of Patent Literature 1 includes a management terminal, a user terminal, and a client terminal, and displays a client avatar set for each client prepared by the management terminal as a listening avatar on the user terminal. The user can instruct purchase of a product on the user terminal based on advertisement information presented by a client avatar who is a seller.

JP 2011-215704 A [G06Q 30/00, G06F 13/00]

  In the system of Patent Document 1, since the request information cannot be transmitted unless the conditions are satisfied, the user cannot call any seller (client) and place an order when necessary. For example, Usability is slightly worse for the elderly and those who are weak.

  Therefore, a main object of the present invention is to provide a novel electronic listening system and program.

  Another object of the present invention is to provide a user-friendly electronic listening system and program.

  The present invention employs the following configuration in order to solve the above problems. The reference numerals in parentheses, supplementary explanations, and the like indicate the corresponding relationship with the embodiments described in order to help understanding of the present invention, and do not limit the present invention.

  The first invention is an electronic listening system using a user terminal, which includes an agent list display unit that displays a list of at least two agent symbols indicating at least two agents on the screen of the user terminal, and an agent list display unit. A selection discriminating unit for discriminating that the user has selected the agent symbol, an agent display unit for displaying the agent corresponding to the agent symbol selected by the user in a state facing the avatar representing the user, and setting the agent in a predetermined scenario. The electronic listening system includes an utterance unit that utters along an option, and an option display unit that displays options according to a predetermined scenario according to the utterance of an agent.

  In the first invention, the electronic listening system (10: reference numerals exemplifying corresponding parts in the embodiment, hereinafter the same) uses the user terminal (14). The agent list display unit (S3) displays a list of at least two agent symbols (small images of agents, names, nicknames, etc.) indicating at least two agents on the screen (30) of the user terminal. The selection determining unit (S5) determines that the user has selected any agent symbol from the agent list display unit. When the selection determining unit determines that the user has selected any agent symbol from the agent list display unit, the agent display unit (S5) displays an agent corresponding to the agent symbol selected by the user as an avatar representing the user. Display in a face-to-face state. The utterance unit (S7-S9) causes the agent to utter along a predetermined scenario, and the option display unit (S11) displays options according to the predetermined scenario according to the utterance of the agent. Therefore, the user can select any of the options as appropriate and interact with the agent.

  According to the first invention, since the user can arbitrarily select an agent that the user wants to interact with, it is convenient for the user.

  A second invention is dependent on the first invention, and the screen of the user terminal includes an agent display unit that displays an agent symbol and an operation unit that displays an agent. An electronic listening system that determines whether a symbol has been dragged to an actuator.

  In the second invention, the screen (30) of the user terminal includes an agent display unit (32) for displaying an agent symbol and an operation unit (34) for displaying the agent. The selection determining unit (S5) determines whether any agent symbol has been dragged from the agent display unit to the operating unit.

  According to the second aspect, since the agent can be selected by dragging the agent symbol from the agent display unit to the operating unit, it is easy for the user to select the agent.

  The third invention is dependent on the second invention, and an agent change determining unit that determines that another agent symbol is dragged from the agent display unit to the operating unit and the agent is changed to another agent, and the agent change The electronic hearing system further includes a scenario changing unit that changes a scenario so that another agent speaks when the determination is made to change the agent.

  In the third invention, the agent change determination unit (S62) determines whether or not another agent symbol is changed from the agent display unit (32) to another agent depending on whether or not another agent symbol is dragged to the operation unit (34). When the agent change determination unit determines the agent change, the scenario change unit (S63) changes the scenario so that another agent speaks.

  According to the third aspect, since the user can arbitrarily change the agent, the usability is further improved.

  A fourth invention is an electronic listening system according to any one of the first to third inventions, further comprising an additional agent display unit for displaying an additional agent together with the agent.

  In the fourth invention, the additional agent display unit (S22, S34, S44, S54) displays the additional agent together with the agent.

  According to the fourth aspect, the agent and the additional agent can cooperate with each other.

