JP6914989B2 - Editing support system and editing support method - Google Patents

Description

An embodiment of the present invention relates to an editing support system, an editing support device, and an editing support method that support data editing related to a dialogue engine.

In recent years, various electronic devices including a dialogue engine that generates a response to a user's utterance input by voice or text and outputs the response by voice or text have been used. The user can obtain an answer corresponding to the question by, for example, uttering a question to such an electronic device.

Japanese Unexamined Patent Publication No. 2011-21718 Japanese Unexamined Patent Publication No. 2005-328172

An object to be solved by the present invention is to provide an editing support system, an editing support device, an editing support method, and a program that can easily edit scenario data used in a dialogue engine.

According to the embodiment, the editing support system is composed of a server device, a first client device, and a second client device. The server device uses the utterance data indicating the utterance by the user, and in order to generate the utterance data indicating the utterance in response to the utterance, a plurality of dialogues including the utterance by the user and the utterance generated by the server device. storing the scenario data indicating a scenario of a first editing scenario data, a scenario data for the second editing in the storage medium. The first editing scenario data includes a first scenario including one first utterance by the user and one second utterance generated by the server device among the plurality of scenarios indicated by the scenario data. the first and state transitions for the intended action at the first scenario, a editable by the first user using the first client device, a range that does not change the first state transition, the The first information indicating the first part of the first utterance and the second part of the second utterance, and the range that cannot be edited by the first user and changes the first state transition . The third part of the first utterance and the second information indicating the fourth part of the second utterance are shown. The second scenario data for editing, the first and the scenario, the first and the state transition, a editable by a second user who uses the second client device, the scope not changing the first state transition in it, a third information showing a sixth part of the fifth portion and said second speech of said first utterance, a non edited by the second user, changing the first state transition range shows a fourth information showing an eighth part of the seventh portion and said second speech of said first utterance. The server device transmits the first editing scenario data to the first client device, and transmits the second editing scenario data to the second client device. Wherein the first client device receives the scenario data for the first editing from the server apparatus. Wherein the first client device, wherein shown in a first editing scenario data, wherein the first scenario and the first information based on the second information, a first object representing the first utterance, A screen is displayed with a first object further indicating the first portion and the third portion, and a second object indicating the second utterance and further indicating the second portion and the fourth portion. Display in. When the first user performs an operation for editing at least one of the first portion and the second portion, the first client device has the edited first portion and the second portion. The first edit data indicating at least one of them is transmitted to the server device. The server device is further configured to receive the first edited data from the first client device, the first edited data using, Oite the first scenario illustrated in the scenario data stored, the first by changing at least one of the said second portion of the first portion and the second utterance of the speech, before rewriting the alkoxy Nariodeta. The second client device receives the scenario data for the second editing from the server apparatus. The second client device, the second indicated in the editing scenario data, based on the first scenario and the third information and the fourth information, a third object representing the first utterance, A screen of a third object further indicating the fifth part and the seventh part, and a fourth object indicating the second utterance and further indicating the sixth part and the eighth part. Display on. When the second user performs an operation for editing at least one of the fifth part and the sixth part, the second client device has the edited fifth part and the sixth part. The second edited data indicating at least one of them is transmitted to the server device. The server device is further configured to receive the second editing data from said second client device, the second editing data using a Oite the first scenario illustrated in the scenario data stored, the first by changing at least one of the said sixth portion of the fifth portion and said second utterance of the speech, before rewriting the alkoxy Nariodeta.

The figure for demonstrating the outline of the editing support system which concerns on embodiment. The block diagram which shows the system configuration of the server apparatus provided in the editing support system of the same embodiment. The block diagram which shows the system configuration of the client apparatus provided in the editing support system of the same embodiment. The block diagram which shows the functional structure of the server apparatus of FIG. 2 and the client apparatus of FIG. The flowchart which shows the example of the procedure of the dialogue processing executed by the server apparatus of FIG. The figure which shows one configuration example of the client data used by the server apparatus of FIG. The figure which shows the 1st example of the scenario edit screen displayed by the client apparatus of FIG. The figure which shows the 2nd example of the scenario edit screen displayed by the client apparatus of FIG. The figure which shows the 3rd example of the scenario edit screen displayed by the client apparatus of FIG. The figure which shows the 4th example of the scenario edit screen displayed by the client apparatus of FIG. The figure which shows the 5th example of the scenario edit screen displayed by the client apparatus of FIG. The figure which shows the sixth example of the scenario edit screen displayed by the client apparatus of FIG. The flowchart which shows the example of the procedure of the scenario editing process executed by the server apparatus of FIG. The flowchart which shows the example of the procedure of the scenario editing process executed by the client apparatus of FIG.

Hereinafter, embodiments will be described with reference to the drawings.
First, the configuration of the editing support system according to the embodiment will be described with reference to FIG. The editing support system 1 supports editing of scenario data for dialogue with a user, which is used in the dialogue system 2. The editing support system 1 includes a server device 10 and a client device 20. The server device 10 can be realized as, for example, a server computer. The client device 20 can be realized as an embedded system built in, for example, a mobile information terminal such as a smartphone, a tablet computer, a mobile phone, a PDA, a personal computer, or various electronic devices. Data can be exchanged between the server device 10 and the client device 20 via the network 40.

The server device 10 includes a dialogue engine 103B for dialogue with the user. The dialogue engine 103B has, for example, a function for understanding the content of an utterance by a user and generating an utterance in response to the utterance by referring to a scenario database 45 in which dialogue rules are defined. The dialogue engine 103B receives the utterance data of the utterance by the user, outputs the utterance data of the response corresponding to the utterance data, and further receives the utterance data of the utterance by the user according to the utterance data of the response. By repeating the exchange of data, the dialogue with the user is realized.

By using such a dialogue engine 103B, the server device 10 constitutes a dialogue system 2 with customer terminals 31 and 32 connected via the network 40. In this dialogue system 2, a dialogue between an utterance by a user using the customer terminals 31 and 32 and an utterance generated by the dialogue engine 103B of the server device 10 is realized. The customer terminal 31 can be realized as an embedded system built in, for example, a mobile information terminal such as a smartphone, a tablet computer, a mobile phone, or a PDA, a personal computer, or various electronic devices. The customer terminals 31 and 32 may be, for example, stationary terminals installed in a shopping mall, an office, or the like.

As shown in FIG. 1, the server device 10 receives the utterance data indicating the utterance by the user from the customer terminals 31 and 32, and transmits the utterance data indicating the utterance generated by the dialogue engine 103B to the customer terminals 31 and 32. .. By transmitting and receiving such utterance data, it is possible to realize a dialogue between the user who uses the customer terminals 31 and 32 and the server device 10.

As described above, the editing support system 1 of the present embodiment supports the editing of the scenario data used in the dialogue system 2 for dialogue with the user. This scenario data is, for example, data in the scenario database 45 referenced by the dialogue engine 103B. Further, the client device 20 is an editing terminal for editing scenario data in cooperation with the server device 10.

