JP2020106746A - Control device, control method, control program, and interactive device - Google Patents

Abstract

To provide an interactive device which can respond to a user at an appropriate timing.SOLUTION: A control device (20) which controls an interactive device (1) includes: an utterance appropriate value management unit (23) which updates an utterance appropriate value indicating the appropriateness of the uttering of the interactive device (1), based on the content of an operation performed by a user to the interactive device (1), the content of the utterance made by the user to the interactive device (1), or the content of a response made by the interactive device (1) to the user; and an utterance appropriateness determination unit (22b) which determines whether or not it is appropriate for the interactive device (1) to speak to the user according to the utterance appropriate value updated by the utterance appropriate value management unit (23).SELECTED DRAWING: Figure 1

Description

The present invention relates to a control device, a control method, and a control program for controlling a dialogue device that interacts with a user. The present invention also relates to a dialogue device for dialogue with a user.

In recent years, interactive devices (smartphones, robots, etc.) that interact with users by performing voice recognition and language processing have been actively developed. For example, for each of a plurality of users who have used the robot in the past, the operation history of the user and the history of the dialogue between the user and the robot are stored in a storage unit together with the emotion value of the robot. Has been proposed (see Patent Document 1).

In the above robot, a person identifying means (for example, a face image recognizing means, a voice recognizing means) for identifying a user encountered from a plurality of users is provided. The emotion value of, the operation history, the dialogue history, and the emotion value stored in the storage unit for the user are referenced to generate a new emotion value of the robot. This enables the robot to take different actions for each user.

JP-A-2003-117866 (Published April 23, 2003)

However, in the robot described in Patent Document 1, the timing at which the robot talks to the user is not optimized. Therefore, there is a problem that the robot may not be able to speak at an appropriate timing for the user.

The present invention has been made in view of the above-mentioned problems, and an object thereof is to provide a control device, a control method, and a control program for realizing a dialogue device capable of speaking at an appropriate timing for a user. To do.

In order to solve the above problems, a control device according to an aspect of the present invention is a control device that controls a dialogue device that interacts with a user, and the utterance appropriateness that indicates the appropriateness of speaking by the dialogue device. The value is the content of the operation performed by the user on the dialog device, the content of the utterance made by the user to the dialog device, or the response made by the dialog device to the user. The utterance appropriate value management unit that updates based on the content of the utterance appropriate value, and whether or not it is appropriate for the dialogue device to utter the user, the utterance appropriate value updated by the utterance appropriate value management unit. An utterance adequacy determining unit that determines according to the utterance.

In order to solve the above problems, a control method according to an aspect of the present invention is a control method for controlling a dialogue device that interacts with a user, and the utterance appropriateness indicates the appropriateness of speaking by the dialogue device. The value is the content of the operation performed by the user on the dialog device, the content of the utterance made by the user to the dialog device, or the response made by the dialog device to the user. Utterance appropriate value update processing for updating based on the content of the, and whether or not it is appropriate for the dialog device to speak to the user, the utterance appropriate value updated in the utterance appropriate value update processing. Utterance suitability determination processing for determining according to.

According to one embodiment of the present invention, a user can speak at an appropriate timing.

1 is a block diagram showing a schematic configuration of a dialogue device according to a first embodiment of the present invention. FIG. 6 is a diagram showing initial values of appropriate utterance values in the dialogue device according to the first embodiment. (A) is a figure which shows the relationship between the user's operation content and the amount of change of a utterance appropriate value in the dialogue apparatus concerning Embodiment 1, (b) is a user's utterance content and the change of an utterance appropriate value. It is a figure which shows the relationship with a quantity, and (c) is a figure which shows the relationship between the response content of a dialog device and the amount of change of an appropriate utterance value. 5 is a graph showing an example of temporal changes in utterance appropriate values in the dialogue device according to the first embodiment. 5 is a flowchart showing a flow of speech processing in the dialogue device according to the first embodiment. It is a flowchart which shows the flow of the utterance process in the dialog apparatus which concerns on Embodiment 2 of this invention.

[Embodiment 1]
Hereinafter, the dialogue device 1 according to the first embodiment of the present invention will be described with reference to FIGS. 1 to 5. For convenience of description, members having the same function will be denoted by the same reference numeral, and description thereof will be omitted as appropriate.

<Structure of dialogue device>
The configuration of the dialogue device 1 will be described with reference to FIG. FIG. 1 is a block diagram showing a schematic configuration of the dialogue device 1 according to the first embodiment. The interactive device 1 is a device (electronic device) that interacts with a user, and may be realized as a non-humanoid device such as a smartphone or a humanoid device such as a robot.

As shown in FIG. 1, the dialogue device 1 includes an input unit 10, a control unit 20 (an example of a “control device” in the claims), a storage unit 30, and an output unit 40. The dialogue device 1 may further include a communication unit (not shown) that communicates with the outside through a communication network. As the communication network, for example, the Internet, a telephone line network, a mobile communication network, a CATV (Cable Television) communication network, a satellite communication network, or the like can be used.

