JP2017162268A - Dialog system and control program - Google Patents

Abstract

CONSTITUTION: A dialog system 10 is configured so that, according to a script which is preset to a dialog database 14, a robot R produces a speech, and one or two or more choices by which a person H responds to the speech production of the robot are displayed on a touch display 12 according to the script. When the person H selects the choice by a touch, a content of the selected response is uttered from a speaker provided in relation to the touch display 12. The choices displayed on the touch display 12 are set according to a phase of a dialog, such as collection of information, construction of relationship, persuasion for decision making, decision making, and idle talk having no purpose.EFFECT: Speech production is performed according to a script, and a person selects a response item, so that the person can continue natural dialog with a robot while feeling actual sensation that the person represented his/her intention.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an interactive system and a control program, and more particularly to a novel interactive system and control program in which an agent or person such as a robot interacts with at least one person through a touch display.

  With the recent development of speech recognition technology, for example, a dialogue system as disclosed in Patent Document 1 has been proposed. The system of Patent Document 1 is intended to realize continuous and natural interaction in a dialogue system between a human and a robot by trying to synchronize the two.

JP 2012-181697 A [G06F 3/16 ...]

  Even in the technique of Patent Document 1, there is a limit to processing based on voice recognition, and it is not easy to give a feeling of dialogue equivalent to that of a person and to appropriately transmit specific contents through dialogue.

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

  Another object of the present invention is to provide a dialog system and a control program capable of sustaining a sense of participating in a dialog to a person.

  Another object of the present invention is to provide an interactive system and a control program capable of transmitting appropriate contents to a question.

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

  The first invention displays a dialog database in which a script is set in advance, an utterance agent that utters according to the script of the dialog database, and one or more options that a person should respond to the utterance by the utterance agent according to the script. When a person selects an option displayed on a display, a dialog system includes an utterance unit that utters the contents of the option according to a script.

  In the first invention, the dialogue system (10: reference numerals exemplifying corresponding parts in the embodiment; the same applies hereinafter) includes a dialog database (14) in which scripts are preset. For example, an utterance agent such as a robot (R) utters a person (H), for example, according to a script of a dialog database by a control means such as a dialogue control manager (16). The person responds to this utterance, but the display means (16, 24) displays one or more options to be answered by the person on the display (12, 122) according to the script of the dialog database. Let When the person selects one of the options by looking at the display of the options, the utterance means (24, 26) utters the contents of the options according to the script.

  According to the first aspect, since the utterance of the utterance agent and the response of the person all follow the script, there is no failure of the dialogue. In addition, since a person selects an option by himself and the content is spoken by voice, the person can recognize that the option (response) reflects his intention even though the script is followed.

  The second invention is an interactive system according to the first invention, wherein the options displayed on the display means include two or more options having similar meanings.

In the second invention, in the dialogue phase of a person's final decision-making, for example, when a dialogue agent asks a person to purchase a product, in the final stage of the decision-making, for example, “I buy because I like”, “Yeah” It is possible to guide decision making by presenting only a reply indicating consent.

  According to the second invention, it is possible to induce human decision making.

  A third invention is an interactive system according to the first or second invention, wherein the number of options displayed on the display means is one.

  In the third invention, by limiting the number of response options to one in order to guide the person in the direction in which he / she wants to finally make a decision, the person is forcibly guided in the direction assumed by the dialogue system.

  According to the third invention, by limiting the number of options to one, it is possible to forcibly induce a person's response and guide the person's subsequent decision making in the direction assumed by the dialogue system.

  According to a fourth aspect of the present invention, there is provided a control program executed by a computer of an interactive system that performs dialogue according to a script preset in a dialog database, and the utterance agent that utters the computer according to the script of the dialog database. A display means for displaying one or more choices to be answered by a person according to the script, and a function that serves as an utterance means for uttering the contents of the choice according to the script when the person selects an option displayed on the display. Is a control program.

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

  According to the present invention, since the utterance of the utterance agent and the reply of the person all follow the script, it is possible to make the person recognize that there is no failure of the dialogue and the reply reflecting his / her intention is made.

