JP6449368B2 - Conversation providing apparatus, conversation providing method, and program - Google Patents

Description

  The present invention relates to a conversation providing device, a conversation providing method, and a program.

  Conventionally, various chatbots and software agents have been proposed that perform dialogues such as providing information and chatting with users.

  For example, a dialogue program that controls insertion behavior of a dynamic dialogue node between fixed dialogue nodes has been proposed (for example, Patent Document 1).

  Further, when a conversation between pseudo-persons is performed, the role of the plurality of pseudo-persons in the conversation is changed according to the progress of the conversation related to the topic information extracted by one of the pseudo-persons, and the pseudo-persons before and after the change are changed. A technology that allows the conversation between people to be shown to the user, so that the role of the personality is changed according to the progress of the conversation, and a conversation space rich in change seen from the user is generated in the system. It has been proposed (for example, Patent Document 2).

  In addition, a technology has been proposed that realizes an agent interface device that maintains an internal state composed of the emotions and desires of an agent and can respond in consideration of the user's emotion based on the user's emotion and the agent's emotion. (For example, Patent Document 3).

JP2015-125198A JP 2003-323388 A JP 2005-222331 A

  Conventionally, various techniques for increasing variations in conversation have been proposed. However, when chatting on a certain theme, there is a problem that it is easy to fall into a dialogue failure such as repeating the same speech or making the context unnatural when switching topics.

  Therefore, an object of the present invention is to improve the continuity of conversation and the naturalness of topic change in a system for conversation with a user.

  The conversation providing apparatus according to the present invention allows a virtual personality and a user to have a conversation. In addition, a scenario storage unit that stores a plurality of conversational partial scenarios including dialogues of a plurality of speakers, and a conversation control unit that outputs virtual personality statements to the user using the partial scenarios, The control unit extracts a response to the virtual personality statement or the user's speech from the partial scenario based on the similarity between the virtual personality statement or the user's speech and the dialogue included in the partial scenario. .

By extracting a response from a partial scenario prepared in advance based on the similarity of the dialogue, it is possible to ensure the naturalness of the content of the conversation and the conversion of the topic. Also, unlike the case where a statement is generated based on the contents of a past topic, it is possible to suppress the occurrence of a breakdown of a dialogue that prevents the user from getting out of any topic by connecting partial scenarios. That is, the continuity of conversation and the naturalness of topic change can be improved.

  In addition, the conversation providing device causes the user to converse with the virtual personality in a format that allows the user to select his / her speech from the options, and the conversation control unit is included in the first partial scenario read from the scenario storage unit. And the dialogue included in the second partial scenario similar to the first partial scenario is presented to the user as an option and the user selects either of them, and the partial scenario including the dialogue selected by the user In the above, a dialogue described as a response to the selected dialogue may be output as a utterance of a virtual personality. In this way, the plurality of options are in line with the ongoing conversation flow, and the naturalness of the conversation flow can be improved.

  Further, the conversation control unit outputs the first speech of the first speaker included in the first partial scenario read from the scenario storage unit as an utterance of a virtual personality, and the first partial scenario To the first speech of the second speaker from the scenario storage unit based on the words included in the first speech of the second speaker described as a response to the first speech of the first speaker in FIG. Extracting the second line of the second speaker included in another similar partial scenario and including the first line of the second speaker and the second line of the second speaker May be presented to the user and one of them may be selected. Specifically, with such a configuration, the plurality of options are in line with the ongoing conversation flow.

  In addition, the conversation control unit performs morphological analysis on the first speech of the second speaker and the speech included in other partial scenarios stored in the scenario storage unit, and the words included in each speech Based on the distributed representation, a second line of the second speaker included in another partial scenario similar to the first line of the second speaker in the first partial scenario is extracted. Also good. In this way, it is possible to calculate a semantic distance between words and dialogues including the words and compare them.

  The second dialogue of the second speaker may be extracted from other than the last dialogue in other partial scenarios. Since the conversation providing apparatus according to the present invention uses the last dialogue in another partial scenario as a virtual personality response to the response selected by the user in the subsequent processing, as described above, It is preferable to extract from other than the last line in the scenario.