(Input section)
The input unit 10 is a device for inputting information to the dialogue device 1. The input unit 10 can be configured by a microphone 11, a camera 12, a touch panel 13, and a sensor 14, as shown in FIG. 1, for example.

The microphone 11 acquires a voice around the dialogue apparatus 1. The microphone 11 provides the control unit 20 with audio data representing the acquired audio. For example, when the user who is speaking is around the dialog device 1, the voice represented by the voice data provided to the control unit 20 includes the utterance sound of the user.

The camera 12 acquires an image of the surroundings of the interactive device 1. The camera 12 provides the control unit 20 with image data representing the acquired image. For example, when the user is around the dialogue device 1, the image of the image data provided to the control unit 20 includes the image of the user.

The touch panel 13 receives a touch operation by the user. The touch panel 13 provides the control unit 20 with operation information indicating the received touch operation. Examples of operations that the touch panel 13 can receive include touch operations for instructing activation of an application, release of sleep, or interruption of speech. Note that, instead of the touch panel 13, for example, a push button type operation device (for example, a physical button) may be used.

The sensor 14 detects the state of the dialogue device 1 or the state around the dialogue device 1. The sensor 14 provides the control unit 20 with state information indicating the detected state. Examples of the state of the dialog device 1 include vibration of the dialog device 1 or acceleration of the dialog device 1. In this case, for example, a vibration sensor or an acceleration sensor is used as the sensor 14. The state around the dialogue device 1 may be, for example, the presence or absence of a user around the dialogue device 1. In this case, for example, a motion sensor or a proximity sensor is used as the sensor 14.

(Control unit)
The control unit 20 is a device for controlling the dialogue device 1. The control unit 20 is realized by, for example, a CPU (Central Processing Unit). The control unit 20 functions as an input control unit 21, an output control unit 22, and an appropriate utterance value management unit 23, as shown in FIG. 1, for example. The program that causes the control unit 20 to function as the input control unit 21, the output control unit 22, and the utterance appropriate value management unit 23 may be stored in the storage unit 30 or may be externally supplied via a communication unit (not shown). It may be supplied.

The input control unit 21 is a configuration for processing information input to the dialogue device 1 using the input unit 10. The input control unit 21 can be configured by, for example, as shown in FIG. 1, a voice recognition unit 21a, an image recognition unit 21b, and an event processing unit 21c.

The voice recognition unit 21a executes a detection process of detecting the user's uttered voice included in the voice represented by the voice data provided from the microphone 11. The uttered voice of the user detected by this detection process may be the uttered voice of a specific user or may be the uttered voice of an unspecified user. Further, when the detection process is successful, the voice recognition unit 21a executes a conversion process for converting the detected utterance voice into utterance content (for example, text data in which the utterance content is represented by text). This conversion process can be realized by a known STT (Speech To Text) technique, for example. When this conversion processing is successful, the voice recognition unit 21a provides the obtained utterance content to the output control unit 22 and the utterance appropriate value management unit 23. The voice recognition unit 21a also notifies the utterance appropriate value management unit 23 of the correctness of the detection process and the correctness of the conversion process.

The image recognition unit 21b executes a detection process of detecting the image of the user included in the image represented by the image data provided by the camera 12. The image of the user detected by this detection process may be an image of a specific user or an image of an unspecified user. The image recognition unit 21b notifies the output control unit 22 of the correctness of this detection process.

The event processing unit 21c executes a specific process for specifying the operation content based on the operation information provided from the touch panel 13. If the identification processing is successful, the event processing unit 21c notifies the utterance appropriate value management unit 23 of the identified operation content. In addition, the event processing unit 21c detects an event that triggers the utterance of the dialog device 1 (hereinafter, referred to as an "utterance trigger event") based on the state information provided by the sensor 14. In the present embodiment, the vibration of the dialogue device 1 caused by the user touching the dialogue device 1 is used as the utterance trigger event. When the event processing unit 21c detects the utterance trigger event, the event processing unit 21c notifies the output control unit 22 to that effect.

The output control unit 22 is configured to cause the dialogue device 1 to execute a reaction according to information input to the dialogue device 1 using the input control unit 21. The output control unit 22 can be configured by a response generation unit 22a, an utterance adequacy determination unit 22b, a voice output control unit 22c, and a display control unit 22d, as shown in FIG. 1, for example.