For example, when the dialogue system 2 is used by a customer who visits a shopping mall, the scenario database 45 includes data for dialogue that provides information about the shopping mall and information about each shop in the shopping mall. .. In such a case, the client device 20 includes, for example, a shopping mall terminal 21 used by the manager of the shopping mall and shop terminals 22 and 23 used by the manager of each shop.

When the client device 20 (for example, terminals 21, 22, 23) requests editing of the data in the scenario database 45, the server device 10 uses the scenario data for editing (hereinafter, also referred to as editing scenario data). It is transmitted to the client device 20. The client device 20 uses the received editing scenario data to display a screen for editing the scenario. The client device 20 transmits the edited data of the scenario based on the operation by the user (administrator) to the server device 10. Then, the server device 10 updates the data in the scenario database 45 based on the edited portion and the edited contents shown in the received edited data.

As described above, the scenario database 45 used by the dialogue engine 103B can be edited not only by the administrator of the server device 10 but also by the administrator of the shopping mall and the administrator of each shop in the shopping mall. However, if an administrator who is not familiar with the technology related to the dialogue engine 103B, such as a manager of a shopping mall or an administrator of each shop, can freely update the scenario database 45, the update can be performed, for example, a dialogue in a scenario. State may not transition properly or an inappropriate response may be generated. Therefore, the update may prevent the dialogue engine 103B from operating properly.

Therefore, in the editing support system 1 of the present embodiment, it is required to realize a new function that can edit the scenario data for dialogue according to the type of the administrator. In the editing support system 1, an administrator who is not familiar with the technology of the dialogue engine 103B speaks within a range that does not change the state transition for the action intended in the scenario (for example, store search, location search, etc.). Control to edit. The editing support system 1 controls, for example, an editable (changeable) part of the scenario data according to the type of the administrator.

Next, FIG. 2 shows the system configuration of the server device 10.
The server device 10 includes a CPU 101, a system controller 102, a main memory 103, a BIOS-ROM 105, a non-volatile memory 106, a communication device 107, an embedded controller (EC) 108, and the like.

The CPU 101 is a processor that controls the operation of various components in the server device 10. The CPU 101 executes various programs loaded from the non-volatile memory 106, which is a storage device, into the main memory 103. These programs include an operating system (OS) 103A and various application programs. The application program includes a dialogue engine 103B and an editing support server program 103C. The dialogue engine 103B provides a group of commands for accepting utterances by the user, understanding utterances to understand utterances, dialogue control for determining a response format, response generation for generating a specific response sentence according to the response format, and output of the response sentence. include. The editing support server program 103C includes a group of instructions for editing a scenario in cooperation with the client device 20. Here, the dialogue engine 103B realized as software and the editing support server program 103C are illustrated, but the functions corresponding to these programs 103B and 103C may be realized as hardware, or a combination of hardware and software. It may be realized as.

The CPU 101 also executes the basic input / output system (BIOS) stored in the BIOS-ROM 105. The BIOS is a program for hardware control.

The system controller 102 is a device that connects the local bus of the CPU 101 and various components. The system controller 102 also has a built-in memory controller that controls access to the main memory 103.

The communication device 107 is a device configured to perform wired or wireless communication. The communication device 107 includes a transmitting unit that transmits a signal and a receiving unit that receives the signal.

The EC108 is a one-chip microcomputer that includes an embedded controller for power management. The EC 108 has a function of turning on or off the power of the server device 10 according to the operation of the power button by the user.

Further, FIG. 3 shows the system configuration of the client device 20. Here, a case where the client device 20 is realized as a tablet computer will be illustrated.

The client device 20 includes a CPU 201, a system controller 202, a main memory 203, a BIOS-ROM 205, a non-volatile memory 206, a communication device 207, an embedded controller (EC) 208, and the like.

The CPU 201 is a processor that controls the operation of various components in the client device 20. The CPU 201 executes various programs loaded from the non-volatile memory 206, which is a storage device, into the main memory 203. These programs include an operating system (OS) 203A and various application programs. The application program includes the editing support client program 203B. The editing support client program 203B includes a group of instructions for editing a scenario in cooperation with the server device 10. Here, the editing support client program 203B realized as software is illustrated, but the function corresponding to this program 203B may be realized as hardware or as a combination of hardware and software.

The CPU 201 also executes the basic input / output system (BIOS) stored in the BIOS-ROM 205. The BIOS is a program for hardware control.

The system controller 202 is a device that connects the local bus of the CPU 201 and various components. The system controller 202 also has a built-in memory controller that controls access to the main memory 203.

The system controller 202 also has a built-in display controller that controls the LCD 204A used as the display monitor of the client device 20. The display signal generated by this display controller is sent to the LCD 204A. The LCD204A displays a screen image based on the display signal. A touch panel 204B is arranged on the upper surface side of the LCD 204A. The touch panel 204B is a capacitive pointing device for inputting on the screen of the LCD 204A. The contact position on the screen to which the finger is touched and the movement of the contact position are detected by the touch panel 204B.

Communication device 207 is a device configured to perform wired or wireless communication. The communication device 207 includes a transmitting unit that transmits a signal and a receiving unit that receives the signal.

The EC208 is a one-chip microcomputer that includes an embedded controller for power management. The EC208 has a function of turning on or off the power of the client device 20 according to the operation of the power button by the user.

The customer terminals 31 and 32 may have the same system configuration as the client device 20, for example. The CPUs of the customer terminals 31 and 32 are configured to execute an application program for exchanging utterance data with the dialogue engine 103B on the server device 10.

Next, FIG. 4 shows the functional configurations of the server device 10 and the client device 20.

First, the dialogue engine 103B executed by the server device 10 includes an utterance reception unit 41, a response control unit 42, a generation unit 43, and an output control unit 44. Further, the server device 10 includes a scenario database 45. The scenario database 45 is allocated, for example, a storage area in the non-volatile memory 106. The scenario database 45 uses the utterance data indicating the utterance by the user to generate the utterance data indicating the utterance in response to the utterance, and the utterance by the user and the utterance generated by the server device 10 (dialogue engine 103B). Includes scenario data 450 indicating a plurality of scenarios of dialogue defined by, and client data 455 relating to the client device 20 for editing the scenario data 450. The scenario data 450 includes, for example, utterance understanding model data 451 and response content data 452, response format data 453, and response generation model data 454.

The utterance comprehension model data 451 includes information for estimating the state of dialogue and understanding the utterance. The utterance understanding model data 451 is used, for example, to generate utterance intention information indicating the intention of utterance. The utterance comprehension model data 451 is, for example, an utterance expression including one or more words, a feature amount (for example, a voice feature amount) corresponding to the utterance expression, and utterance intention information indicating the content intended by the utterance expression. Contains information associated with. The feature quantity is, for example, a feature vector.