  In addition, the conversation providing device allows the user to freely input a speech, and allows the user to have a conversation with a virtual personality, and the conversation control unit describes a speech similar to the speech input by the user and the speech. A partial scenario may be extracted, and in the extracted partial scenario, a dialogue described as a response to the extracted dialogue may be output as a virtual personality response to the utterance input by the user. In this way, it is possible to guarantee the naturalness of dialogue content and topic conversion to some extent by extracting responses based on the similarity from partial scenarios prepared in advance even for remarks that are freely input by the user. can do.

  In addition, the conversation control unit determines the similarity between the combination of the virtual personality speech or the first speech of the user and the second speech responding to the speech and the combination of the two dialogues included in the partial scenario. Based on this, a response to the second utterance may be extracted from the partial scenario. By evaluating the degree of similarity for a combination of two statements, it can be determined whether or not they are similar in a larger context, and a more appropriate response can be extracted.

The partial scenario may be described in advance so that the story line is consistent. By preparing such a partial scenario in advance, it is possible to improve the continuity of conversation and the naturalness of topic change.

  The contents described in the means for solving the problems can be combined as much as possible without departing from the problems and technical ideas of the present invention. The contents of the means for solving the problems can be provided as a device such as a computer or a system including a plurality of devices, a method executed by the computer, or a program executed by the computer. A recording medium that holds the program may be provided.

  In a system for conversation with a user, the continuity of conversation and the naturalness of topic change can be improved.

It is a figure which shows an example of a structure of the whole system. It is a block diagram which shows an example of a structure of a user terminal and a conversation provision server. It is a figure which shows an example of a user interface. It is a figure which shows an example of the scenario previously memorize | stored in the memory | storage device of the conversation provision server. It is a processing flowchart which shows an example of a conversation provision process. It is a processing flowchart which shows an example of a similar scenario extraction process. FIG. 6 is a diagram illustrating an example of a partial scenario in progress and other partial scenarios stored in a storage device.

  Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. In addition, the structure of embodiment is an illustration and this invention is not limited to the structure shown in embodiment.

<System configuration>
FIG. 1 is a diagram illustrating an example of the entire system according to the embodiment. This system is a virtual personality (corresponding to a “second speaker” according to the present invention) in which a speech is output by a computer and has a conversation with a user (corresponding to the “first speaker” according to the present invention). Provide). Further, the content of the conversation includes a chat, and for example, the conversation is continued until the user stops using the system. This system includes a user terminal 1 and a conversation providing server 2, which are connected via a network 3. The network 3 is a communication network such as the Internet, and computers connected to the network 3 can communicate with each other based on various protocols. The user terminal 1 is a computer such as a smartphone, a tablet, or a PC (Personal Computer), and the user of the conversation providing system according to the present embodiment by executing a program (also referred to as software or application) according to the present embodiment. Functions as an interface. Note that a plurality of user terminals 1 may be connected to the network 3. The conversation providing server 2 is a general computer, and by executing a program according to the present embodiment, a speech of a virtual personality is output to the user terminal 1 via the network 3, and the user terminal 1 sends the user to the user. Repeat the process to get the response. Note that the user's remarks are input to the user terminal 1 and transmitted to the conversation providing server 2 via the network 3 when the user selects from the options output by the conversation providing server 2.

<Device configuration>
FIG. 2 is a block diagram illustrating an example of the configuration of the user terminal 1 and the conversation providing server 2.

The user terminal 1 is a general computer, and includes an input / output interface (I / F) 11, a storage device 12, a communication interface (I / F) 13, a processor 14, and a bus 15. The input / output I / F 11 is a user interface such as a touch panel. The input / output I / F 11 displays an utterance of a virtual personality to the user, displays user response options, and accepts a user response selection based on a user operation. The storage device 12 includes a main storage device such as a RAM (Random Access Memory) and a ROM (Read Only Memory), and an auxiliary storage device (secondary storage) such as an HDD (Hard-disk Drive), an SSD (Solid State Drive), and a flash memory. Device). The main storage device temporarily stores programs read by the processor and received dialogue data, and secures a work area for the processor. The auxiliary storage device stores a program executed by the processor. The communication I / F 13 is a network module that performs communication with, for example, a mobile phone base station or a wireless LAN (Local Area Network) access point, and performs communication based on a predetermined protocol. The processor 14 is an arithmetic processing unit such as a CPU (Central Processing Unit), and performs each process according to the present embodiment by executing a program. In the example of FIG. 2, functional blocks are shown in the processor 14. Specifically, the processor 14 functions as the input / output control unit 141. The input / output control unit 141 displays an utterance of a virtual personality on the input / output I / F 11 and accepts a user response input via the input / output I / F 11. The components as described above are connected via the bus 15.