The response generation unit 22a (1) determines the utterance content and the display content of the dialogue device 1 when the user speaks to the user and when the utterance trigger event occurs, and (2) appropriately utters the determined utterance content. The value management unit 23 is notified, (3) audio data representing the determined utterance content is provided to the audio output control unit 22c, and (4) image data representing the determined display content is provided to the display control unit 22d. When the user speaks, the speech recognition unit 21a provides the response generation unit 22a with the content of the user's utterance. In this case, the response generation unit 22a determines the utterance content (of the dialogue device 1) and the display content according to the utterance content (of the user) provided from the voice recognition unit 21a. On the other hand, when the utterance trigger event occurs, the event processing unit 21c notifies the response generation unit 22a of that fact. In this case, the response generation unit 22a determines the utterance content (display of the dialog device 1) and the display content according to the situation at that time.

The utterance adequacy determination unit 22b determines whether or not it is appropriate for the dialog device 1 to utter when the utterance trigger event occurs, according to the utterance appropriate value stored in the storage unit 30. The appropriate utterance value is a numerical value indicating the appropriateness of the dialogue device 1 to speak, and takes a relatively large value when it is appropriate to speak, and relatively when it is inappropriate to speak. Take a small value. The utterance adequacy determining unit 22b compares the appropriate utterance value with a predetermined threshold value to determine whether or not it is appropriate for the dialog device 1 to utter. Specifically, when the appropriate utterance value is equal to or higher than the threshold value (or larger than the threshold value), it is determined that it is appropriate to speak, and when the appropriate utterance value is lower than the threshold value (or equal to or lower than the threshold value), It is determined that speaking is inappropriate.

The voice output control unit 22c controls the speaker 41 included in the output unit 40 so as to audibly output the utterance content determined by the response generation unit 22a based on the voice data provided from the response generation unit 22a. The display control unit 22d controls the display 42 included in the output unit 40 so as to visually output the display content determined by the response generation unit 22a based on the image data provided from the response generation unit 22a.

The appropriate utterance value management unit 23 is stored in the storage unit 30 based on the utterance content (of the user) provided by the voice recognition unit 21a and the operation content (of the user) provided by the event processing unit 21c. Manage the appropriate value of the uttered speech. A method of managing the appropriate utterance value by the appropriate utterance value management unit 23 will be described later with reference to the drawings.

(Storage unit)
The storage unit 30 is a device for storing information in the dialogue device 1. The storage unit 30 is realized by, for example, a semiconductor memory. In addition to the utterance appropriate value described above, the storage unit 30 stores an operation content-change amount table, an utterance content-change amount table, and a response content-change amount table, which will be described later. The contents of these tables will be described later with reference to the referenced drawings.

(Output section)
The output unit 40 is a device for outputting information from the dialogue device 1. The output unit 40 is realized by a speaker 41 and a display 42 as shown in FIG. 1, for example.

The speaker 41 audibly outputs the utterance content determined by the response generation unit 22a under the control of the voice output control unit 22c. The display 42 visually outputs the display content determined by the response generation unit 22a under the control of the display control unit 22d. The display 42 forming the output unit 40 may be integrated with the touch panel 13 forming the input unit 10.

<Proper utterance value and its management>
Next, the utterance appropriate value and its management will be described. The appropriate utterance value is a numerical value indicating the appropriateness of the dialogue device 1 to speak at that time. The appropriate utterance value is updated by the appropriate utterance value management unit 23 when the user operates the interactive device 1 and when the user interacts with the interactive device 1.

In this embodiment, the appropriate utterance value is an integer value of 0 or more and 100 or less. When the dialog device 1 is the most inappropriate to speak, the appropriate speech value takes 0, and when the interactive device 1 is the most appropriate to speak, the appropriate speech value is 100. In addition, in the present embodiment, it is assumed that predetermined utterance appropriate values for a plurality of users are individually managed for each user. Further, in the present embodiment, it is assumed that the appropriate utterance values in a plurality of predetermined time zones are individually managed for each time zone.

(Initial value of appropriate utterance value)
The initial value of the appropriate utterance value will be described with reference to FIG. FIG. 2 is a table showing initial values of appropriate utterance values.

In the table shown in FIG. 2, appropriate utterance values for user A, user B, user C,... In time zones 0 to 4 o'clock, time zones 4 to 8 o'clock, time zones 8 to 12 o'clock,. The initial value is stored.

For example, the appropriate utterance value for the user A in the time zone of 8 to 12 o'clock is the initial value “50”. It is assumed that the user A is likely to accept the utterance by the dialogue device 1 because the user A is often at home during the time zone from 8 to 12 o'clock, so the initial value of the appropriate utterance value is , Such a relatively high value is set. On the other hand, the appropriate utterance value for the user A in the time zone 0 to 4:00 is the initial value "0". Since it is assumed that the user A is difficult to accept the utterance by the dialogue device 1 because the user A is often sleeping at the time of 0 to 4 o'clock, the initial value of the appropriate utterance value is , Such a relatively low value is set.

(Update appropriate utterance value)
The utterance appropriate value management unit 23 executes an update process for updating the utterance appropriate value when the user operates the dialog device 1. This update process is executed as follows, for example. That is, when the user operates the dialogue device 1, the event processing unit 21c notifies the utterance appropriate value management unit 23 of the operation content. The appropriate utterance value management unit 23 updates the appropriate utterance value according to the notified operation content.