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

FIG. 1 is a schematic diagram showing an outline of an interactive system according to an embodiment of the present invention. FIG. 2 is a block diagram showing an example of the configuration of the dialog control manager in the FIG. 1 embodiment. FIG. 3 is a schematic view showing an example of the robot in the embodiment of FIG. FIG. 4 is a block diagram showing an example of the configuration of a robot controller that controls the robot of FIG. FIG. 5 is a block diagram showing an example of the configuration of the touch display I / F in the FIG. 1 embodiment. FIG. 6 is a flowchart showing an example of the operation of the dialog control manager shown in FIG. FIG. 7 is a flowchart showing an example of the operation of the robot controller shown in FIG. FIG. 8 is a flowchart showing an example of the operation of the touch display I / F shown in FIG. FIG. 9 is an illustrative view illustrating options (response items) displayed on the touch display in the embodiment of FIG. 1. FIG. 10 is a schematic diagram showing an outline of a dialog system according to another embodiment of the present invention.

  Referring to FIG. 1, a robot R as a dialogue agent and one person H exist at a dialogue place of the dialogue system 10 of this embodiment. However, the number of robots R and the number of people H may be two or more. In the interactive system 10 of this embodiment, the robot R and the person H interact through a touch display (touch display) 12.

  In short, the dialog control manager 16 causes the robot R to speak a question about the person H according to a dialog or script (script) preset or prepared in the dialog database 14, and when the person H answers the touch display 12. Displays one or more response items that can be selected.

  The person H selects one of the response items on the touch display 12. In response, the touch display 12 utters the utterance corresponding to the selected response item as the response of the person H. The dialogue system 10 of the embodiment is a novel dialogue system that repeats such a dialogue.

  When a person speaks, if the other person's utterance is not generated in response to the utterance, it is difficult to continue the conversation itself, so a robot whose speech recognition and natural language processing capabilities are not perfect is human. It is not easy to continue to speak appropriately for the utterances, and the dialogue is likely to break down.

  Therefore, in this embodiment, when the robot speaks, it is not always necessary to make the person respond by speaking, but only by the touch display 12. By doing so, it is possible to avoid the failure of the dialogue due to insufficient robot capabilities. In addition, a response item suitable as a response to the utterer H of the robot R is displayed on the touch display 12, and by causing the person H to select it, the sensation that the person H is participating in the dialogue is maintained. You can do this by returning an appropriate response to the question from.

  The dialogue system 10 as in this embodiment can be used as a system such as an information collection service, an information provision service, an advertisement service, and a sales service.

  The dialog system 10 is a dialog (Dialog) database for causing the robot R to speak to the person H and displaying response items for the reply of the person H on the touch display 12 as described above. 14. Here, “dialog” means a sequence of utterances and non-verbal motion commands to be performed during a dialogue, and the dialogue database 14 is a collection of dialogues (for example, what a banana likes and what a robot care provides. It includes a series of commands for dialogue on each topic, such as dialogue about importance. “Script” is a character string representing a series of commands, and script data is a character string representing the commands. Therefore, the script data series is a script. In this embodiment, the script is described by State Chart XML (SCXML: w3c.org), and is all stored in the dialog database 14.

  As an example, in the script “T = XX, RH SPEAKER = RL = HUMAN TEXT = ……, NONVERVAL = ……”, “T = XX” is the time (or time) XX at which the script should be executed. “RH SPEAKER = RL = HUMAN” means that the robot R speaks to the person H, and “TEXT = ……” indicates a text sentence to be spoken. “NONVERVAL =...” Indicates a non-linguistic operation such as the operation of the robot R, watching the person H, whispering, shaking his / her head, or curling his / her neck.

  Such a script is transmitted from the dialog queue 18 to the robot controller 20 for controlling the robot R by the dialog control manager 16.

  The dialogue control manager 16 shown in FIG. 2 includes a CPU 16a, and a communication device 16c is connected to the CPU 16a via an internal bus 16b. The communication device 16c includes, for example, a network interface (NIC). The CPU 16a can communicate with the robot controller 20, the touch display I / F 24, and the like via the communication device 16c, and can exchange data between them. .