More specifically, the utterance understanding model data 451 associates, for example, the utterance expression "I want", the feature amount corresponding to this utterance expression, and the utterance intention information indicating the user's action "store search". Contains the information provided. Further, for example, the utterance understanding model data 451 includes information associated with the utterance expression "where", the feature amount corresponding to this utterance expression, and the utterance intention information indicating the user's action "location search". .. Further, for example, the utterance understanding model data 451 includes information in which the utterance expression "shoes", the feature amount corresponding to the utterance expression, and the utterance intention information indicating the object "shoes" are associated with each other. Further, for example, the utterance understanding model data 451 includes information in which the utterance expression "toilet", the feature amount corresponding to the utterance expression, and the utterance intention information indicating the object "toilet" are associated with each other.

The response content data 452 includes information provided as a response to the user's utterance. For example, when the dialogue engine 103B is used for a dialogue that provides various information about a store, the response content data 452 includes store information associated with the store name, products handled, price range, business hours, location, and the like. .. The response content data 452 includes, for example, information associated with the store name "shoe store A", the products handled indicating "leather shoes" and "sneakers", and the price range indicating "luxury".

The response format data 453 includes information for determining the format of the response sentence (response format) based on the utterance intention information. Response formats include, for example, requests, greetings, questions, responses, suggestions, confirmations, and answers.

More specifically, the response format data 453 includes, for example, information in which a combination of the action "unknown" and the target "unknown" is associated with the question format (request (X = action, Y = target)). Further, for example, the response format data 453 includes information in which a combination of the action “store search” and the target “unknown” is associated with the proposal format (offer (X = target, Y = store)). Further, for example, the response format data 453 includes information in which a combination of the action “location search” and the target “unknown” is associated with the question format (request (X = target)).

The response generation model data 454 includes information for generating a response statement. The response generation model data 454 includes, for example, information in which the response format and the response statement are associated with each other.
More specifically, the response generation model data 454 includes, for example, information in which the question format (request (X = target)) and the response sentence "I have a request for X" are associated with each other. Further, for example, the response generation model data 454 includes, for example, information in which the confirmation format (confirm (X = x)) and the response sentence "Is it okay to say that X is x?" Are associated with each other. Further, for example, the response generation model data 454 is associated with, for example, the proposal format (offer (X = target, Y = store)) and the response sentence "Y was found as a store handling X". Contains information.

The dialogue engine 103B uses such utterance understanding model data 451 and response content data 452, response format data 453, and response generation model data 454 to transmit and receive utterance data to and from customer terminals 31 and 32.

More specifically, the utterance receiving unit 41 receives the utterance data from the customer terminals 31 and 32 used by the user via the communication device 107. The utterance data indicates the utterance by the user. The utterance data has an arbitrary data format, for example, voice data, voice feature amount data, text data, and the like. When the utterance data is voice data, the utterance reception unit 41 may extract a feature amount using the voice data, or obtains text data indicating the utterance by performing voice recognition processing on the voice data. You may.

The response control unit 42 uses the utterance understanding model data 451 to specify the utterance intention information corresponding to the utterance expression included in the utterance data in order to estimate the utterance intention by the user. For example, when the utterance includes the utterance expression "shoes" and the utterance expression "wants", the response control unit 42 includes the utterance intention information indicating the target "shoes" and the "store search". The utterance intention information indicating the action is specified. That is, the response control unit 42 can estimate that the intention of the user's utterance is "store search" of "shoes" based on the utterance expression "I want shoes".

Next, the response control unit 42 determines the response format based on the specified utterance intention information by using the response format data 453 and the response content data 452. The response control unit 42 uses the response format data 453 and the response content data 452 to respond based on, for example, the utterance intention information indicating the target "shoes" and the utterance intention information indicating the action "store search". As the format, the proposal format (offer (X = shoes, Y = shoe store A)) is determined.

The generation unit 43 uses the response generation model data 454 to generate a response sentence based on the determined response format. More specifically, the generation unit 43 has found, for example, the response sentence "Shoe store A as a store that handles shoes" corresponding to the proposal format (offer (X = shoes, Y = shoe store A)). "Is generated.

The output control unit 44 outputs the generated response statement. The output control unit 44 can output the response statement in any format. The output control unit 44 transmits, for example, the read-aloud voice data of the response sentence to the customer terminals 31 and 32, which are the sources of the utterance data by the user. Further, the output control unit 44 may transmit text data indicating a response sentence to the customer terminals 31 and 32. Further, for example, the output control unit 44 may transmit both the read-aloud voice data of the response sentence and the text data indicating the response sentence to the customer terminals 31 and 32.

An example of the procedure of the dialogue processing executed by the server device 10 will be described with reference to the flowchart of FIG. This dialogue processing is realized, for example, by the CPU 101 of the server device 10 executing the instruction group included in the dialogue engine 103B.

First, the utterance reception unit 41 determines whether or not the utterance data of the utterance by the user has been received from the customer terminals 31 and 32 (step S11). If the utterance data has not been received, the process returns to step S11, and it is determined again whether or not the utterance data has been received.

When the utterance data is received (YES in step S11), the response control unit 42 generates the divided data 492 by dividing the utterance data 491 into words (step S12). The utterance data 491 includes, for example, voice data corresponding to the utterance "I want shoes". By analyzing the utterance data 491, the response control unit 42 generates divided data 492 indicating the words "shoes", "ga", "want", and "i" included in the utterance.

The response control unit 42 extracts the feature amount 493 for each word included in the divided data 492 (step S13). The feature amount 493 for each word is, for example, a feature vector sequence.

Next, the response control unit 42 executes the utterance understanding process using the utterance understanding model data 451 (step S14). The response control unit 42 generates the utterance intention information 494 from the feature amount 493 by executing the utterance understanding process. The utterance intention information 494 includes information indicating the action intended by the utterance and the target, and includes, for example, information indicating that the action is "store search" and the target is "shoes".

Then, the response control unit 42 executes the response control process using the response content data 452 and the response format data 453 (step S15). The response control unit 42 determines the response format 495 based on the feature amount 493 and the utterance intention information 494 by executing the response control process. The response format 495 includes information indicating the type of response and its contents (for example, target, store, etc.), for example, the type is "offer", the target is "shoes", and the store It contains information indicating that it is "shoe store A".

The generation unit 43 generates the response sentence 496 from the response format 495 using the response generation model data 454 (step S16). The generation unit 43 generates, for example, "Shoe store A was found as a store that handles shoes" as a specific response sentence 496 for communicating the contents of the response format 495 to the user.

The output control unit 44 transmits the response sentence 496 to the customer terminals 31 and 32 via the communication device 107 (step S17). The output control unit 44 may transmit the response sentence 496 as voice data or text data. In the customer terminals 31 and 32, the response sentence 496 may be output as voice from the speaker or earphone, or may be displayed as text on the screen of the display.