In the present embodiment, the operation content-change amount table as shown in FIG. 3A is stored in the storage unit 30. The utterance appropriate value management unit 23 obtains the updated utterance appropriate value by adding the change amount corresponding to the operation content notified in this table to the utterance appropriate value before the update. For example, in this table, the amount of change corresponding to the operation content of “start application” is “+5”. Therefore, the appropriate utterance value management unit 23 increases the appropriate utterance value by “+5” when the user operates the dialog device 1 to activate the application. Alternatively, in this table, the amount of change corresponding to the operation content “interrupt speech” is “−1”. Therefore, the utterance appropriate value management unit 23 increases the utterance appropriate value by "-1" (decreases by 1) when the user operates the dialog device 1 to interrupt the utterance.

Further, the utterance appropriate value management unit 23 executes an update process for updating the utterance appropriate value when the user interacts with the dialogue device 1. This update process is executed as follows, for example. That is, when the user interacts with the interactive device 1, (1) the utterance content of the user is notified from the voice recognition unit 21a to the appropriate utterance value management unit 23, and (2) the response content of the interactive device 1 responds. The generation unit 22a notifies the appropriate utterance value management unit 23. The appropriate utterance value management unit 23 (1) updates the appropriate utterance value according to the notified utterance content of the user, and (2) updates the appropriate utterance value according to the notified response content of the interactive device 1. To do.

In the present embodiment, the utterance content-change amount table as shown in FIG. 3B is stored in the storage unit 30. The appropriate utterance value management unit 23 obtains the appropriate utterance value after update by adding the amount of change corresponding to the utterance content notified in this table to the appropriate utterance value before update. For example, in this table, the change amount corresponding to the utterance content of "Good morning" is "+5". Therefore, the utterance appropriate value management unit 23 increases the utterance appropriate value by “+5” when the user utters “Good morning”.

Further, in the present embodiment, a response content-change amount table as shown in FIG. 3C is stored in the storage unit 30. The utterance appropriate value management unit 23 obtains the updated utterance appropriate value by adding the amount of change corresponding to the response content notified in this table to the utterance appropriate value before the update. For example, the amount of change corresponding to the response content "Welcome to you" in this table is "-3". Therefore, the utterance appropriate value management unit 23 increases the utterance appropriate value by -3 (decreases by 3) when the interactive device 1 responds by saying "Welcome."

The appropriate utterance value management unit 23 performs (1) whether the voice recognition unit 21a detects the utterance voice correctly or not (2) the voice recognition unit 21a converts the utterance voice into the utterance content (text). The appropriate utterance value may be updated depending on whether the utterance is correct or not and (3) whether or not the determination process in which the response generation unit 22a determines the response content according to the utterance content is correct. Examples of the case where the determination process fails include a case where the utterance content obtained by the conversion process is not significant utterance content. When the detection process is successful and the conversion process is unsuccessful, the utterance appropriate value management unit 23 increases the utterance appropriate value by +1, for example. When the detection process succeeds, the conversion process succeeds, and the determination process fails, the utterance appropriate value management unit 23 increases the utterance appropriate value by “+3”, for example. When the detection process is successful, the conversion process is successful, and the determination process is successful, the utterance appropriate value management unit 23 increases the utterance appropriate value by “+5”, for example.

In the present embodiment, the appropriate utterance value management unit 23 manages the appropriate utterance value for each time period. That is, the utterance appropriate value management unit 23 resets the utterance appropriate value to the initial value at the start point of each time period. Here, the initial value of the appropriate utterance value corresponding to each time zone is determined with reference to the table shown in FIG. However, the present invention is not limited to this. For example, a configuration may be adopted in which the appropriate utterance value is managed for each arbitrary time unit (day, week, month, year, etc.). In this case, the appropriate utterance value is reset to the initial value once every time unit.

Further, in the present embodiment, the utterance appropriate value management unit 23 manages the utterance appropriate value for each user. That is, the utterance appropriate value management unit 23 provides the utterance appropriate value corresponding to each user when the user operates the dialogue device 1, when the user utters the dialogue device 1, or the dialogue device 1 Is updated when the user responds to the user (when another user operates the dialogue device 1, when another user speaks to the dialogue device 1 or when the dialogue device 1 responds to the other user). I will not update it when I do). However, the present invention is not limited to this. For example, when someone operates the dialogue device 1, regardless of the user who operates the dialogue device 1, it is possible to adopt a configuration in which the appropriate utterance value is updated according to the operation content. The same applies when someone speaks to the dialogue device 1 and when the dialogue device 1 responds to someone.

(Example of time-dependent change in appropriate utterance value)
FIG. 4 is a graph showing an example of the change over time of the appropriate utterance value. For example, looking at the time change of the appropriate utterance value in the time zone from 8 to 12 o'clock in the graph, the following can be seen.