  Further, a memory 16d and an input device 16e are connected to the CPU 16a via an internal bus 16b. The memory 16d includes a ROM and a RAM. A script (dialog) is read from the dialog database 14 through the memory I / F 16f and temporarily stored in the memory 16d.

  Further, programs (such as OS and sensor signal acquisition program) necessary for the dialogue control manager 16 are stored in the memory 16d. The dialog control manager 16 operates according to a program stored in the memory 16d.

  The dialog queue 18 is also an area in the memory 16d, for example. In this dialog queue 18, the queue data is loaded in the next dialog candidate pool 22 so that the robot R can next execute the script data. Can be loaded in the form.

  Referring to FIG. 3, FIG. 3 shows an appearance of the robot R according to the embodiment, and the robot R is provided on the base 30 so as to be able to rotate back and forth and right and left with respect to the base 30. In other words, the body 32 has two degrees of freedom.

  From the left and right positions corresponding to the shoulders of the body 32 of the human body, the right arm 34R and the left arm 34L are provided so as to be rotatable forward and backward and left and right by shoulder joints (not shown). That is, two degrees of freedom are set for each of the right arm 34R and the left arm 34L.

  A neck 36 is provided at the center of the upper end of the body 32, and a head 38 is provided thereon. The neck 36, that is, the head 38, has three degrees of freedom with respect to the body 32 in terms of a roll angle (left-right tilt), pitches (front-back tilt), and yaw (left-right rotation).

  A right eye 40R and a left eye 40L are provided on the front surface of the head 38, that is, a surface corresponding to a human face, and eyeballs 42R and 42L are provided on the right eye 40R and the left eye 40L. The right eye 40R and the left eye 40L can close and open the eyelids, and the eyeballs 42R and 42L can rotate up, down, left and right, respectively. That is, one degree of freedom is set for the right eye 40R and the left eye 40L, that is, the eyelid, and two degrees of freedom are set for the eyeballs 42R and 42L.

  The face is further provided with a mouth 44, which can be closed or opened. That is, one degree of freedom is set for the mouth 44.

  At the position of the chest of the human body 32, a speaker 46 that makes a speech for the person H to hear in the dialogue system 10 and a microphone 48 that listens to the environment, particularly the voice of the person H, are provided.

  A camera 50 capable of shooting a moving image or a still image is built in a portion corresponding to the forehead of the head 38. The camera 50 can photograph the person H who is facing, and the camera signal (video signal) from the camera 50 is sent to the CPU 22a via the sensor I / F of the sensor manager 18 in the same manner as the environmental camera 16 (FIG. 1). , May be entered.

  FIG. 4 is a block diagram showing the robot controller 20 that controls the operation of the robot R (such as speech and non-language operation). Referring to FIG. 4, the robot controller 20 includes a CPU 20a, and a communication device 20c is connected to the CPU 20a via an internal bus 20b. The communication device 20c includes, for example, a network interface, etc., and the CPU 20a can communicate with the dialog control manager 16 and further with an external computer or another robot (both not shown) via the communication device 20c. Data can be exchanged.

  A memory 20d is further connected to the CPU 20a via an internal bus 20b. The memory 20d includes a ROM and a RAM. Control data and script data sent from the dialogue control manager 16 are temporarily stored in the memory 20d.

  In addition, programs (such as an OS and a sensor signal acquisition program) necessary for robot control are stored in the memory 20d. The robot controller 20 controls the operation of the robot R according to the program stored in the memory 20d.

In other words, the CPU 20a of the robot controller 20 is further connected to an output board 20e made of, for example, a DSP, and this output board 20e controls the operation of the actuators provided in the above-described units (FIG. 3) of the robot R. . However, since the specific operation of each part of the robot R is not related to the embodiment, detailed description is omitted here. Note that the speaker 46 of the robot R shown in FIG. 3 is connected to the output board 20e of the robot controller 20. The Therefore, the CPU 20a utters (speaks) from the speaker 46 according to the script data given from the dialogue control manager 16 and stored in the memory 20d as necessary.