  A fifth invention is an electronic listening program executed by a computer (20, 26) of an electronic listening system (10) using a user terminal, wherein the computer is provided on at least two screens (30) of the user terminal. An agent list display unit (S3) that displays a list of at least two agent symbols indicating agents, a selection determination unit (S5) that determines that the user has selected any agent symbol from the agent list display unit, and a user selected An agent display unit (S5) for displaying an agent corresponding to the agent symbol in a state facing the avatar representing the user, an utterance unit (S7-S9) for causing the agent to utter along a predetermined scenario, and depending on the utterance of the agent To display options according to a given scenario択肢 display unit to function as a (S11), an electronic order taker cyber program.

  According to the fifth invention, the same effect as that of the first invention can be expected.

  According to this invention, since the user can arbitrarily select an agent that the user wants to interact with, an electronic listening system that is convenient for the user can be obtained.

  The above object, other objects, features, and advantages of the present invention will become more apparent from the following detailed description of embodiments with reference to the drawings.

FIG. 1 is a schematic diagram showing an outline of an electronic listening system according to an embodiment of the present invention. FIG. 2 is an illustrative view showing one example of an initial screen of the user terminal in FIG. 1 embodiment. FIG. 3 is an illustrative view showing one example of a subsequent screen of the user terminal in FIG. 1 embodiment. FIG. 4 is a flowchart showing an example of the operation of the embodiment in FIG. FIG. 5 is an illustrative view showing one example of an order form that may be used in the embodiment of FIG. FIG. 6 is a flowchart showing another example of the operation of FIG. 1 embodiment. FIG. 7 is an illustrative view showing one example of screen transition of the user terminal according to the operation shown in FIG. FIG. 8 is a flowchart showing another example of the operation of the FIG. 1 embodiment. FIG. 9 is a flowchart showing another example of the operation of the FIG. 1 embodiment. FIG. 10 is an illustrative view showing one example of screen transition of the user terminal according to the operation shown in FIG. FIG. 11 is an illustrative view showing one example of a screen transition of the user terminal according to the operation shown in FIG. 9, and shows an example of a screen subsequent to FIG. FIG. 12 is a flowchart showing another example of the operation of the FIG. 1 embodiment. FIG. 13 is a flowchart showing another example of the operation of the FIG. 1 embodiment. FIG. 14 is an illustrative view showing one example of screen transition of the user terminal according to the operation shown in FIG.

  With reference to FIG. 1, an electronic listening system (hereinafter simply referred to as “system”) 10 of this embodiment includes a matching site 12 and one or more user terminals 14 connected via a network 16. The user terminal 14 is a touch screen display terminal such as a tablet terminal, and the user can basically interact by a touch operation. However, as is well known, the location on the touch screen that the user touches is determined based on the touch coordinates indicating the touch position.

  The system 10 may further be connected with one or more vendor terminals 18 via the network 16. Arbitrary computers can be used as the trader terminal 18, and a mobile terminal such as a smartphone can also be used. However, any traders such as merchandise dealers and service providers can be considered as traders.

  The matching site 12 basically responds to the user's request while repeating the dialogue with the user of the user terminal 14, and further matches the user of the user terminal 14 and the supplier of the supplier terminal 18 when necessary. is there.

  The matching site 12 includes a chat server 20 that serves as a window to the network 16 of the user terminal 14. The chat server 20 is a server for constructing a chat mechanism, and accepts access from all user terminals 14. The utterance from the user terminal 14 is temporarily stored in the dialog queue 22 each time. However, all chat logs are stored in the chat database 24.

  A dialog server 26 that interacts with the user terminal 14 and the like in cooperation with the chat server 20 is provided. The dialog server 26 mainly instructs the chat server 20 for an utterance sentence and the like. For this purpose, a scenario database 28 is provided with the dialog server 26. The scenario is created by another system and set in advance in this scenario database 28.

  The dialog server 26 acquires user utterance data (contents of option buttons) from the dialog queue 22, retrieves a scenario corresponding to the data from the scenario database 28, and sends the scenario to the chat server 20. At that time, the dialog server 26 transmits to the chat server 20 the content of the utterance (sentence), the number of the utterance, the number of the agent (described later), the action of the agent, and the content and the number of the option button described later.