That is, at the start point (that is, 8:00) of the time zone from 8 to 12 o'clock, the appropriate utterance value is set to the initial value 50 corresponding to the time zone. Next, at the time when the user performs the sleep release operation on the dialog device 1, the appropriate utterance value increases from 50 to 53. This is because in the operation content-change amount table shown in FIG. 3A, the operation content “release sleep” corresponds to the change amount “+3”. Next, when the user utters "Good morning" to the dialog device 1, the appropriate utterance value increases from 53 to 58. This is because the utterance content “good morning” corresponds to the change amount “+5” in the utterance content-change amount table shown in FIG. 3B. Next, when the interactive device 1 responds to the user with "Good morning", the appropriate utterance value increases from 58 to 59. This is because the response content "Good morning" corresponds to the variation amount "+1" in the response content-variation amount table shown in (c) of FIG. Next, when the user utters “I come” to the dialog device 1, the appropriate utterance value increases from 59 to 64. This is because, in the utterance content-change amount table shown in FIG. 3B, the utterance content “I come” corresponds to the change amount “+5”. Next, at the time when the interactive device 1 responds to the user by saying "Welcome," the appropriate utterance value decreases from 64 to 61. This is because, in the response content-variation amount table shown in FIG. 3C, the response content "Welcome" corresponds to the variation amount "-3". At the end point of time zone 8 to 12 o'clock (that is, 12 o'clock), the appropriate utterance value is set to the initial value 30 corresponding to the time zone 12 to 16 o'clock next to the time zone.

As described above, in the interactive device 1, the user operates the interactive device 1 according to the operation content, the utterance content uttered by the user to the interactive device 1, and the response content returned by the interactive device 1 to the user. , The appropriate utterance value is updated. For example, in a situation in which the user easily accepts the utterance by the dialogue device 1, such as after the user utters "good morning" to the dialogue device 1, the utterance appropriate value takes a relatively large value and the dialogue device 1 In a situation in which the user has difficulty accepting the utterance by the dialogue device 1, such as after the user responds to the user by saying “Welcome,” the appropriate utterance value takes a relatively small value. Therefore, by determining whether or not the dialogue device 1 executes the utterance according to the utterance appropriate value, the dialogue device 1 that utters at an appropriate timing can be realized.

In addition, in the present embodiment, the configuration in which the appropriate utterance value updated by the appropriate utterance value management unit 23 is immediately referred to by the appropriate utterance determination unit 22b has been described, but the present invention is not limited to this. That is, it is also possible to adopt a configuration in which the utterance appropriateness value updated by the utterance appropriateness value managing unit 23 is referred to by the utterance appropriateness determining unit 22b on and after the next day of the day. In this case, the utterance appropriateness value referred to by the utterance adequacy determining unit 22b in a certain time zone on a certain day is, for example, the utterance appropriateness value last updated by the utterance appropriateness value managing unit 23 in the time zone on the previous day of the day. Further, in this case, the utterance appropriateness value referred to by the utterance adequacy determining unit 22b in a certain time zone on a certain day is compared with the total utterance appropriateness value last updated by the utterance appropriateness value managing unit 23 in each time zone on the previous day of the day. It is also possible to set it as the ratio value of the utterance appropriate value last updated by the utterance appropriate value management unit 23 in the time zone of the day before the day.

<Utterance processing when an utterance trigger event occurs>
The speech process S2 executed by the dialogue device 1 when a speech trigger event occurs will be described with reference to FIG. FIG. 5 is a flowchart showing the flow of the utterance processing S2. Note that this flowchart is an example, and the present invention is not limited to this.

The utterance process S2 includes an event determination process S11, an appropriate utterance value acquisition process S12, an appropriateness utterance determination process S13, a response content generation process S14, an utterance execution process S15, and a user presence/absence determination process S16. There is. The contents of each process are as follows.

The event determination process S11 is a process in which the event processing unit 21c determines, based on the state information provided from the sensor 14, whether or not a speech trigger event has occurred. The event determination process S11 is repeated until the event processing unit 21c determines that a speech trigger event has occurred. When the event processing unit 21c determines that the utterance trigger event has occurred, the subsequent utterance appropriate value acquisition process S12 is executed.

The appropriate utterance value acquisition process S<b>12 is a process in which the appropriate utterance determination unit 22 b reads the appropriate utterance value from the storage unit 30. The utterance adequacy determination process S13 is a process in which the utterance adequacy determination unit 22b determines whether or not it is appropriate for the dialogue device 1 to speak, according to the appropriate utterance value acquired in the appropriate utterance value acquisition process S12. is there. The utterance adequacy determination process S13 determines that it is appropriate to speak, for example, when the appropriate utterance value acquired in the appropriate utterance value acquisition process S12 is equal to or greater than (or larger than) a predetermined threshold value. If the appropriate utterance value acquired in the appropriate utterance value acquisition process S12 is smaller than (or equal to or less than) the threshold value, it is determined that utterance is inappropriate. When the utterance suitability determination unit 22b determines that it is appropriate to speak, a response content generation process S14 described below is executed. On the other hand, when the utterance suitability determination unit 22b determines that it is inappropriate to speak, the user presence/absence determination process S16 described below is executed.