  The conversation providing server 2 is also a general computer and includes an input / output I / F 21, a storage device 22, a communication I / F 23, a processor 24, and a bus 25. The input / output I / F 21 is a user interface such as a keyboard, a mouse, or a display. The storage device 22 is a main storage device such as a RAM or a ROM, and an auxiliary storage device such as an HDD, an SSD, or a flash memory. The main storage device temporarily stores programs read by the processor and scenario data stored in the auxiliary storage device, and secures a work area for the processor. The auxiliary storage device stores a program executed by the processor. The communication I / F 23 is a wired network card, for example, and performs communication based on a predetermined protocol. The processor 24 is an arithmetic processing unit such as a CPU, and performs each process according to the present embodiment by executing a program. As shown in FIG. 2, the processor 24 functions as a conversation control unit 241. The conversation control unit 241 outputs an utterance of a virtual personality for the user and an option of the user's response to the utterance using a scenario prepared in advance. The components as described above are connected via the bus 15.

<User interface>
FIG. 3 is a diagram illustrating an example of an image displayed on the input / output I / F 11 of the user terminal 1. In the example of FIG. 3, an input / output I / F 11 that is a touch panel includes an image 111 representing a virtual personality, an image 112 representing a user, a balloon 113 representing a speech of a virtual personality, and user response options. Three balloons 114 (114a to 114c) are displayed. The user can have a conversation with a virtual personality by, for example, tapping and selecting his / her speech from such options. The number of options is plural and is not limited to three.

<Scenario (partial scenario)>
FIG. 4 is a diagram illustrating an example of a scenario stored in advance in the storage device 22 of the conversation providing server 2. In the present embodiment, it is assumed that a plurality of scenarios including a plurality of dialogue-style dialogues as shown in FIG. 4 are stored in advance in the storage device 22 of the conversation providing server 2. In addition, lines are described in each line included in the scenario, and alphabets A and B added to the beginning of each line represent a person who emits the lines described in each line. For convenience, continuous lines assigned to the same person will be referred to as a single statement. In the example of FIG. 4, the utterance 221 (first utterance) of the person B, the utterance 222 (second utterance) of the person A responding thereto, and the utterance 223 (3 (3) of the person B responding further) The first remark) is described. One statement includes one or more lines. One line may be one sentence, and a line indicating some intention is described in one line. In addition, the topic may be changed between the lines, but the description is made so that the story line is consistent. As shown in FIG. 4, an exchange including about 3 to 4 statements with context is described as one unit, and this is also referred to as a “partial scenario”. In the present embodiment, the conversation is continuously performed by connecting at least a part of the partial scenario to the previous conversation between the virtual personality and the user.

<Conversation processing>
FIG. 5 is a process flow diagram showing an example of the conversation providing process executed in the present system. First, the conversation control unit 241 of the conversation providing server 2 selects an initial partial scenario based on some condition (FIG. 5: S1). In this step, any partial scenario for starting a conversation is selected. For example, it is assumed that the partial scenario shown in FIG. 4 is selected from a plurality of partial scenarios stored in the storage device 22.

  And conversation control part 241 outputs one remark of virtual personality (Drawing 5: S2). In this step, for example, the utterance 221 described at the beginning of the partial scenario is transmitted to the user terminal 1 as an utterance of a virtual personality. In the example of FIG. 4, the lines of person B, “I'll be disliked because of rain recently” and “I think it's about to enter the rainy season” are transmitted to the user terminal 1 via the communication I / F 23.

  On the other hand, when the input / output control unit 141 of the user terminal 1 receives an utterance of a virtual personality via the communication I / F 13, the input / output I / F 11 outputs the received utterance (FIG. 5: S3). In this step, the utterances of the virtual personality “I'll be disliked because of rain recently” and “I think it's about to enter the rainy season” are displayed in the balloon 113 of FIG. 3 in order, for example.

  Further, the conversation control unit 241 of the conversation providing server 2 performs a similar scenario extraction process (FIG. 5: S4).