  As will be described later, the matching site 12 interacts with the user terminal 14 (the user) through cooperation of the chat server 20 and the dialog server 26, and introduces a supplier that provides a product or service desired by the user to the user. Fulfills the function.

  FIG. 2 is an illustrative view showing an example of an initial screen of the screen 30 of the user terminal 14 immediately after logging in to the matching site 12. The configuration of the screen 30 is a vertically oriented layout so that an elderly person can operate while holding a tablet with one hand.

  The screen 30 of the user terminal 14 includes an upper agent display unit 32 and an operation unit 34 below it. The agent display unit 32 displays a list of two or more agents that can interact with the user, in this embodiment, four agents A, B, C, and D. In the embodiment, Agent A is an agent representing a Yorozuya with the name “Mikawaya”, Agent B is an agent representing a staff member of a municipal office where the user lives, and Agent C is an agent for this interview. For example, it is an agent representing a salesclerk of “Michi no Eki” who is an operating entity of the system 10, and an agent D is an agent representing a taxi driver, for example. These agents have a one-to-one correspondence with the support content.

  However, it should be pointed out in advance that the agent AD and the specific job type here are merely examples, and that the symbol AD is irrelevant to the agent symbol that will be described later together with the flowchart. . Furthermore, the number of agents may be two or more.

  An avatar 36 serving as a user agent is disposed in front of the screen in the operation unit 34, and a drag region 38 is formed in a direction in which the avatar 36 faces. The drag area 38 is set as an area for dragging any of the agents AD that the user wants to talk about and wants to ask. If the user drags one of the agents to the drag area 38 If it is dragged, the screen 30 of the user terminal 14 changes as shown in FIG.

  Here, the user's avatar 36 is displayed backward on the screen. Thereby, the user can easily be sure that the avatar 36 is his / her agent. That is, by displaying the avatar 36 backward, the user can strongly recognize that he is facing the agent displayed in an enlarged manner as shown in FIG.

  In FIG. 3, as an example, a screen when the agent C who is a clerk is dragged to the drag region 38 is shown. In this screen, the agent C of the clerk is enlarged to face the agent 36 across the drag region 38. Is displayed. That is, the agent called by the user is enlarged and displayed so as to face the user avatar 36 in the screen of the operation unit 34. On the other hand, the image of agent C in the agent display unit 32 is deleted, and it can be seen that the agent C is called by the operating unit 34. At the same time, a balloon 40 from the agent C is displayed, and an utterance sentence such as a question to the user is displayed therein. In the example, it is shown that the user speaks “What is your business?”

  In contrast, a plurality of (two in the embodiment of FIG. 3) option buttons 42a and 42b are displayed below the user avatar 36. In this example, the button 42a indicates a user's request “I want to ask for shopping”, and the button 42b indicates a request “I want to talk”. Therefore, by touching any button, the user can inform the store clerk, that is, the agent C, of the item desired by the user at that time. However, the number of option buttons 42 is determined to be a maximum of four. This is because when the amount is too large, the selection becomes unclear, and when the amount is too small, there is no choice.

  In this way, when you drop the agent you want to talk to from the agent list in the center of the screen, that agent appears and the corresponding conversation scenario starts. However, the agent you want to interact with may be in the middle of a conversation with another agent. , Can be switched at any timing. At that time, naturally, the scenario set for the utterance is also switched. However, even if the user changes the agent, the utterance (dialogue) may be continued in the same scenario set as before the change.

  However, unlike this embodiment, it may be possible to select an agent by simply touching a desired agent among the agents AD displayed on the agent display unit 32. However, this method increases the possibility of erroneous selection due to erroneous touch. In contrast, in the method of selecting an agent by dragging a desired agent image to the drag region 38 as in the embodiment, the agent image is touched on, dragged while being touched on, and then touched off in the region 38. Therefore, not only can the possibility of erroneous selection be reduced as much as possible, but also the user can be fully aware of his / her selection. For example, the wrong selection can be stopped at an intermediate stage.