  Further, the microphone 48 and the camera 50 of the robot R shown in FIG. 3 are input to the CPU 20a via the input board 20f. The CPU 20a sends the input data to the dialogue control manager 16. The dialogue control manager 16 stores the voice data captured from the microphone 48 in the memory 16d (FIG. 2), and executes voice recognition processing as necessary. The dialogue control manager 16 also processes the camera signal from the camera 50 to sense the situation of the dialogue place.

  Further, the robot R used in the dialogue system 10 of FIG. 1 embodiment is not limited to the robot described above with reference to FIG.

  FIG. 5 is a block diagram showing a touch display I / F 24 that controls display and speech of the touch display 14. The touch display I / F 24 includes a CPU 24a, and a communication device 24c is connected to the CPU 24a via an internal bus 24b. The communication device 24c includes, for example, a network interface. The CPU 24a can communicate with the dialogue control manager 16 and the like via the communication device 24c, and can exchange data between them.

  Further, a memory 24d is connected to the CPU 24a via an internal bus 24b. The memory 24d includes a ROM and a RAM. Control data and script data sent from the dialogue control manager 16 are temporarily stored in the memory 24d.

  A program (such as an OS or a sensor signal acquisition program) necessary for controlling the touch display 12 is stored in the memory 24d. The touch display I / F 24 controls the operation of the touch display 12 according to a program stored in the memory 24d.

  In other words, the CPU 24a of the touch display I / F 24 is further connected with an output board 24e configured by, for example, a DSP. The output board 24e outputs sound from the speaker 26 provided on the touch display 12 and touches the touch display I / F 24. Display data is output to the display 12.

  However, although the speaker 26 may be provided in the touch display 12, it may be provided in the vicinity of it separately from the touch display 12.

  Further, an input board 24f is connected to the CPU 24a of the touch display I / F 24. The input board 24f acquires coordinate data (touch coordinate data) from the coordinate detection circuit 28 provided in the touch display 12, and the CPU 24a. To enter. The CPU 24a sends the input coordinate data to the dialogue control manager 16. The dialog control manager 16 receives the coordinate data and can know which response item on the touch display 12 the person H touched at that time.

  In the dialogue system 10 of FIG. 1, the robot R at the dialogue place proceeds with the dialogue according to the script described above. The dialogue control manager 16 controls the utterance of the robot R and the utterance by the touch display 12 in an integrated manner. It is.

  The flowchart shown in FIG. 6 shows the operation of the CPU 16a (FIG. 2) of the dialog control manager 16. The operation of FIG. 6 is repeatedly executed at a speed of about the frame rate, for example.

  In the first step S1, the CPU 16a performs initialization such as reading the script data as described above from the dialog database 14 (FIG. 1).

  In subsequent step S3, the CPU 16a updates sensor data such as a camera signal and a microphone signal captured from the robot controller 20, for example.

  In the next step S5, based on the sensor data updated in step S3, the CPU 16a of the dialogue control manager 16 determines whether or not there is a person H at the dialogue place. If “NO” in the step S5, the process returns to the step S3 and waits.

  When “YES” is determined in the step S5, the CPU 16a reads the script data of the next candidate from the next candidate pool 22 in the next step S7. In step S9, it is determined whether there is a next candidate script.

  If there is a next candidate script, in step S11, the CPU 16a determines whether the script is a script for controlling the robot R. If “YES” in the step S11, the robot controller 20 executes a control operation of the robot R in a step S13. If “NO” in the step S11, it is further determined whether or not the next candidate script is for the touch display 12 in a step S15. If “YES” in the step S15, the touch display I / F 24 executes a control operation of the touch display 12 in a step S17.

  That is, from the information registered in the next candidate pool 22, the dialog control manager 16 determines whether to control the robot R or display on the touch display 12 according to the control rules of the dialog, and the robot controller 20 or the touch display I / A control signal is transmitted to F24. For example, when the robot R is to be controlled, in step S11, a control signal and reproduction voice information that are optimum for the content to be uttered are transmitted to the robot controller 20. When the touch display 12 is to be controlled, information on response items to be displayed on the touch display 12 and to be selected by the person H is transmitted.