In the response content generation process S14, the response generation unit 22a determines the utterance content and the display content according to the situation at that time, and then (a) provides the voice output control unit 22c with the voice data representing the determined utterance content. In addition, (b) is a process of providing image data representing the determined display content to the display control unit 22d. In the utterance execution process S15, (a) the voice output control unit 22c controls the speaker 41 so as to audibly output the utterance content determined by the response generation unit 22a based on the voice data provided from the response generation unit 22a. At the same time, (b) the display control unit 22d controls the display 42 so as to visually output the display content determined by the response generation unit 22a based on the image data provided from the response generation unit 22a. ..

The user presence/absence determination process S16 is a process in which the event processing unit 21c determines, based on the state information provided from the sensor 14, whether or not a user is present around the dialogue apparatus 1. When the event processing unit 21c determines that there is a user around the dialogue apparatus 1, the response content generation process S14 and the utterance execution process S15 described above are executed, and then the flow ends. When the event processing unit 21c determines that there is no user around the interactive apparatus 1, the flow ends without executing the response content generation process S14 and the utterance execution process S15 described above.

[Embodiment 2]
Hereinafter, the dialogue device 1 according to the second embodiment of the present invention will be described with reference to FIG. For convenience of description, members having the same function will be denoted by the same reference numeral, and description thereof will be omitted as appropriate. In the second embodiment, the flow of the utterance processing S2 by the dialogue device 1 is different from that of the first embodiment.

FIG. 6 is a flowchart showing the flow of the utterance processing S2 of the dialog device 1 according to the second embodiment. Note that this flowchart is an example, and the present invention is not limited to this. The utterance process S2 of the dialogue apparatus 1 according to the second embodiment includes an event determination process S21, an appropriate utterance value acquisition process S22, an utterance appropriateness determination process S23, a response content generation process S24, an utterance execution process S25, and presence/absence of a user. The determination process S26 further includes an operation execution process S27 and a voice input presence/absence determination process S28. The contents of each process are as follows.

The event determination process S21 is a process in which the event processing unit 21c determines, based on the state information provided from the sensor 14, whether or not a speech trigger event has occurred. The event determination process S21 is repeated until the event processing unit 21c determines that a speech trigger event has occurred. When the event processing unit 21c determines that the utterance trigger event has occurred, the subsequent utterance appropriate value acquisition process S22 is executed.

The utterance appropriateness value acquisition process S22 is a process in which the utterance appropriateness determination unit 22b reads the utterance appropriateness value from the storage unit 30. The utterance adequacy determination process S23 is a process in which the utterance adequacy determination unit 22b determines whether or not it is appropriate for the dialog device 1 to utter according to the utterance appropriate value acquired in the utterance appropriate value acquisition process S22. is there.

In the utterance adequacy determination process S23, it is appropriate to utter when the utterance appropriate value acquired in the utterance appropriate value acquisition process S22 is a high value and is equal to or larger than (or larger than) a predetermined first threshold value. Judge that there is. In this case, after the response content generation process S24 similar to the response content generation process S14 of the first embodiment is executed, the utterance execution process S25 similar to the utterance execution process S15 of the first embodiment is executed.

On the other hand, when the appropriate utterance value acquired in the appropriate utterance value acquisition processing S22 is a medium value, which is smaller than the first threshold value and equal to or larger than the second threshold value (or larger than the second threshold value), the above-described first embodiment is performed. The user presence/absence determination process S26 similar to the user presence/absence determination process S16 is executed. Here, the second threshold is set to a value smaller than the first threshold.

In the user presence/absence determination process S26, when the event processing unit 21c determines that a user is present around the dialogue apparatus 1, the response content generation process S24 and the utterance execution process S25 described above are executed, and then the flow is executed. Ends. When the event processing unit 21c determines that there is no user around the dialogue apparatus 1, the flow ends without executing the response content generation process S24 and the utterance execution process S25.

If the utterance appropriate value acquired in the utterance appropriate value acquisition process S22 is a low value and smaller than the second threshold value (or equal to or less than the second threshold value), the operation execution process S27 and the voice input presence/absence determination process S28 described later are performed. Is executed.