  3 is read from the scenario database 28 by the dialog server 26 shown in FIG. 1 and the user terminal 14 is read from the chat server 20 by the dialog server 26 shown in FIG. It is displayed as a result of sending to.

  For this reason, the scenario database 28 of the matching site 12 stores in advance a scenario set (a series of scenarios) for each of the agents AD, and further includes a scenario corresponding to each option button 42 therein. Yes. Therefore, the agent 36, that is, the user selects any option button, an appropriate response is made from the matching site 12.

  Interview agent dialogues can be broadly divided into two types: task dialogues as a window for requesting support and non-task dialogues that entertain users such as elderly people as communication partners. And in order to use these appropriately, it is necessary to confirm a user's request first. Also, in order to promote the use of potential services, scenarios that naturally lead to non-task dialogues to task dialogues are also considered useful.

  Therefore, in the embodiment, the dialogue scenarios stored in the scenario database 28 in FIG. 1 are classified into four categories, “start scenario”, “task scenario”, “non-task scenario”, and “guide scenario”, and are used properly according to the purpose. It was decided to do.

  The start scenario is a scenario that is executed at the start of a conversation. A simple greeting and confirmation of a message are performed, and a scenario to be branched is determined. For example, option buttons “I want to ask for shopping” and “I want to talk” are presented, and the task branches to a task scenario for requesting support, a non-task scenario or a guidance scenario for chatting, depending on the option.

  The task scenario is a scenario for requesting support, confirming the minimum requirements necessary for the request, and notifying the supporter. For example, in the case of shopping support, option buttons such as “groceries”, “daily necessities”, “medical supplies”, and “others” are presented, and a supporter who can respond according to the selected content is notified. .

  In the regional comprehensive care system assumed by the listening system 10 of the embodiment, the supporter is a volunteer who lives in the same area as the user, and does not always meet the user's request. For this reason, in the dialogue with the listening agent, only the information necessary for assigning the person in charge and determining the priority is heard, and the details of the request and the availability of the response are confirmed again by the supporter and determined by the user. For example, the support desired by the user is determined by calling the user directly from the supporter (trader).

  The non-task scenario is a scenario for establishing a trust relationship between the user and the interview agent, and performs a chat that is not related to the support service. For example, the dialogue that conveys the person of the listening agent is conducted, for example, the dialogue that worries about the physical condition of the user, or the user asks the complaint of the listening agent.

  The guidance scenario is a scenario for encouraging the use of the support service. The user's trouble is heard through chat, and if necessary, the use of the support service is guided. For example, a question “Do you have a problem with shopping?” Is returned, and if the user is in trouble, a response “You can buy it instead” is returned. If the user desires to use the support service on the spot, the task scenario is transferred to notify the supporter.

  In the selection-type dialogue using the option button 42 adopted by the system of the embodiment, the user selects one close to his / her idea from the presented options. Therefore, it is considered that the user's utterance can be induced by appropriately designing the option presentation method and the agent's utterance. A specific method for guiding the user's utterance is proposed below.

  It is known that one side can disclose deep information about the self and the other side can disclose the same level of deep information (returnability of self-disclosure). Talking about privacy has a sense of resistance for the user, but it is thought that the listening agent can talk about the user's troubles without resistance by talking about his concerns first.

  By presenting a plurality of synonymous choices, even if there is only one scenario development, it is possible to give the user a sense of decision making. For example, if you want to get an approval from the supporter, please say “Thank you” and “Let me know” for the utterance “Can I call you later?”

  If none of the utterances presented with ambiguous choices matches the user's idea, the user hesitates to make a choice. Therefore, by providing an opportunity to avoid the selection once, it is possible to give a margin for reconsidering the response within the range of the presented options. For example, when you want to ask if you are having trouble with shopping, you can also present utterances that do not select either “I'm in trouble” or “I'm not in trouble”, such as “Hmm…”. If an utterance that does not select either is selected, an utterance such as “I am really in trouble” or “I am not in trouble” is presented and a question is asked again.