  Referring to FIG. 7, the robot controller 20 performs initialization in step S31. The robot R controlled by the robot controller 20 has two types of operations: an idle operation and an operation according to a control signal. Execution of an idle operation means that even if an explicit control signal is not transmitted from the dialogue control manager 16, a basic operation such as blinking, breathing, looking at a nearby person is performed without stopping. Point to. At that time, independent of the operation flow of the robot controller 20, by acquiring information such as the presence / absence of the person H and the position of the person using sensors such as the microphone 48 and the camera 50 of the robot R, it is more social. An operation suitable for the operation is realized (step S33).

  Next, when the control signal is received from the dialogue control manager 16 while the idle operation is being executed, “YES” is determined in step S35, and in step S37, the CPU 20a of the robot controller 20 determines the control signal. The robot R is caused to execute the operation according to the above. Specifically, the utterance of the instructed dialogue (text sentence) and the operation according to the utterance are executed.

  When the operation of step S37 ends, the CPU 20a of the robot controller 20 sends a robot end flag (not shown) indicating the end of the operation of the robot R to the dialog control manager 16, and the control by the dialog control manager 16 of FIG. Transition to. Thereby, the dialogue control manager 16 shifts to a display task on the touch display 12. That is, the dialogue control manager 16 waits for the robot end flag from the robot controller 20 and shifts to control of the touch display 12.

  Referring to FIG. 8, the operation of touch display 12 is controlled by touch display I / F 24 shown in FIGS. The CPU 24a of the touch display I / F 24 performs initialization in step S41. Then, it is determined whether or not a control signal has been received from the dialog control manager 16 in step S43. When “YES” is determined in step S43, 1 or 2 to be displayed in that state in accordance with the transmitted control signal in step S45. The above options (response items) are displayed. However, in the embodiment, there are options for asking a question from the person H to the robot R, but such an option may also be referred to as a “response item” for convenience.

  Thereafter, when it is determined in step S47 that the user has touched any displayed response item, in step S49, the selected response item is read out by voice.

  When one is selected by one's own intention from among a plurality of options (response items) displayed on the screen of the touch display 12 as in this embodiment, a limited number of options set by the system are given. Despite being there, it is possible to make the person H feel that the selected response is his / her opinion. In addition, the robot R can interact without using voice recognition by having the person H select from the items displayed on the touch display 12.

  If the option (response item) selected by the person H is read (speaks) from the touch display 12 or the speaker 26 provided in the vicinity thereof, the person H feels as if he / she spoke the response. , The natural feeling of dialogue is not impaired.

  Here, as described above, which response item the person H has selected (touched) is the touch position detected by the coordinate detection circuit 28 (FIG. 5) provided in association with the touch display 12. Is determined by the CPU 24a. Then, this determination result (selected response item) is returned to the dialog control manager 16 for the progress of the subsequent dialog.

  Thereafter, after determining the end of the read-out voice (step S51), the process returns to FIG.

  Here, when the operation of step S49 is finished, the CPU 24a of the touch display I / F 24 sends a touch display end flag (not shown) that the operation on the touch display 12 is finished to the dialog control manager 16, and FIG. Transition to control of the robot R by the dialogue control manager 16 is made.

  In this way, the robot R starts the next utterance after the utterance voice of the response selected by the person H on the touch display 12 is finished, so that the robot R understands the voice from the touch display 12, that is, the person H. After that, you can feel as if the next utterance.

  Here, a process for determining an operation command transmitted from the dialog control manager 16 to the touch display I / F 24, that is, the touch display 12 will be described.

  First, a series of dialogues are classified into five phases: “information collection”, “relationship building”, “persuasion for decision making”, “decision making”, and “unintentional chat”. However, in the following, the “partner” refers to the robot R viewed from the person H.

  The “information collection” is intended to collect information such as the thoughts and preferences of the person H through daily conversation. Specifically, questions such as "Where did you come from today" or "What is your name?" The information collected here can be used in subsequent “persuades for decision making”.

  When collecting information, since it is necessary to accurately acquire information about the person H, four questions having different meanings are displayed on the touch display. At that time, the wording (dialog or script) is designed so that a matter that applies to the person H always matches one of the options. For example, for the question “Where did you come from?” As shown in Figure 9 (A), the options are “Is Osaka,” “Hyogo,” “Kyoto,” and “Somewhere in Japan.” Response item) is displayed on the touch display 12.