Next, the utterance reception unit 41 determines whether or not the dialogue has ended (step S18). If the dialogue is not completed (NO in step S18), the process returns to step S11. When the dialogue ends, this dialogue processing ends.

As a result, dialogue processing can be performed between the server device 10 constituting the dialogue system 2 and the customer terminals 31 and 32.

Returning to FIG. 4, the functional configuration of the editing support server program 103C executed by the server device 10 will be described. The editing support server program 103C includes a reception control unit 46, a transmission control unit 48, and an editing processing unit 47.

The reception control unit 46 receives, for example, a request for scenario data for editing (scenario data for editing) from the client device 20 via the communication device 107.

In response to this request, the editing processing unit 47 and the transmission control unit 48 transmit the editing scenario data corresponding to at least one scenario to the client device 20. More specifically, the editing processing unit 47 acquires editing scenario data according to the requesting client device 20. The editing processing unit 47, for example, by referring to the client data 455, makes a specific edit from the scenario database 45 based on the client ID assigned to the client device 20, the authority given to the client device 20, and the like. Read the scenario data for the client.

FIG. 6 shows a configuration example of client data 455. The client data 455 includes a plurality of records corresponding to the plurality of client devices 20. Each record contains, for example, a client ID, editing scenario data, and permissions.

In the record corresponding to a certain client device 20, the "client ID" indicates the identification information given to the client device 20. The "client ID" may be set with the identification information given to the user (administrator) who uses the client device 20.

“Editing scenario data” indicates editing scenario data for the client device 20. More specifically, in the "editing scenario data", for example, a storage location (for example, a file name, a file path, etc.) of the editing scenario data can be set. Further, in the "editing scenario data", in order to specify the data used as the editing scenario scenario data among the speech understanding model data 451 and the response content data 452, the response format data 453, and the response generation model data 454. Information may be set. Further, in the "editing scenario data", at least one of the utterance understanding model data 451 and the response content data 452, the response format data 453, and the response generation model data 454 is used to generate the editing scenario data. Information may be set.

“Authority” indicates the authority to edit the scenario given to the client device 20. In "authority", for example, the authority level of the client device 20 is set. In the editing support system 1, the editable part and the non-editable part in the scenario can be controlled according to the level of this authority.

By using such client data 455, the editing processing unit 47 can acquire editing scenario data according to the requesting client device 20. The editing scenario data is a part of the scenario data 450, and indicates a scenario in which a user using the client device 20 can edit at least a part of the scenario data 450. The editing scenario data may be created in advance, and in response to a request from the client device 20, the editing processing unit 47 sends the scenario data 450 (that is, the speech understanding model data 451 and the response content data 452, Specific data may be read out as editing scenario data from the response format data 453 and the response generation model data 454). Further, the editing processing unit 47 may generate editing scenario data using the scenario data 450, if requested.

The transmission control unit 48 transmits the read editing scenario data to the client device 20 via the communication device 107. As a result, the server device 10 can cause the client device 20 to display the scenario editing screen based on the editing scenario data and present the object for editing.

Further, the reception control unit 46 receives the edited data indicating the edited scenario from the client device 20. The editing processing unit 47 uses this editing data to update the data in the scenario database 45. The edited data includes, for example, a part of the scenario shown in the transmitted editing scenario data edited by the user of the client device 20 and its edited contents, and a part added to the scenario by the user of the client device 20 and its addition thereof. The contents etc. are shown. The editing processing unit 47 rewrites (updates) the scenario data 450 based on the editing data. The editing processing unit 47 is included in at least one of the utterance understanding model data 451 and the response content data 452, the response format data 453, and the response generation model data 454 based on the edited portion of the scenario and the edited content thereof, for example. Rewrite the corresponding data.

With the above configuration, the editing support system 1 can rewrite the scenario data 450 according to the editing by the user of the client device 20. Then, in the dialogue system 2, a dialogue can be performed between the server device 10 and the customer terminals 31 and 32 based on the rewritten scenario data 450.

Next, the functional configuration of the editing support client program 203B executed by the client device 20 will be described. The editing support client program 203B includes a reception control unit 51, a display control unit 52, an operation reception unit 53, and a transmission control unit 54.

The operation reception unit 53 receives various operations by the user who uses the client device 20. The operation reception unit 53 receives, for example, an operation for requesting editing scenario data. The operation reception unit 53 determines that the editing scenario data has been requested, for example, when the editing support client program 203B is started or when a specific operation is performed by the user.

When the editing scenario data is requested, the transmission control unit 54 requests the editing scenario data from the server device 10 via the communication device 207. The reception control unit 51 receives the editing scenario data from the server device 10 via the communication device 207.

The display control unit 52 displays the scenario editing screen on the LCD 204A using the received editing scenario data. The display control unit 52 indicates, for example, a first utterance by the user, a first object indicating an editable first portion of the first utterance, and a second utterance generated by the server device 10. A second object indicating the editable second part of the second utterance is displayed (presented) on the screen of the LCD204A. An example of the scenario editing screen will be described later with reference to FIGS. 7 to 12.

Further, the operation reception unit 53 and the display control unit 52 accept editing of the scenario according to the operation on the displayed scenario editing screen. This operation is an operation using, for example, a keyboard or a pointing device such as the touch panel 204B. The display control unit 52 changes at least a part of the object indicating the utterance on the scenario editing screen, or adds a new object indicating the utterance on the scenario editing screen according to the operation by the user. Further, the operation reception unit 53 generates edit data indicating the edited part and the edited content in the scenario based on the operation by the user.

The edited data includes, for example, when at least a part of the object is changed on the scenario edit screen by the user of the client device 20, data indicating the changed part and the changed content, and on the scenario edit screen. When an object is added in, it contains data indicating the added part and the added contents. As described above, the server device 10 can update the data in the scenario database 45 based on the edited data.

The scenario shown in the scenario data may include not only utterances but also images. This image is, for example, an image displayed on the screens of the customer terminals 31 and 32 in accordance with the utterance in response to the utterance by the user. For the image, for example, an image of a character of a shopping mall or each shop may be used. Therefore, in the editing support system 1, the image can be edited (changed) in the same manner as the utterance included in the scenario.

Here, some examples of specific operations in the editing support system 1 are shown.
First, the editing processing unit 47 and the transmission control unit 48 of the server device 10 make a dialogue in the shopping mall terminal 21 (first client device) including the utterance by the user and the utterance generated by the server device 10 (dialogue engine 103B). The case where the first scenario data indicating the first scenario of the above is transmitted will be illustrated. This first scenario data is obtained by the user of the shopping mall terminal 21 in the first part of the first utterance by the user that can be edited by the user of the shopping mall terminal 21 and the second utterance generated by the server device 10. It may contain information indicating an editable second part. These first and second parts can be arbitrarily set according to, for example, the type of the client device 20 and the type of the user who uses the client device 20.