  Regarding the scenario, the following devices that further increase the feeling of dialogue may be considered.

  In order to deepen the relationship between the user and the listening agent, it is important to give the impression that the user is actually interacting with the agent. A method for this is proposed below.

  I think that an agent can give an impression of speaking to the user himself by calling the user's name. The user registers his name at the start of use, and the agent embeds the name in the utterance and calls the user.

  I think that it is possible to deepen the familiarity between the user and the agent who listens by expressing favors to the other person expressing emotions and expressing their own feelings. For example, an agent utters an impression on a user such as “Because xxx looks fine” or “I am happy to talk with xxx”. Also, by presenting utterances such as “I am happy” as a user's choice, I think that it is possible to clearly make people feel ambiguous feelings.

  Fillers such as “Ano” and “Ut” that express the mental state indicate a mental state such as choosing a word for utterance or thinking about it. Let it be predicted. By adding a filler to the utterance of the agent, we can give the impression that the agent is interacting with the user in mind.

  It is thought that the user feels friendly and sympathetic to the agent when the agent is clogged up in words or mistaken. For example, when telling that you want to make a call, choose a word and say it again, such as "Telephone ... I'll call you."

  Here, with reference to FIG. 4, the operation in the system 10, that is, the matching site 12, will be described. However, in the matching site 12, as already described, the chat server 22 and the dialog server 26 cooperate to advance a dialog with the user (user terminal 14).

  The first step S1 is started when there is an access (login) from the user terminal 14 to the chat server 20. As is well known, user authentication is executed by inputting a login password or the like from a user terminal. However, if the user authentication is not passed, the process cannot proceed to the next step S3.

  In step S3 after passing the user authentication, the chat server 20 displays an agent list capable of listing a plurality of agents on the agent display unit 32 (FIG. 2) of the user terminal 14. At this time, the name of the agent and the scenario ID of the agent are stored in the chat database 24 for each agent. Step S3 functions as an agent list display unit that displays a list of agents indicating at least two agents on the screen of the user terminal. The agent list display unit displays each agent image in the embodiment, but it may be possible to display only the characters such as names and nicknames. Such agent images, names, nicknames, and the like can be referred to as agent symbols that can identify each agent individually.

  At this time, the user avatar 36 is also displayed on the operating unit 34, but the image data is transmitted from the chat server 20 to the user terminal 14 from the chat server 20 after the user authentication. To display. However, since it is useless to send an image from the chat server 20 to the user terminal 14 every time the user logs in, the image received once is stored in a memory (not shown) in the user terminal 14 and is displayed along with the successful login. May be displayed on the operating unit 34.

  In the next step S5, whether any one agent has been selected from the agent display unit 32 and dragged to the drag area 38 (FIG. 2) by the method described above, that is, on the user terminal 14. To determine whether an agent is selected. Since the signal indicating the agent selected by the user terminal 14 is input from the user terminal 1 to the chat server 20, the chat server 20 can know which agent has been selected on the user terminal 14. Step S5 functions as a selection determination unit that determines that the user has selected any of the agents displayed on the agent list display unit.

  The chat server 20 writes to the dialog queue 22 which agent has been selected by the user terminal 14, thereby informing the dialog server 26 which agent has been selected by the user, and requests to start a scenario for the agent. .

  In step S <b> 7, the dialog server 26 reads the scenario set of the corresponding scenario ID from the scenario database 28 and transmits the first utterance sentence to the dialog queue 22. At this time, as described above, the dialog server 26 sends chat contents (words), utterance numbers, agent numbers, and selection button contents and numbers to the chat chat server 20 as necessary. Send etc.

  When “YES” is determined in the step S5, for example, as shown in FIG. 3, the agent selected by the user is displayed so as to face the avatar 36 representing the user. Therefore, step S5 also functions as an agent display unit.