In the operation execution process S27, for example, an operation that attracts the user, such as outputting a predetermined voice ("Let's talk together") from the speaker 41, is continuously performed for a predetermined time. In the voice input presence/absence determination process S28, for example, it is determined whether or not the voice input by the user is performed by the voice recognition unit 21a, assuming a state in which the user is moving but the user is moving and thus waits for voice input. To do. When the voice recognition unit 21a determines that the user's voice is input, the response content generation process S24 and the utterance execution process S25 described above are executed. On the other hand, if there is no voice input by the user even after the lapse of the predetermined time, the flow ends without executing the response content generation process S24 and the utterance execution process S25.

[Example of software implementation]
Each functional block of the control unit 20 (an example of the “control device” in the claims) may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or by software. You may.

In the latter case, the control unit 20 includes a computer that executes the instructions of a program that is software that realizes each function. The computer includes, for example, at least one processor, and at least one computer-readable recording medium that stores the program. Then, in the computer, the processor reads the program from the recording medium and executes the program to achieve the object of the present invention. As the processor, for example, a CPU (Central Processing Unit) can be used. As the recording medium, a "non-transitory tangible medium" such as a ROM (Read Only Memory), a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, or the like can be used. Further, a RAM (Random Access Memory) for expanding the program may be further provided. The program may be supplied to the computer via any transmission medium (communication network, broadcast wave, etc.) capable of transmitting the program. Note that one aspect of the present invention can also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[Summary]
(1) The control device (control unit 20) according to the first aspect of the present invention is a control device that controls the dialog device 1 that interacts with the user, and the utterance that represents the appropriateness of the dialog device 1 to speak. The appropriate value is set to the content of the operation performed by the user on the dialog device 1, the content of the utterance made by the user to the dialog device 1, or the dialog device 1 to the user. The appropriate utterance value management unit 23 for updating based on the content of the response made by the user and the appropriate utterance value management unit 23 for determining whether or not it is appropriate for the dialog device 1 to utter the user. An utterance adequacy determining unit 22b that determines the utterance adequacy value according to the updated utterance adequacy value.

According to the above configuration, it is determined whether or not it is appropriate for the dialogue device 1 to speak to the user, the contents of the operation performed by the user on the dialogue device 1, and the user. Can be determined based on the content of the utterance made by the dialogue device 1 or the contents of the response made by the dialogue device 1 to the user. Therefore, it is possible to execute the utterance at an appropriate timing for the user by determining whether or not to execute the utterance to the user based on the determination result.

(2) In the control device (control unit 20) according to the second aspect of the present invention, in the first aspect, the utterance appropriate value is managed for each time zone, and the utterance appropriate value management unit 23 sets each time. The appropriate utterance value corresponding to the time zone is updated by adding the amount of change depending on the content of the operation, the content of the utterance, or the content of the response to the current value of the appropriate utterance value corresponding to the zone. To do.

According to the above configuration, whether or not it is appropriate for the dialog device 1 to speak to the user can be independently determined for each time period. Therefore, it becomes possible for the user to perform the utterance at a more appropriate timing.

(3) In the control device (control unit 20) according to Aspect 3 of the present invention, in the above Aspect 1 or Aspect 2, the utterance appropriate value is managed for each user, and the utterance appropriate value management unit 23 is , The utterance appropriate for the user by adding the amount of change according to the content of the operation, the content of the utterance, or the content of the response to the current value of the utterance appropriate value corresponding to each user. Update the value.

According to the above configuration, it is possible to independently determine, for each user, whether it is appropriate for the dialogue device 1 to speak to each user. Therefore, it becomes possible for each user to execute the utterance at a more appropriate timing.

(4) In the control device (control unit 20) according to Aspect 4 of the present invention, in the above Aspects 1 to 3, when the utterance suitability determination unit 22b determines that the utterance suitability is appropriate, or the utterance suitability determination unit 22b is inappropriate. If it is determined that the user is present in the vicinity of the dialog device 1, a voice output control unit 22c that executes a speech to the user is further included.

According to the above configuration, when the user is present in the vicinity of the dialogue device 1, it is possible to perform the utterance to the user regardless of the result of the determination based on the appropriate utterance value. Become. As a result, it is possible to prevent the utterance by the dialogue device 1 from being excessively suppressed.

(5) In the control device (control unit 20) according to the fifth aspect of the present invention, when the utterance adequacy determination unit 22b determines that the utterance adequacy determination unit 22b is inappropriate in the first to fourth aspects, it is necessary to obtain the user's attention beforehand The apparatus further includes a speaker 41 and a display 42, which are operation execution units that cause the interactive apparatus 1 to execute a predetermined operation.
According to the above configuration, it is possible to change a situation where the utterance to the user is inappropriate to a situation where the utterance to the user is appropriate.

(6) The control method according to the sixth aspect of the present invention is a control method for controlling the dialogue device 1 for interacting with a user, wherein the utterance appropriate value representing the appropriateness of the utterance of the dialogue device 1 is The contents of the operation performed by the user on the dialogue device 1, the contents of the utterance made by the user to the dialogue device 1, or the response made by the dialogue device 1 to the user. The appropriate utterance value updating process that is updated based on the content, and whether or not it is appropriate for the dialog device 1 to utter the user, the appropriate utterance value updated by the appropriate utterance value updating process. Utterance suitability determination processing for determining according to.