The reception control unit 51 of the shopping mall terminal 21 receives the first scenario data. The display control unit 52 is a first object indicating the first utterance by the user based on the first scenario shown in the first scenario data, and further indicates an editable first portion of the first utterance. One object and a second object indicating the second utterance generated by the server device 10 and further indicating the editable second part of the second utterance are displayed on the screen of the LCD 204A. .. Then, when the user using the shopping mall terminal 21 performs an operation for editing at least one of the first portion and the second portion, the operation reception unit 53 and the transmission control unit 54 are edited first. The first edit data indicating at least one of the portion and the second portion is transmitted to the server device 10.

The reception control unit 46 of the server device 10 receives the first edit data. The editing processing unit 47 uses the first editing data to change at least one of the first part of the first utterance and the second part of the second utterance included in the first scenario, and the first 1 Rewrite the scenario data.

Thereby, only the first part and the second part in the first scenario can be controlled to be edited by the user of the shopping mall terminal 21.

The first scenario data includes a third part of the first utterance that cannot be edited by the user of the shopping mall terminal 21 and a fourth part of the second utterance that cannot be edited by the user of the shopping mall terminal 21. It may include information to indicate. In that case, the first object displayed by the display control unit 52 of the shopping mall terminal 21 further indicates the non-editable third part in the first utterance, and the second object is the edit in the second utterance. The impossible fourth part may be further shown.

In addition, the first scenario data may include information indicating a position where a new utterance can be added to the dialogue of the first scenario. In that case, the display control unit 52 of the shopping mall terminal 21 displays the fifth object for adding a new utterance to the first scenario on the screen. The operation reception unit 53 and the transmission control unit 54 are additional data indicating the third utterance when the user using the shopping mall terminal 21 performs an operation for adding the third utterance using the fifth object. Is transmitted to the server device 10.

The reception control unit 46 of the server device 10 receives the additional data. The editing processing unit 47 uses this additional data to rewrite the first scenario data so as to add the third utterance to the first scenario.

As a result, a new utterance can be added to the first scenario, and the position where the new utterance is added can be controlled.

Next, a case where the editing processing unit 47 and the transmission control unit 48 of the server device 10 transmit the first scenario data to the shop terminal 22 (second client device) will be illustrated. The transmitted first scenario data includes the fifth part of the first utterance by the user, which can be edited by the user using the shop terminal 22, and the shop terminal 22 in the second utterance generated by the server device 10. Contains information indicating a sixth part that can be edited by the user. The first part of the first utterance editable by the user using the shopping mall terminal 21 may differ from the fifth part of the first utterance editable by the user using the shop terminal 22. Similarly, the second part of the second utterance editable by the user using the shopping mall terminal 21 may differ from the sixth part of the second utterance editable by the user using the shop terminal 22. This is because the authority for the user using the shopping mall terminal 21 to edit (change) the first scenario is different from the authority for the user using the shop terminal 22 to edit the first scenario.

The display control unit 52 of the shop terminal 22 is a third object indicating the first utterance based on the first scenario shown in the first scenario data, and the third object further indicating the fifth part and the second utterance. The fourth object indicating the sixth part and the fourth object further indicating the sixth part are displayed on the screen of the LCD 204A.

Then, when the user who uses the shop terminal 22 performs an operation for editing at least one of the fifth part and the sixth part, the operation reception unit 53 of the edited fifth part and the sixth part. The second edit data indicating at least one of them is transmitted to the server device 10.

The reception control unit 46 of the server device 10 receives the second edited data. The editing processing unit 47 uses the second editing data to change at least one of the fifth part of the first utterance and the sixth part of the second utterance included in the first scenario, and the second 1 Rewrite the scenario data.

As a result, only the fifth and sixth parts of the first scenario can be controlled to be edited by the user of the shop terminal 22.

Next, the editing processing unit 47 and the transmission control unit 48 of the server device 10 tell the shopping mall terminal 21 the first scenario data showing the first scenario and the second scenario showing the second scenario of the dialogue including the first speech. An example is given in the case of transmitting data. That is, both the first scenario and the second scenario include the first utterance.

The reception control unit 51 of the shopping mall terminal 21 receives the first scenario data and the second scenario data from the server device 10. The display control unit 52 displays the first page, which is the first page based on the first scenario shown in the first scenario data and includes the first object and the second object, on the screen of the LCD 204A. As described above, the first object represents a first utterance by the user and an editable first portion of the first utterance. The second object also indicates a second utterance generated by the server device 10 and an editable second portion of the second utterance.

Next, the operation reception unit 53 and the display control unit 52 are based on the second scenario shown in the second scenario data after the operation for editing the first portion is performed by the user who uses the shopping mall terminal 21. The second page is the sixth object indicating the first utterance, and the editable third part of the first utterance includes the sixth object modified based on the edited contents for the first part. Is displayed on the screen.

More specifically, the operation reception unit 53 receives an operation for editing the first portion by the user who uses the shopping mall terminal 21, and the display control unit 52 displays the edited first portion. After that, when the operation reception unit 53 receives an operation for displaying the second page based on the second scenario shown in the second scenario data by the user who uses the shopping mall terminal 21, the operation reception unit 53 is based on the second scenario. The sixth object indicating the first utterance, and the editable third part of the first utterance, displays the sixth object modified based on the edited contents for the first part on the screen. Further, the transmission control unit 54 transmits the third edited data indicating the edited first portion and the modified third portion to the server device 10.

The reception control unit 46 of the server device 10 receives the third edited data from the shopping mall terminal 21. The editing processing unit 47 uses the third editing data to change the first part of the first utterance included in the first scenario, rewrite the first scenario data, and include it in the second scenario. The third part of the first utterance is changed to rewrite the second scenario data.

With such a configuration, the editing content on the first page based on the first scenario is also reflected on the second page based on the second scenario, so that the complexity of the editing work by the user can be reduced. ..

Next, FIGS. 7 to 12 show some examples of the scenario editing screen displayed by the client device 20.

First, the scenario editing screen 6 shown in FIG. 7 includes a plurality of scenario editing pages 61, 62, 6N for editing a plurality of scenarios. These plurality of scenario edit pages 61, 62, 6N correspond to, for example, a plurality of scenarios using the same type of template. The template is, for example, a template in which at least one of utterance and state transition is defined. The plurality of scenario editing pages 61, 62, 6N are arranged so as to overlap in the scenario editing screen 6. The user selects one of the plurality of scenario edit pages 61, 62, 6N, for example, by tapping any position in a certain page, the page is displayed on the plurality of scenario edit pages 61, 62. , Can be placed at the top of 6N. The user can edit the scenario for one page arranged at the top of the page.

On each scenario edit page 61, 62, 6N, the utterance can be edited within the range in which the state transition for the action intended in the scenario (for example, store search, place search, etc.) is not changed. Editing this utterance includes, for example, changing the content of the utterance or adding an utterance.