  Thereafter, in step S9 and subsequent steps, the agent called by the operating unit 34 (FIG. 2) outputs a conversation according to the scenario. When necessary, in step S11, for example, an option button 42 as shown in FIG. Is displayed. When the conversation is output, not only the wording is displayed in the balloon 40 as shown in FIG. 3, but also the text data is used to output as a voice from the speaker (not shown) of the user terminal 14. Thereby, the user can grasp the contents of the utterance of the agent more surely. Here, step S9 and the like function as an utterance unit that utters the enlarged agent according to a predetermined scenario. Furthermore, since step S11 displays the option button 42 which the user avatar 36 should select, it functions as an option display part.

  When any option button is selected in step S13, the information is transmitted to the chat server 20, and the chat server 20 writes the information in the dialog queue 22. However, in step S13, as described above, it is determined whether or not the touch coordinates on the touch screen of the user terminal 14 are within the display range of the corresponding option button.

  When the user selects any option button in step S13, the dialog server 26 extracts a scenario according to the content of the option button selected by the user from the scenario database 28 and writes it in the dialog queue 22. Therefore, the chat server 20 causes the agent to speak based on the scenario (step S15).

  However, when any option button is selected in step S13, not only the display of the option button is changed to indicate that the option button has been selected, but also the user terminal according to the wording in the option button. The sound is output from 14 speakers. The user can easily understand which option button the user has selected.

  In subsequent step S17, the chat server 20 determines whether or not the scenario set has been completed by looking at the dialog queue 22. If "NO", the process returns to the previous step S11, and the steps after step S11 are repeated. If “YES”, the process ends.

  When “NO” is determined in the previous step S13, the chat server 20 determines whether or not the agent repeats the utterance of the same sentence in step S19. For example, when the agent outputs the same utterance three times or more, “YES” is determined in this step S19, and in this case, the process is terminated. If “NO”, the process returns to step S13.

  When the selection of the option button 42 in step S13 or the like is an option indicating that something is ordered from the supplier, for example, an instruction is issued from the matching site 12 to the supplier terminal 18 shown in FIG. However, for example, an order form 50 as shown in FIG. 5 may be transmitted to the user terminal 14. As an example, this order form is an order form for ordering a lunch box, and, for example, an ID entry field 52 for entering a user ID and an order number entry field 54 for entering the number of orders are provided. However, the ID entry field 52 may be in an already filled state when transmitted from the trader terminal 18 to the user terminal 14. Further, in this ID entry field 52, not only the user ID but also the user's name may be written. Then, it can be easily confirmed that the user is the order form for himself.

  The typical operation of the embodiment of FIG. 1 has been described above with reference to FIG. 4, but various operation examples will be described below together with the screen transition of the user terminal 14.

  In the operation example illustrated in FIG. 6, an additional or auxiliary agent B different from the main agent A selected by the user appears in the operation unit 34 of the screen 30 of the user terminal 14.

  When agent A is selected in step S20 as in steps S3 and S5 of FIG. 4, in step S21, chat server 20 and dialog server 26 have a conversation from the data of scenario AA for agent A to agent A. Output. Thereafter, in step S22, the chat server 20 and the dialog server 26 cause the agent B to appear in the operating unit 34 regardless of the user's selection. That is, in the state where the agent A is displayed together with the user avatar 36 on the operation unit 34 as shown in FIG. 7A, the agent B is close to the agent A as shown in FIG. , The size is smaller than that of Agent A. At this time, the chat server 20 and the dialog server 26 cause the agent B to output a conversation from the data of the scenario AA for the agent A because the user is interacting with the agent A. Step S22 functions as an additional agent display unit for displaying an additional or auxiliary agent together with the main agent.

  As shown in FIG. 6, in addition to agent A, agent B is displayed on the same operation unit 34, and speaking is performed based on the same scenario set, for example, agent A and agent B can talk directly. By having two agents talk directly, the user can indirectly recognize the advantages of a specific product or service through the conversation, or persuade the user by two people. Furthermore, it is possible to produce a scene in which two agents collaborate to consult the user.

  FIG. 8 shows another example of a method for causing another agent to appear on the operating unit 34 as shown in FIG. Steps S30 and S31 in FIG. 8 are the same as steps S20 and S21 in FIG.