According to the above method, it is determined whether it is appropriate for the dialogue device 1 to speak to the user, the contents of the operation performed by the user on the dialogue device 1, and the user. Can be determined based on the content of the utterance made by the dialogue device 1 or the contents of the response made by the dialogue device 1 to the user. Therefore, it is possible to execute the utterance at an appropriate timing for the user by determining whether or not to execute the utterance to the user based on the determination result.

(7) A control program according to aspect 7 of the present invention is a control program for operating a computer as a control device for controlling the dialogue apparatus 1 according to aspects 1 to 5, wherein the computer is used as each part of the control device. Make it work.

According to the control program, whether or not it is appropriate for the dialogue device 1 to speak to the user, the contents of the operation performed by the user on the dialogue device 1, The determination can be made based on the content of the utterance made by the user to the dialogue device 1 or the contents of the response made by the dialogue device 1 to the user. Therefore, it is possible to execute the utterance at an appropriate timing for the user by determining whether or not to execute the utterance to the user based on the determination result.

(8) The dialogue apparatus 1 according to the eighth aspect of the present invention is a dialogue apparatus that interacts with a user, and includes the control device (control unit 20) according to the first to fifth aspects that controls the dialogue apparatus.

According to the above configuration, it is determined whether or not it is appropriate for the dialogue device 1 to speak to the user, the contents of the operation performed by the user on the dialogue device 1, and the user. Can be determined based on the content of the utterance made by the dialogue device 1 or the contents of the response made by the dialogue device 1 to the user. Therefore, it is possible to execute the utterance at an appropriate timing for the user by determining whether or not to execute the utterance to the user based on the determination result.

The present invention is not limited to the above-described embodiments, but various modifications can be made within the scope of the claims, and embodiments obtained by appropriately combining the technical means disclosed in the different embodiments Is also included in the technical scope of the present invention. Furthermore, a new technical feature can be formed by combining the technical means disclosed in each embodiment.

1 Dialogue device 10 Input unit 20 Control unit (control device)
21 input control unit 21a voice recognition unit 21b image recognition unit 21c event processing unit 22 output control unit 22a response generation unit 22b utterance adequacy determination unit 22c voice output control unit 22d display control unit 23 utterance appropriate value management unit 30 storage unit 40 output unit 41 speaker 42 display

Claims (8)

A control device for controlling an interactive device for interacting with a user,
The utterance appropriate value representing the appropriateness of the dialogue device to speak, the content of the operation performed by the user on the dialogue device, the content of the utterance performed by the user on the dialogue device, or , An utterance appropriate value management unit that updates based on the content of the response that the dialog device makes to the user,
An utterance adequacy determination unit that determines whether or not it is appropriate for the dialog device to utter the user according to the utterance appropriate value updated by the utterance appropriate value management unit. A control device characterized by the above. The appropriate utterance value is managed for each time period,
The utterance appropriate value management unit adds the amount of change depending on the content of the operation, the content of the utterance, or the content of the response to the current value of the utterance appropriate value corresponding to each time zone, The control device according to claim 1, wherein the utterance appropriate value corresponding to the time zone is updated. The utterance appropriate value is managed for each user,
The utterance appropriate value management unit adds, to the current value of the utterance appropriate value corresponding to each user, the amount of change according to the content of the operation, the content of the utterance, or the content of the response, The control device according to claim 1 or 2, wherein the appropriate utterance value corresponding to the user is updated. If the utterance adequacy determining unit determines that it is appropriate, or if the utterance adequacy determining unit determines that the utterance appropriateness is inappropriate and it is confirmed that the user is present in the vicinity of the dialogue device, the user The control device according to any one of claims 1 to 3, further comprising a voice output control unit that executes an utterance to the user. The operation execution unit that causes the dialog device to execute a predetermined operation for obtaining the attention of the user when the utterance adequacy determination unit determines that the utterance suitability is inappropriate. 5. The control device according to claim 4. A control method for controlling an interactive device for interacting with a user, comprising:
The utterance appropriate value representing the appropriateness of the dialogue device to speak, the content of the operation performed by the user on the dialogue device, the content of the utterance performed by the user on the dialogue device, or , An utterance appropriate value update process for updating based on the content of a response made by the dialog device to the user,
Utterance adequacy determination processing for determining whether or not it is appropriate for the dialogue device to speak to the user according to the utterance appropriate value updated in the utterance appropriate value update processing. A control method characterized in that A control program for operating a computer as the control device according to any one of claims 1 to 5, the control program causing the computer to function as each unit of the control device. An interactive device for interacting with a user, comprising the control device according to any one of claims 1 to 5 as a control unit for controlling the interactive device.

Images