As shown in FIG. 7, on the first scenario edit page 61 arranged at the top, the utterance object 71 showing the utterance generated by the system (that is, the utterance generated by the server device 10) is displayed on the left side of the page. , 73, 75 are included, and on the right side of the page, utterance objects 72, 74 indicating utterances by the user are included. In addition, the first scenario edit page 61 includes additional objects 791,793 for adding system-generated utterances on the left side of the page, and additions for adding user utterances on the right side of the page. An additional object 792 is included. The additional objects 791,792,793 are arranged, for example, at positions in the scenario where utterances can be added.

In each utterance object 71-75, in the text indicating the content of the utterance, a part 711,722,732,752 that can be edited by the user and a part 712,721,731,741,751 that cannot be edited are shown. .. For example, in the utterance object 71, the editable part "Happy New Year" 711 and the non-editable part "What happened" 712 are shown.

The editable portion 711,722,732,752 and the non-editable portion 712,721,731,741,751 are displayed so as to be identified by the user, respectively. In the example shown in FIG. 7, editable portions 711,722,732,752 are underlined and non-editable portions 712,721,731,741,751 are shown by being enclosed in a rectangle. The display form for distinguishing the editable part 711,722,732,752 from the non-editable part 712,721,731,741,751 is not limited to this example, and different colors, highlights, bold, etc. Any display form may be used as long as these two types of parts can be identified.

Further, depending on the user performing a specific operation (for example, a tap operation) on the editable portion 711,722,732,752, any position of the editable portion 711,722,732,752. A cursor may be displayed at and the editing of the text representing the content of the utterance may be started. It should be noted that any operation on the non-editable part 712,721,731,741,751 may not be accepted. As a result, it is possible to control the scenario to be edited within the range of authority given to the user of the client device 20.

If the user is, for example, an administrator of a shopping mall, the editable content included in the object 71 is to change the greeting of the time set at the time of the New Year's sale to the greeting of the time of the spring sale. You can perform an operation to change the part "Happy New Year" 711 to "It looks like spring."

As shown in FIG. 8, the editable part "Happy New Year" 711 has been changed to "It looks like spring" 711A. The editable portion 711A may be displayed, for example, in different colors, highlights, bold, etc. so that the changes can be identified. The user can also further modify this editable portion 711A.

The plurality of scenario editing pages 61, 62, 6N correspond to, for example, a plurality of scenarios using the same type of template and a plurality of scenarios including the same utterance. Therefore, the change to the editable part 711 on the first scenario editing page 61 may be reflected in the corresponding part on the other scenario editing pages 62 and 6N. As a result, as shown in FIG. 9, on the second scenario editing page 62, the editable part “It looks like spring” 811A corresponding to the greeting of the time is shown on the object 81.

In addition, when the user is, for example, the manager of a certain shop in a shopping mall, a specific value included in the utterance, for example, a product that fluctuates in a period specified by a month, a week, a day, an hour, or the like. It is also possible to change the value of the discount rate, or change the ending of the utterance in order to change the impression of the utterance to an impression based on the concept of the shop (for example, a polite impression, a casual impression, etc.). It is also possible to prevent the manager of such a shop from changing the utterance including the information of the entire shopping mall and the information of other shops.

For example, a user who is a shop administrator can change the editable part "20% off" 854 as shown in FIG. 9 to "40% off" 854A shown in FIG. It can be performed. Further, on the scenario editing screen 6 shown in FIG. 10, utterances (parts) other than the editable part 854A are displayed so that they cannot be edited. Thereby, the editable part in the scenario can be controlled according to the type of the client device 20 or the type of the user who uses the client device 20.

Next, FIG. 11 shows an example in which an utterance is added to the scenario. Utterances can be added to the scenario as long as the state transition for the action intended in the scenario (eg, store search, location search, etc.) is not changed. For example, it is possible to add a chat utterance (dialogue) in which the state transition is not changed to the scenario.

In the example shown in FIG. 11, an object 76 indicating the added utterance is shown. The object 76 may be displayed so that it can be identified that it has been newly added.

Further, when a new utterance is added, information 76A indicating whether or not the added utterance is appropriate for the scenario may be displayed in association with the object 76. In this information 76A, for example, the input utterance is already used in another scenario, the input utterance may change the template for the action intended in this scenario, and so on. Shown. The user can also perform an operation for changing the content of the utterance or an operation for canceling the addition of the utterance according to the information 76A.

Further, as shown in the scenario edit screen 9 of FIG. 12, if the scenario of the scenario edit page 91 including the objects 96 and 97 indicating the added utterances is already registered in the server device 10, it is registered in the user. A dialog 99 asking whether or not to do so may be displayed. The dialog 99 includes, for example, a button 991 for registering a scenario and a button 992 for canceling the registration of the scenario. The user can instruct to register the scenario or cancel the registration by selecting any of the buttons 991 and 992.

Next, an example of the procedure of the scenario editing process executed by the server device 10 will be described with reference to the flowchart of FIG. This scenario editing process is realized, for example, by the CPU 101 of the server device 10 executing the instruction group included in the editing support server program 103C.

First, the reception control unit 46 determines whether or not the scenario data for editing is requested from the client device 20 (step S21). More specifically, the reception control unit 46 determines whether or not data indicating a request for a scenario for editing has been received from the client device 20 via the communication device 107. If the scenario for editing is not requested from the client device 20 (NO in step S21), the process returns to step S21, and it is determined again whether or not the scenario for editing is requested from the client device 20.

When a scenario for editing is requested from the client device 20 (YES in step S21), the editing processing unit 47 reads out the editing scenario data corresponding to the requested client device 20 from the scenario database 45 (step S22). .. Then, the transmission control unit 48 transmits the read editing scenario data to the client device 20 (step S23). The client device 20 displays a scenario editing screen based on the editing scenario data.

Next, the reception control unit 46 determines whether or not the edit data indicating the edited scenario has been received from the client device 20 (step S24). If the edited data has not been received (NO in step S24), the process returns to step S24, and it is determined again whether or not the edited data has been received.

On the other hand, when the edited data is received (YES in step S24), the editing processing unit 47 updates the data in the scenario database 45 using the edited data (step S25).

Next, an example of the procedure of the scenario editing process executed by the client device 20 will be described with reference to the flowchart of FIG. This scenario editing process is realized, for example, by the CPU 201 of the client device 20 executing the instruction group included in the editing support client program 203B.

First, the operation reception unit 53 determines whether or not the display of the scenario for editing is requested (step S31). The operation reception unit 53 determines that the display of the scenario for editing is requested, for example, when the editing support client program 203B is started, or when a specific operation is performed by the user. If the display of the scenario for editing is not requested (NO in step S31), the process returns to step S31, and it is determined again whether or not the display of the scenario for editing is requested.