  In the next step S32, the agent A displays a selection button 42 as shown in FIG. 3 based on the data of the scenario AA. In step S33, it is determined whether there is an option button for causing agent B to appear in the option buttons and whether the user has selected the option button. The determination in step S33 is, for example, the same determination as in step S13 in FIG.

  If “YES” is determined in step S33, the chat server 20 and the dialog server 26 add the agent A to the operating unit 34 of the user terminal 14 in step S34, as shown in FIG. The agent B that the user has accepted is displayed. Then, the agent B outputs a subsequent conversation according to the data of the scenario AA for the agent A. Also in this case, since the agent B speaks according to the data set in the scenario AA for the agent A as in the previous embodiment of FIG. 6, the agent B reinforces the utterance of the agent A, You can help. Here, step S34 functions as an additional agent display unit for displaying an additional or auxiliary agent together with the main agent.

  In the operation examples of FIGS. 6 to 8, at least two agents are displayed on the operation unit 34 at the same time, so that collaboration (cooperation) of these agents can be expected.

  On the other hand, in the operation example shown in FIGS. 9 to 11, at least two agents are simultaneously displayed on the actuating unit 34, but then the auxiliary displayed agents are exited from the actuating unit 34. As a result, the dialogue between the main agent and the user is continued and a conclusion is drawn.

  Steps S40 to S43 in FIG. 9 are the same as steps S20 to S23 in FIG. 6, and at the time of step S43, as in FIGS. 7A and 7B, FIG. 10A and FIG. As shown in FIG. 10 (B), in addition to the main agent A, an additional agent B appears in the operating unit 34. Here, step S43 functions as an additional agent display unit for displaying an additional or auxiliary agent together with the main agent.

  Thereafter, similar to steps S32 and S33 of FIG. 8, in the next step S44, the agent A displays the option button 42 (FIG. 3) based on the data of the scenario AA. It is determined in step S45 whether or not there is an option button for leaving Agent B and the user has selected the option button. The determination in step S45 is, for example, the same determination as in step S13 of FIG.

  When “YES” is determined in step S45, in step S46, the chat server 20 and the dialog server 26 leave (erase) Agent B from the operating unit 34 of the screen 30 of the user terminal 14. The state is shown in FIG. That is, agent B is displayed in FIG. 11B, whereas agent B is displayed in FIG. Therefore, step S46 functions as an additional agent leaving section for leaving additional or auxiliary agents.

  Thereafter, in step S47 and thereafter, agent A continues the dialog with the user based on scenario AA. Similarly, when “NO” is determined in the step S45, the agent A continues the dialogue with the user based on the scenario AA in a step S48.

  In this operation example, agent B appears along with agent A on the way, and only agent A remains thereafter.

  On the other hand, the operation example shown in FIG. 12 is an operation example in which agent B serves as the main agent in place of agent A in the middle.

  That is, Steps S50 to S53 in FIG. 12 are the same as Steps S40 to S43 in FIG. 9. As a result, in the stage of Step S53, as shown in FIG. Agent B is displayed at the same time. Here, step S53 functions as an additional agent display unit for displaying an additional or auxiliary agent together with the main agent.

  Thereafter, the agent B displays an option button for leaving the agent A (step S54), and determines whether the user has selected the option button (step S55). If “YES” is determined in the step S55, the agent A is left (not displayed) in a step S56, contrary to the step S46 of FIG.

  In step S57, the chat server 20 and the dialog server 26 display the agent B at the position where the agent A of the operating unit 34 was displayed in FIG. That is, the main agent is changed from A to B. Accordingly, in the subsequent step S58, the agent B continues the utterance (conversation) along the scenario AA for the agent A. Therefore, step S58 functions as an agent changing unit for changing the main agent.

  However, if “NO” is determined in the step S55, the agent A continues the dialog with the user based on the scenario AA in a step S59.

  The operation examples shown in FIGS. 13 and 14 are operation examples when the agent is changed at the user's will.

  Steps S60 and S61 in FIG. 13 are the same as steps S20 and S21 in FIG. 6, for example. That is, for example, as shown in FIG. 14A, the agent A appears in the operation unit 34 of the user terminal 14, and the agent A and the user (user avatar 36) interact with each other.