  However, the number of response items (options) may be four or more, but is four in the embodiment. The reason is that if there are five or more options, it takes time to understand the displayed options (contents), and the feeling of dialogue is hindered.

  Next, “building a relationship” is intended to deepen the relationship between the person H and the opponent (robot R). There are various ways to deepen the relationship with each other, but the most effective method is to make sure that each other has a good will. However, since the emotion is ambiguous, it is difficult to present a candidate that matches the emotion of the person H with the four options used in “collecting information”. Accordingly, in the “relationship building” phase, the touch display 12 automatically reads out a word (text sentence) prepared in advance in the dialog system 10, that is, the dialog database 14 at an appropriate timing. Specifically, in response to the question “(Robot R) looks very happy”, the response “(Human H) Yes, it ’s very fun” is automatically read out. In this way, recognition of feelings that I am happy now is performed. Conversely, the question “(Human H) is it fun to talk to me?” Is automatically read out, and “(Robot R) Yes, it is fun” is automatically answered to Robot R. By doing so, it is possible to build a situation in which each other is friendly. Emotion is an ambiguous definition, and it is difficult for people to understand clearly what emotions they have. Therefore, even if it is automatically read out, it is considered that it is easy to be recognized as the emotion that I have.

  Next, “persuasion toward decision making” refers to obtaining consent for a question to guide the person H in a direction in which he / she wants to finally make a decision. Specifically, the person H is forcibly guided in the direction assumed by the system side by limiting the number of options for replying to a question from a partner that affects subsequent decision making. In order for a person H to make a decision, a certain basis is necessary. For example, if the purpose of the dialogue between the robot R and the person H is to purchase a product, even if it is said that it should be suddenly purchased, it is difficult to make a decision to purchase. On the other hand, when it is revealed through conversation between each other that the color of the recommended product matches the preference of the person H, the conversation becomes the basis for the purchase.

  For example, if you answered “(Robot R) This color is perfect for your customers. Do you like your customers?” Is expected to influence subsequent decisions. In the case of such a question that affects the subsequent decision making, as shown in FIG. 9B, the answer of the person H is forcibly induced by limiting the number of options to one. This makes it possible to guide decision making in the direction assumed by the system thereafter.

  Next, “decision” refers to the final decision-making statement of the person H. Specifically, when urging the person H to make a final decision, it has the meaning of agreeing on the proposal from the robot R, but the response is guided by presenting multiple options with different expressions. Nevertheless, you can feel that you have made your own decisions.

  For example, “(Robot R) customer, do you want to buy this product?” As shown in Fig. 9 (C), both parties agree, “I want to buy because I like it”, “Yeah, I will” It is possible to guide decision making by presenting only the response indicating. Here, the reason why a plurality of synonymous responses are displayed is that by prompting the person H to decide to select one of the plurality of responses, responsibility for the selection is generated. Making a final decision is an action that requires more responsibilities and requires more consideration than the previous choices in the “persuade to decision” and “deepen relationships” phases. Therefore, methods such as “automatically read out” or “present only one option” make it difficult for the person H to give a sense of decision making. Therefore, in the “decision-making” phase, a method of selecting from a plurality of options having different synonymous expressions was adopted.

  Note that, as a condition necessary for a natural conversation between the robot R and the person H, there is a turn taking of the conversation. In other words, it is necessary to speak at an appropriate timing with respect to the other party's speech. For this purpose, it is necessary to appropriately inform the person H of the timing at which the person H responds on the touch display 12. In the embodiment, nothing is displayed on the touch display 12 when a response from the person H is not required, and options are displayed on the touch display 12 only when a response is required. At that time, a beep sound is generated to cause the person H to pay attention to the displayed option.

  Further, for example, by giving non-verbal information such as the direction of the line of sight of the robot R, that is, eye contact, the turn taking timing can be notified to the person H more effectively.

  Furthermore, since the person H must understand the contents of the options displayed on the touch display 12, it may take time to reply. However, depending on the content of the dialogue, there are cases where utterances are expected at socially correct timings such as “reply immediately” or “dare to say”.