When the display of the scenario for editing is requested (YES in step S31), the transmission control unit 54 requests the server device 10 for the scenario for editing via the communication device 207 (step S32). The reception control unit 51 receives the editing scenario data from the server device 10 via the communication device 207 (step S33). Then, the display control unit 52 displays the scenario editing screen on the LCD 204A using the received editing scenario data (step S34). An example of this scenario editing screen is as described above with reference to FIGS. 7 to 12.

Next, the operation receiving unit 53 and the display control unit 52 accept editing of the scenario according to the operation on the displayed scenario editing screen (step S35). The display control unit 52 changes at least a part of the object indicating the utterance on the scenario editing screen, or adds a new object indicating the utterance on the scenario editing screen according to the operation.

Then, the operation reception unit 53 determines whether or not the editing of the scenario is completed (step S36). The operation reception unit 53 determines that the editing of the scenario is completed based on a specific operation indicating the completion of editing (for example, an operation of tapping the editing completion button). If the scenario editing is not completed (NO in step S36), the process returns to step S35, and further scenario editing is accepted.

When the editing of the scenario is completed (YES in step S36), the transmission control unit 54 transmits the editing data indicating the edited content to the server device 10 (step S37).

As described above, according to the present embodiment, the scenario data used in the dialogue engine can be easily edited. The scenario database 45 of the server device 10 uses the utterance data indicating the utterance by the user to generate the utterance data by the user and the utterance generated by the server device 10 in order to generate the utterance data indicating the utterance in response to the utterance. Stores scenario data showing multiple scenarios of the included dialogue. The transmission control unit 48 transmits the first scenario data indicating the first scenario among the plurality of scenarios to the client device 20. The reception control unit 51 of the client device 20 receives the first scenario data from the server device 10. The display control unit 52 is a first object indicating the first utterance by the user based on the first scenario shown in the first scenario data, and further indicates an editable first part of the first utterance. The first object and the second object indicating the second utterance generated by the server device 10 and further indicating the editable second part of the second utterance are displayed on the screen of the LCD 204A. do. When the user who uses the client device 20 performs an operation for editing at least one of the first portion and the second portion, the operation reception unit 53 and the transmission control unit 54 have the edited first portion and the first portion. The first edit data indicating at least one of the two parts is transmitted to the server device 10. The reception control unit 46 of the server device 10 further receives the first edit data from the client device 20, and the edit processing unit 47 uses the first edit data to include the first utterance included in the first scenario. The first scenario data is rewritten by changing at least one of the first part and the second part of the second utterance.

The first part and the second part can be arbitrarily set according to, for example, the type of the client device 20 and the administrator. Therefore, since the editable (changeable) part of the scenario data for dialogue is controlled according to the type of the client device 20 and the administrator, the scenario data 450 used in the dialogue engine 103B can be easily and appropriately used. Can be edited to.

Further, each of the various functions described in the present embodiment may be realized by a circuit (processing circuit). Examples of processing circuits include programmed processors such as central processing units (CPUs). This processor executes each of the described functions by executing a computer program (instruction group) stored in the memory. This processor may be a microprocessor including an electric circuit. Examples of processing circuits also include digital signal processors (DSPs), application specific integrated circuits (ASICs), microcontrollers, controllers, and other electrical circuit components. Each of the components other than the CPU described in this embodiment may also be realized by a processing circuit.

Further, since various processes of the present embodiment can be realized by a computer program, the present embodiment and the present embodiment can be obtained only by installing and executing the computer program on a computer through a computer-readable storage medium in which the computer program is stored. A similar effect can be easily achieved.

Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other embodiments, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are also included in the scope of the invention described in the claims and the equivalent scope thereof.

1 ... Editing support system, 2 ... Dialogue system, 10 ... Server device, 20 ... Client device (editing terminal), 31, 32 ... Customer terminal, 40 ... Network, 103B ... Dialogue engine, 103C ... Editing support server program, 41 ... Speech reception unit, 42 ... Response control unit, 43 ... Generation unit, 44 ... Output control unit, 45 ... Scenario database, 450 ... Scenario data, 451 ... Speech understanding model data, 452 ... Response content data, 453 ... Response format data, 454 ... Response generation model data, 455 ... Client data, 46 ... Reception control unit, 47 ... Editing processing unit, 48 ... Transmission control unit, 203B ... Editing support client program, 51 ... Reception control unit, 52 ... Display control unit, 53 … Operation reception unit, 54… Transmission control unit.

Claims (5)