  FIG. 6 is a process flow diagram illustrating an example of a similar scenario extraction process. The conversation control unit 241 of the conversation providing server 2 performs morphological analysis on the ongoing partial scenario (FIG. 6: S11). In this step, for example, in the scenario 1 in which the dialogue is first output to the user terminal 1, the dialogue “Sone, here is the rain continuation” described as the utterance in response to the utterance of the virtual personality output to the user terminal 1. For “Ne”, morphological analysis is performed and it is broken down into parts of speech.

  Moreover, the conversation control part 241 performs a morphological analysis about the other partial scenario memorize | stored in the memory | storage device 22 of the conversation provision server 2 (FIG. 6: S12). In this step, morphological analysis is performed on speech included in partial scenarios other than the ongoing partial scenario stored in the storage device 22 and decomposed into parts of speech.

  Then, the conversation control unit 241 performs a partial scenario similar to the ongoing partial scenario (corresponding to the “first partial scenario” in the present invention) based on the feature vector (“second part in the present invention”). (Corresponding to “scenario scenario”) (FIG. 6: S13). In this step, text mining is performed using words included in each dialogue, and similar partial scenarios and similar dialogue are extracted. For example, for an existing partial scenario, machine learning using a neural network such as deep learning is used to estimate words that are likely to appear in the surrounding context, or other words that appear in the context from a certain word. A prediction model for estimation is generated in advance. Various existing technologies can be used for text mining, and synonyms and usage forms may be treated as the same word.

For example, FastText, Word2Vec, CBOW (Continuous Bag-of-Words), Skip-gram W
Techniques such as MD (Word Mover's Distance) and cosine similarity can be used. When using WMD, for example, words having similar meanings included in two dialogues are associated with each other, the distance of the distributed expression is calculated as a cost for each association, and the total cost between the two dialogues is calculated between the two dialogues. Distance. And it is judged that the similarity is higher as the distance is closer. When using cosine similarity, generate a distributed representation of the words contained in each dialogue and cosine with the words contained in the dialogue described as utterances in response to the utterances of the virtual personality in the ongoing partial scenario. Extract lines containing words with high similarity. Further, the similarity of words may be calculated using nouns in particular. Using text mining technology as described above, in the ongoing scenario, the sentences described as utterances responding to the utterances of the virtual personality “sore, here ’s the rain lasts”. Extract lines of numbers.

  At this time, an utterance of a virtual personality that is the previous utterance may also be used for calculating the similarity. That is, with respect to an exchange consisting of two lines of a certain utterance (virtual personality utterance) and a response to this, an exchange consisting of two similar lines is extracted from a partial scenario using WMD, for example. . In this way, since a subsequent utterance can be determined based on a similar partial scenario in a longer context, a more natural response can be extracted.

  In this step, for example, two lines are extracted as options.

  FIG. 7 is a diagram illustrating an example of an ongoing partial scenario and other partial scenarios stored in the storage device 22. FIG. 7 illustrates scenarios 1-5, and scenario 1 is the ongoing scenario also shown in FIG. Scenarios 2 to 5 are partial scenarios related to some topic. Note that scenario 2 includes a fourth statement 224 and a fifth statement 225, and scenario 4 includes a fourth statement 224.

  In the ongoing scenario, “Sone, here's the last rain” in the scenario 1, which is a dialogue described as a statement in response to a statement of a virtual personality (FIG. 7: solid rounded rectangle) is S11 in FIG. Is broken down into parts of speech. Dialogues included in scenarios 2 to 5 are decomposed into parts of speech in S12 of FIG. Further, it is assumed that the dialogue “Since it's raining from noon” in scenario 2 and the dialogue “It's cloudy, is it raining?” In scenario 3 are extracted as similar dialogues. The dialogue described in response to the utterance of the virtual personality in the ongoing scenario and a predetermined number of dialogues that are judged to be similar to the dialogue (FIG. 7: dotted rounded rectangle) are user response options. Extracted as It should be noted that similar lines are not extracted from scenarios 4 and 5.

  Thereafter, the similar scenario extraction process is terminated, and the process returns to the process of FIG. Of the processes shown in FIG. 6, morphological analysis (S11, S12) and calculation of similarity between dialogues of similarity are performed for all dialogues included in the partial scenario stored in the storage device 22. The combination is executed in advance, and in the similar scenario extraction process, a process described as a comment that responds to a comment of a virtual personality in a scenario in progress and a process that extracts a high similarity between them You may make it perform only.