  In such a case, the option is not displayed on the touch display 12, and the predetermined content is automatically read out at an appropriate timing. Thereby, natural conversation can be realized more effectively. As a problem in this case, there is a possibility of giving a feeling to the person H that it is not one's choice by being automatically read out, but since most of the conversation is chosen by oneself, when necessary If you just answer automatically, you won't think your intention has been hampered.

  The above-described embodiment is an interactive system using the robot R which is a physical agent. However, the present invention is not limited to such a physical agent, but for example, an avatar or character displayed on a display screen. A dialogue system between the agent and the person H as described above may also be applied. In this case, the robot controller 20 in FIG. 1 is replaced with a display I / F (not shown) for displaying such avatars and characters.

  Furthermore, the person H does not need to be an agent, and the person may also be a person. An embodiment in that case is shown in FIG.

  In this embodiment, it should be understood that the person H1 corresponds to the robot R of FIG. 1 embodiment, and the person H2 corresponds to the person H of FIG. 1 embodiment. That is, since the person H1 takes the place of the robot R, the dialogue control manager 16 uses the touch display 121 for the person H1 or a speaker (not shown: corresponding to the speaker 26 in FIG. 5) provided in the vicinity thereof. The utterance sentence of the robot R of one embodiment is uttered. In response to such an utterance from the touch display 121 of the person H1, the person H2 touches an option (response item) displayed on his / her touch display 122 in the same manner as the person H in the embodiment of FIG. select. Then, the selected item is uttered from the touch display 122 or a speaker (not shown: corresponding to the speaker 26 in FIG. 5) provided in the vicinity thereof.

  According to the embodiment of FIG. 10, in a dialogue between people, a person who has difficulty in speaking, for example, a person who is hesitant to speak words psychologically, or a person who has difficulty in speaking But you can talk. In addition, when the dialogue system reads out all the utterances on behalf of the person, it can feel as if the utterance is in line with the intention of the person.

  In the embodiment in FIG. 10, contrary to the above, the person H2 may correspond to the robot R in the embodiment in FIG. 1, and the person H1 may correspond to the person H in the embodiment in FIG.

  For example, the robot R in the embodiment of FIG. 1 utters an utterance sentence according to the script from its own speaker 46, and from the speaker provided in association with the touch display 121 or 122 in the embodiment of FIG. Since the uttered sentence is uttered, the dialog control manager 16, the robot controller 20, the touch display I / F 24, etc., which are uttered using the robot R or the touch display 121 (122), can be referred to as utterance agents.

  In the above-described embodiment, the response item (option) of the person H or the person H2 or the person H1 is displayed on the touch display and is selected by the touch. However, the display is not limited to the touch display, and other displays are used for display. And other pointing devices can be used for selection.

  Further, in FIG. 1, the dialogue control manager 16, the robot controller 20, and the touch display I / F 24 are illustrated as being separate from each other, but these may be realized by one computer.

10 ... Dialog system R ... Robot R
H, H1, H2 ... person 12, 121, 122 ... touch display 14 ... dialog database 16 ... dialog control manager 18 ... dialog queue 20 ... robot controller 22 ... next candidate pool 24 ... touch display I / F
26, 46 ... speaker 28 ... coordinate detection circuit

Claims (4)

Dialog database with preset scripts,
An utterance agent that utters according to the dialog database script;
Display means for displaying one or more options to be answered by the person in response to the utterance by the utterance agent according to the script, and when the person selects the option displayed on the display means, the content of the option is scripted A dialogue system comprising utterance means for uttering according to the above.   The dialog system according to claim 1, wherein the options displayed on the display means include two or more options having similar meanings.   The interactive system according to claim 1, wherein the number of options displayed on the display unit is one. A control program that is executed by a computer of an interactive system that interacts according to a script that is preset in a dialog database, the utterance agent that utters the computer according to the script of the dialog database,
Display means for displaying one or more options to be answered by the person in response to the utterance by the utterance agent according to the script, and when the person selects the option displayed on the display, the content of the option is determined according to the script. A control program that functions as a means of speaking.

Images