It is an editing support system composed of a server device, a first client device, and a second client device.
The server device uses the utterance data indicating the utterance by the user, and in order to generate the utterance data indicating the utterance in response to the utterance, a plurality of dialogues including the utterance by the user and the utterance generated by the server device. and scenario data indicating a scenario of a first editing scenario data is configured to store a second editing scenario data in a storage medium,
The first editing scenario data is
Among the plurality of scenarios indicated by the scenario data, a first scenario including one first utterance by the user and one second utterance generated by the server device, and
The first state transition for the action intended in the first scenario ,
The first part of the first utterance and the second of the second utterances, which are editable by the first user using the first client device and do not change the first state transition. The first information showing the part and
With the second information indicating the third part in the first utterance and the fourth part in the second utterance, which is not editable by the first user and is a range for changing the first state transition. , Indicates
The second editing scenario data is
The first scenario and
With the first state transition
The fifth part of the first utterance and the sixth part of the second utterance, which are editable by the second user using the second client device and do not change the first state transition. The third information showing the part and
With the fourth information indicating the seventh part in the first utterance and the eighth part in the second utterance, which is not editable by the second user and is a range for changing the first state transition. , Indicates
The server device
The first editing scenario data is transmitted to the first client device,
It is configured to transmit the second editing scenario data to the second client device.
The first client device is
Receiving said scenario data for the first editing from the server device,
It is shown in the first editing scenario data, based on the first scenario and the first information and the second information, a first object representing the first utterance, the said first portion second A first object further indicating the three parts and a second object indicating the second utterance, which further indicates the second part and the fourth part, are displayed on the screen.
When the first user performs an operation for editing at least one of the first portion and the second portion, the first edit indicating at least one of the edited first portion and the second portion is performed. It is configured to send data to the server device
The server device further
The first edit data is received from the first client device, and the first edit data is received.
Using the first editing data, Oite the first scenario illustrated in the scenario data stored, at least before Symbol said second portion of said first portion and of the second utterance of the first utterance by changing one, it is constructed prior to rewriting the carboxymethyl Nariodeta,
The second client device is
Receiving said scenario data for the second editing from the server apparatus,
The second is shown in the editing scenario data, based on the first scenario and the third information and the fourth information, a third object representing the first utterance, the said fifth portion second A third object further indicating the seven parts and a fourth object indicating the second utterance, which further indicates the sixth part and the eighth part, are displayed on the screen.
When the second user performs an operation for editing at least one of the fifth part and the sixth part, a second edit indicating at least one of the edited fifth part and the sixth part is performed. It is configured to send data to the server device
The server device further
The second edit data is received from the second client device, and the second edit data is received.
Using said second editing data, Oite the first scenario illustrated in the scenario data stored before Symbol of the sixth portion of said fifth portion of said second utterance of the first utterance by changing at least one, before editing support system configured to rewrite the carboxymethyl Nariodeta. The second scenario of the plurality of scenario illustrated by the scenario data are scenarios dialogue including the first utterance,
The server device is further configured to store the third editing scenario data in the storage medium.
The third editing scenario data is
Said second scenario,
The second state transition for the action intended in the second scenario ,
The fifth information indicating the ninth part of the first utterance, which is editable by the first user and does not change the second state transition , is shown .
The server device is further configured to transmit the first editing scenario data and the third editing scenario data to the first client device.
The first client device further
The first editing scenario data and the third editing scenario data are received from the server device, and the data is received.
It is shown in the first editing scenario data, based on the first scenario and the first information and the second information, to display the first page and a second object and the first object on the screen ,
By the first user, after the operation for editing the first portion is performed, the third is shown in the editing scenario data, on the basis of said fifth information and the second scenario, the first Claim 1 configured to display on the screen a second page of a sixth object indicating an utterance, the ninth portion of which includes a sixth object modified based on an edit to the first portion. Described editing support system. The first client device is further configured to transmit third edited data indicating the edited first portion and the modified ninth portion to the server device.
The server device further
The third edit data is received from the first client device, and the third edit data is received.
Using said third editing data, Oite the first scenario illustrated in the scenario data stored by previous SL modify the first portion of the first utterance, before rewriting a carboxymethyl Nariodeta by changing the ninth portion of said scenario data Oite before Symbol first utterance to the second scenario illustrated in storing, before carboxymethyl Nariodeta claim 2 configured to rewrite the Editing support system. It is an editing support method of an editing support system composed of a server device, a first client device, and a second client device.
The editing support method uses utterance data indicating utterances by the user, and in order to generate utterance data indicating utterances in response to the utterances, a plurality of dialogues including utterances by the user and utterances generated by the server device. and scenario data indicating a scenario of stores a first editing scenario data, a scenario data for the second editing in a storage medium in the server apparatus,
The first editing scenario data is
Among the plurality of scenarios indicated by the scenario data, a first scenario including one first utterance by the user and one second utterance generated by the server device, and
A first state transition for the action intended by the first scenario,
A editable by the first user using the first client device, a range that does not change the first state transition, a second of the first portion and the second utterance of the first utterance The first information showing the part and
With the second information indicating the third part in the first utterance and the fourth part in the second utterance, which is not editable by the first user and is a range for changing the first state transition. , Indicates
The second editing scenario data is
The first scenario and
With the first state transition
The fifth part of the first utterance and the sixth part of the second utterance, which are editable by a second user using the second client device and do not change the first state transition. The third information indicating
With the fourth information indicating the seventh part in the first utterance and the eighth part in the second utterance, which is not editable by the second user and is a range for changing the first state transition. , Indicates
The editing support method further
The first editing scenario data is transmitted from the server device to the first client device,
The second editing scenario data is transmitted from the server device to the second client device.
It is shown in the first editing scenario data, based on the first scenario and the first information and the second information, a first object representing the first utterance, the said first portion second A first object further indicating the three parts and a second object indicating the second utterance and further indicating the second part and the fourth part are displayed on the screen of the first client device. Display and
When the first user performs an operation for editing at least one of the first portion and the second portion, the first edit indicating at least one of the edited first portion and the second portion is performed. Data is transmitted to the server device,
Using the first editing data, Oite the first scenario illustrated in the scenario data stored in the storage medium, before Symbol of said second utterance and the first portion of the first utterance by changing at least one of said second portion, rewriting said sheet Nariodeta,
The second is shown in the editing scenario data, based on the first scenario and the third information and the fourth information, a third object representing the first utterance, the said fifth portion second A third object further indicating the seven parts and a fourth object indicating the second utterance, which further indicates the sixth part and the eighth part, are displayed on the screen of the second client device. Display and
When the second user performs an operation for editing at least one of the fifth part and the sixth part, a second edit indicating at least one of the edited fifth part and the sixth part is performed. Data is transmitted to the server device,
Using said second editing data, Oite the first scenario illustrated in the scenario data stored in the storage medium, before Symbol of said fifth portion and said second utterance of the first utterance by changing at least one of said sixth portion, edit support method before rewriting the alkoxy Nariodeta. An editing support system consisting of a server device and a first client device.
The server device
Using the speech data representing a speech by the user, in order to generate the speech data representing a speech responsive to the speech, a plurality of scenarios of interaction and a speech produced by the speech with the server device by a user shows The scenario data and the scenario data for the first editing are configured to be stored in the storage medium.
The first editing scenario data is
Among the plurality of scenarios indicated by the scenario data, a first scenario including one first utterance by the user and one second utterance generated by the server device, and
A first state transition for the action intended by the first scenario,
The first part of the first utterance and the second of the second utterances, which are editable by the first user using the first client device and do not change the first state transition. The first information showing the part and
With the second information indicating the third part in the first utterance and the fourth part in the second utterance, which is not editable by the first user and is a range for changing the first state transition. ,
The third information indicating the position where a new utterance can be added to the first scenario and the third information are shown .
The server device is configured to transmit the first editing scenario data to the first client device.
The first client device is
Receiving said scenario data for the first editing from the server device,
The first represented the editing scenario data, based on the first scenario and the first information and the second information and the third information, a first object representing the first utterance, the first A first object further indicating the first portion and the third portion, a second object indicating the second utterance, and a second object further indicating the second portion and the fourth portion, and the first object. Display a third object on the screen to add a new utterance to the scenario,
When an operation for editing at least one of the first portion and the second portion is performed by a user who uses the first client device, at least one of the edited first portion and the second portion is performed. Is configured to transmit the first edit data indicating the above to the server device.
The server device further
The first edit data is received from the first client device, and the first edit data is received.
Using the first editing data, Oite the first scenario illustrated in the scenario data stored before Symbol of the second portion of the first portion and of the second utterance of the first utterance by changing at least one, it is constructed prior to rewriting the carboxymethyl Nariodeta,
The first client device further
Wherein the first user, the third case of operation to add one third utterance using an object is performed, the third utterance is used in the second scenario Runa mules, the third Display information indicating that the utterance is not appropriate,
When the first user performs an operation for registering the first scenario to which the third utterance is added, additional data indicating the third utterance is configured to be transmitted to the server device.
The server device further
Upon receiving the additional data from the first client device,
The additional data using a stored said by adding the third utterance to the first scenario illustrated in the scenario data, before editing support system configured to rewrite the carboxymethyl Nariodeta.

Images