  In step S62, the chat server 20 and the dialog server 26 cause the user to drop another agent, for example, agent B, displayed on the agent display unit 32 (FIG. 3) onto the drag area 38 on the user terminal 14. Judge whether or not. That is, it is determined whether the user has instructed to change the agent.

  If “YES” in the step S62, only the agent B is displayed on the operation unit 34 as shown in FIG. That is, the agent is changed from A to B. Therefore, after step S63, agent B outputs a conversation according to scenario BB for agent B.

  Specifically, in step S62, the chat server 20 can detect that the agent displayed on the operating unit 34 has been changed from a series of user touch operations. When such a change of the agent is written to the dialog queue 22 by the chat server 20, the dialog server 26 reads the scenario set BB for the changed agent B from the scenario database 28 in step S63, and the dialog queue 22 Write to. That is, the chat server 20 in step S62 functions as an agent change determination unit, and the dialog server 26 in step S63 functions as a scenario change unit that changes a scenario in response to such an agent change.

  If “NO” in the step S62, the agent is not changed, and the agent A continues the conversation according to the scenario AA for the agent A in the step S64 and thereafter.

  According to the operation example, the agent to interact with can be changed at an arbitrary timing according to the intention of the user. Moreover, since the changed agent interacts with the scenario for that agent, a problem solving method different from the previous agent can be expected.

  In the experiment by the inventors, the possibility of such arbitrary change of the user was well received by the test subject, and in particular, there was an evaluation that “It was good to be able to choose the person whom he / she wants to talk to”.

  4 and the like are basically preset as programs in a memory (not shown) of the chat server 20 and / or the dialog server 26, and these memories are an example of a program storage medium. is there.

  However, the chat server 20 and the dialog server 26 share work by function as described above. That is, the chat server 20 manages and controls the chat with the user (user terminal 14), whereas the dialog server 26 has a function of managing and controlling the contents of the chat. Such function sharing enables optimal control of the servers 20 and 26. However, it is of course possible to manage and control the functions of the two servers 20 and 26 in a single server. That is, the servers 20 and 26 may be configured as a single server to be integrated.

DESCRIPTION OF SYMBOLS 10 ... Electronic listening system 12 ... Matching site 14 ... User terminal 16 ... Network 18 ... Contractor terminal 20 ... Chat server 26 ... Dialog server

Claims (5)

An electronic listening system using a user terminal,
An agent list display unit for displaying a list of at least two agent symbols each individually indicating an agent on the screen of the user terminal;
A selection determining unit for determining that the user has selected any agent symbol from the agent list display unit;
An agent display unit for displaying an agent corresponding to the agent symbol selected by the user in a state facing the avatar representing the user;
An electronic listening system comprising: an utterance unit that utters the agent according to a predetermined scenario; and an option display unit that displays an option according to the predetermined scenario according to the utterance of the agent. The screen of the user terminal includes an agent display unit that displays the agent symbol and an operation unit that displays the agent,
The electronic listening system according to claim 1, wherein the selection determination unit determines whether any agent symbol is dragged from the agent display unit to the operation unit. An agent change determination unit that determines that another agent symbol is dragged from the agent display unit to the operation unit and the agent is changed to another agent, and when the agent change determination unit determines an agent change, The electronic listening system according to claim 2, further comprising a scenario changing unit that changes a scenario so that the other agent speaks.   The electronic listening system according to claim 1, further comprising an additional agent display unit that displays an additional agent together with the agent. An electronic listening program executed by a computer of an electronic listening system using a user terminal, wherein the computer displays a list of at least two agent symbols each indicating an agent individually on the screen of the user terminal Part,
A selection determining unit for determining that the user has selected any agent symbol from the agent list display unit;
An agent display unit for displaying an agent corresponding to the agent symbol selected by the user in a state facing the avatar representing the user;
An electronic listening program that functions as an utterance unit that utters the agent according to a predetermined scenario, and an option display unit that displays an option according to the predetermined scenario according to the utterance of the agent.

Images