  The conversation control unit 241 outputs the user's response options (FIG. 5: S5). In this step, the conversation control unit 241 transmits the dialogue extracted in S13 of FIG. 6 to the user terminal 1 via the communication I / F 23.

On the other hand, when the input / output control unit 141 of the user terminal 1 receives the user's response options via the communication I / F 13, the input / output I / F 11 displays the received options (FIG. 5: S6). In this step, scenario 1 shown in FIG.
The dialogue “It's raining from noon” and the scenario 3 “It ’s cloudy, is it raining?” Are displayed in the balloons 114 a to 114 c in FIG. 3.

  The input / output control unit 141 receives a user operation via the input / output I / F 11, receives an input for selecting a user response, and displays the selected response on the input / output I / F 11 as a user's message. (FIG. 5: S7). In this step, when the user taps any of the balloons 114a to 114c in FIG. 3, the dialogue of the tapped balloon is displayed as the user's response. In addition, the input / output control unit 141 transmits information indicating which response has been selected to the conversation providing server 2 via the communication I / F 13. In this case, it is assumed that the user selects the dialogue of scenario 3 “It's cloudy, is it raining?” From the options.

  Thereafter, the conversation control unit 241 of the conversation providing server 2 determines a further response of the virtual personality according to the response selected by the user (FIG. 5: S8). In this step, in the partial scenario including the dialogue selected by the user, the dialogue described as a response to the dialogue is extracted as an utterance of a virtual personality. In the example of FIG. 7, in the scenario 3, the lines of person B's lines “It seems to be so” and “But it seems to be fine tonight” “Do n’t think” (Figure 7: Rounded rectangle with dashed line) is decided to be a statement of virtual personality.

And it returns to S2 of FIG. 5 and repeats a process. That is, the conversation control unit 241 performs S8.
The utterance of the virtual personality determined in is transmitted to the user terminal 1 via the communication I / F 23.

<Effect>
According to the present embodiment, partial scenarios describing contextual exchanges are connected, and a virtual personality statement and user response options are output based on the plurality of connected partial scenarios. Also, the user's response options are dialogues that are evaluated as having a high degree of similarity using so-called artificial intelligence, and the story path is consistent regardless of which is selected. Further, the further virtual personality response to the selected user response is a dialogue according to a scenario described in advance as a contextual interaction in the partial scenario, and the story line is made consistent. Therefore, even if the topic is changed, it is done depending on whether the topic naturally changes in a partial scenario prepared in advance or whether the user changes the topic by selecting an option. It will be natural. Moreover, since the conversation can be continued by connecting the partial scenarios, the continuity of the conversation is also improved.

<Modification>
As shown in FIG. 7, the person A and the person B appearing in the partial scenario may be assigned to a user and a virtual personality, respectively. That is, in S13 of FIG. 6, in the partial scenario, the user's response options are extracted from the dialogue of the person A that is the user's speech. In this way, in the partial scenario prepared in advance, the user and virtual personality can be given some personality and the content and tone of the dialogue can be described.

  On the other hand, the user's response options may be extracted without distinguishing between the person A and the person B appearing in the partial scenario. Even in such a mode, the continuity of conversation and the naturalness of topic change can be improved by connecting partial scenarios prepared in advance.

In addition, the extraction of user response options is not limited to the first statement in the partial scenario,
The remarks described after the second in the partial scenario may be targeted. In the partial scenario, since the utterance described as the response is used as the utterance of the virtual personality after the utterance related to the user's response, the last utterance in the partial scenario is extracted as an option of the user's response Not subject to. That is, the user's response options in S13 are extracted from statements other than the last one described in the partial scenario.

  In addition, the user's response choices are extracted based on past statements in the ongoing scenario, not the ones described as responses to the utterances of the virtual personality in the ongoing scenario. You may make it do.

  Further, the user may be allowed to freely input the content of the utterance instead of the user interface in the form of allowing the user to select his / her utterance from the options. In this case, the speech similar to the speech entered by the user is extracted from the partial scenario, and the speech described as a response to the similar speech extracted in the partial scenario is the next speech of the virtual personality. As output.

  Also, for the combination of the speech freely input by the user and the speech of the virtual personality before that, a combination of similar dialogues is extracted from the partial scenario, and for the dialogue combination extracted in the partial scenario Further, the dialogue described in the form of response may be output as the next statement of the virtual personality.

<Others>
The present invention includes a computer program that executes the above-described processing. Furthermore, a computer-readable recording medium on which the program is recorded also belongs to the category of the present invention. With respect to the recording medium on which the program is recorded, the above-described processing can be performed by causing the computer to read and execute the program on the recording medium.

  Here, the computer-readable recording medium refers to a recording medium in which information such as data and programs is accumulated by electrical, magnetic, optical, mechanical, or chemical action and can be read from a computer. Examples of such a recording medium that can be removed from the computer include a flexible disk, a magneto-optical disk, an optical disk, a magnetic tape, and a memory card. In addition, examples of the recording medium fixed to the computer include a hard disk drive and a ROM.

1: User terminal 11: Input / output I / F
12: Storage device 13: Communication I / F
14: Processor 141: Input / output control unit 15: Bus 2: Conversation providing server 21: Input / output I / F
22: Storage device 23: Communication I / F
24: Processor 241: Conversation control unit 25: Bus 3: Network

Claims (7)

A conversation providing device for conversing a virtual personality and a user in a form that allows the user to select his / her speech from options ,
A scenario storage unit for storing a plurality of conversational partial scenarios including dialogues of a plurality of speakers;
A conversation control unit that outputs the utterance of the virtual personality for the user using the partial scenario;
With
The conversation control unit
Based on the similarity between the utterance of the virtual personality or the utterance of the user and the dialogue included in the partial scenario, the response to the utterance of the virtual personality or the utterance of the user from the partial scenario. Extract and
The dialogue included in the first partial scenario read from the scenario storage unit and the dialogue included in the second partial scenario similar to the first partial scenario are presented to the user as the options. And select one with
A conversation providing apparatus for outputting a dialogue described as a response to a selected dialogue as a statement of the virtual personality in a partial scenario including the dialogue selected by the user . The conversation control unit
Outputting the first speech of the first speaker included in the first partial scenario read out from the scenario storage unit as an utterance of the virtual personality;
Based on the words included in the first speech of the second speaker described as a response to the first speech of the first speaker in the first partial scenario, the first from the scenario storage unit Extracting a second line of the second speaker included in another partial scenario similar to the first line of the two speakers;
The conversation providing apparatus according to claim 1 , wherein an option including a first line of the second speaker and a second line of the second speaker is presented to the user and one of the options is selected. The conversation control unit
A morphological analysis is performed on the first speech of the second speaker and speech included in another partial scenario stored in the scenario storage unit, and based on a distributed representation of words included in each speech , according to claim 2 for extracting a second speech of the second speaker to be included in other partial scenario similar to the first words of the second speaker in the first partial scenario Conversation providing device. The conversation providing apparatus according to claim 3 , wherein the second dialogue of the second speaker is extracted from other than the last dialogue in the other partial scenario. The conversation providing device according to any one of claims 1 to 4 , wherein the partial scenario is described in advance so that a story line is consistent. A computer-aided conversation providing method in which a virtual personality and a user are conversed in a form that allows the user to select his / her speech from options ,
Using the partial scenario stored in a scenario storage unit that stores a plurality of conversational partial scenarios including dialogues of a plurality of speakers, the virtual personality speech or the user speech, and the partial Extracting from the partial scenario a response to the utterance of the virtual personality or the utterance of the user based on the similarity to the dialogue included in the scenario;
The dialogue included in the first partial scenario and the dialogue contained in the second partial scenario similar to the first partial scenario, read from the scenario storage unit, are presented to the user as the options. And selecting one of them, and
Outputting in a partial scenario including the dialogue selected by the user a speech described as a response to the selected dialogue as an utterance of the virtual personality;
Conversation providing method including A virtual personality and users in the form to select their own speech to the user from the options, a program for executing processing for conversation computer,
Using the partial scenario stored in a scenario storage unit that stores a plurality of conversational partial scenarios including dialogues of a plurality of speakers, the virtual personality speech or the user speech, and the partial Extracting from the partial scenario a response to the utterance of the virtual personality or the utterance of the user based on the similarity to the dialogue included in the scenario;
The dialogue included in the first partial scenario and the dialogue contained in the second partial scenario similar to the first partial scenario, read from the scenario storage unit, are presented to the user as the options. And selecting one of them, and
Outputting in a partial scenario including the dialogue selected by the user a speech described as a response to the selected dialogue as an utterance of the virtual personality;
A program that executes

Images