JP2023163600A - Baby-type interactive robot, baby-type interactive method and baby-type interactive program - Google Patents

Abstract

To provide a possible baby-type interactive robot, baby-type interactive method and baby-type interactive program.SOLUTION: A baby-type interactive robot comprises: a head part without any face design drawn thereon; a body part connected to the head part and having no movable portion; an input part including at least one of a microphone disposed inside of at least one of the head part and the body part and inputting voice, and one or a plurality of physical quantity sensors for detecting a physical quantity; an output part disposed inside thereof and including a speaker for outputting voice; an information processing part disposed inside thereof and outputting a speech signal to a speaker so as to output a speech of a baby in response to at least one of a speech of an elderly person with dementia input by the microphone and the physical quantity detected by the physical quantity sensor; and clothing detachably attached to the body part.SELECTED DRAWING: Figure 1

Description

The present disclosure relates to a baby-type conversation robot, a baby-type conversation method, and a baby-type conversation program for elderly people with dementia.

There are dolls that have a rehabilitation effect on elderly people suffering from dementia or depression, and elderly people living alone (see Patent Document 1). The doll has the ability to cry when a shock is applied to it.

Japanese Patent Application Publication No. 2003-093749

As described in Patent Document 1, there are dolls that cry when a shock is applied to them.
By the way, the applicant's research has revealed that when elderly people are allowed to use crying dolls, many of them develop negative impressions due to the crying, and many of them stop using the dolls after a relatively short period of time. Ta. Furthermore, it was found that not only the elderly people who use the dolls but also the people around them have a negative impression due to the doll's crying. In nursing care facilities, etc., if a doll used by an elderly person continues to cry, the caregiver may urge the elderly person to comfort the doll, which is a burden to the caregiver.

The present disclosure provides a baby-type conversation robot, a baby-type conversation method, and a baby-type conversation program for elderly people with dementia that can be used for a relatively long period of time.

In one embodiment, a baby-like interactive robot includes a head without a face design, a body without a movable part connected to the head, and an interior of at least one of the head and the body. , an input section including at least one of a microphone for inputting sound and one or more physical quantity sensors for detecting a physical quantity, an output section disposed inside including a speaker for outputting sound; an information processing unit that outputs a speech signal to a speaker so as to output the baby's speech according to at least one of the speech of the elderly person with dementia inputted by the microphone and the physical quantity detected by the physical quantity sensor; , and clothing that is detachably attached to the torso.

According to one aspect, a speech signal is output to the speaker so as to output the baby's speech in response to at least one of the speech of the elderly person with dementia inputted by the microphone and the physical quantity detected by the physical quantity sensor. , can be used for a relatively long period of time for elderly people with dementia.

FIG. 1 is a diagram for explaining a baby-type interactive robot according to an embodiment. FIG. 1 is a block diagram for explaining a baby-type interactive robot according to an embodiment. 1 is a flowchart for explaining an information processing method according to an embodiment.

An embodiment will be described below.

[Overview of baby-type conversational robot 1]
First, an overview of the baby-type interactive robot 1 will be explained.
FIG. 1 is a diagram for explaining a baby-type interactive robot 1 according to an embodiment.

The baby-type interactive robot 1 of this embodiment may be suitable for use by elderly people with dementia, for example, or may be suitable for use by elderly people, children, etc. It may be suitable for use by users of various ages.
The baby-shaped interactive robot 1 may be, for example, a baby-shaped doll (stuffed toy) including a head 11 and a body 12. The head 11 has no face design. The body portion 12 is connected to the head 11. Joints such as arms and legs of the body part 12 do not need to be movable. That is, the body portion 12 does not need to have any movable parts such as joints. Clothing 13 may be detachably attached to torso section 12 . Various methods can be used to put on and take off the clothes 13.

An input unit 113 (see FIG. 2) is arranged inside at least one of the head 11 and the body 12. The input unit 113 may include, for example, at least one of a microphone 113a for inputting audio and one or more physical quantity sensors 113b for detecting a physical quantity. The physical quantity sensor 113b may include, for example, a camera device.
An output unit 114 (see FIG. 2) is disposed inside at least one of the head 11 and the body 12. The output unit 114 may include, for example, a speaker 114a that outputs audio.

An information processing unit 103 (see FIG. 2) is arranged inside at least one of the head 11 and the body 12.
The information processing unit 103 outputs a speech signal to the speaker 114a to output the baby's speech according to at least one of the result of input to the microphone 113a and the result of detection by the physical quantity sensor 113b. The input to the microphone 113a may be, for example, the speech of an elderly person with dementia. The physical quantity sensor 113b may detect one or more various physical quantities when, for example, an elderly person with dementia touches the baby interactive robot 1. The speaker 114a outputs the baby's speech based on the speech signal.
Note that the information processing unit 103 may have a configuration in which the volume of the baby's speech output from the speaker 114a can be adjusted by, for example, a volume adjustment button (not shown) arranged on the body part 12 such as a hand. good.

[Details of baby-type conversational robot 1]
Next, the baby-type interactive robot 1 according to one embodiment will be described in detail.
FIG. 2 is a block diagram for explaining the baby-type interactive robot 1 according to one embodiment.

The baby-type interactive robot 1 includes, for example, a head 11, a body 12, clothing 13, a communication section 111, a storage section 112, an input section 113, an output section 114, a control section 101, and the like. The control unit 101 includes, for example, an acquisition unit 102 (for example, a first acquisition unit 102a and a second acquisition unit 102b, etc.), an information processing unit 103, and the like. The control unit 101 may be configured by, for example, an arithmetic processing device. The control unit 101 (for example, an arithmetic processing unit, etc.) reads and executes various programs stored in the storage unit 112, etc. as appropriate, thereby controlling the acquisition unit 102 (the first acquisition unit 102a and the second acquisition unit 102b). ) and the information processing unit 103, etc. may also be implemented.

The communication unit 111 is, for example, a communication interface that can send and receive various information to and from a device (external device) (not shown) outside the baby-type interactive robot 1 . The external device may be, for example, a server, a terminal, or the like. The terminal may be, for example, a desktop, laptop, tablet, smartphone, smartwatch, etc.

The storage unit 112 may store various information and programs, for example. An example of the storage unit 112 may be a memory, a solid state drive, a hard disk drive, or the like. Note that the storage unit 112 may be, for example, a storage area on a cloud, a server, or the like.
The storage unit 112 may store multiple types of baby's utterances.

The plurality of types of baby utterances stored in the storage unit 112 may be, for example, utterances indicating positive emotional expressions. The baby's utterance indicating a positive emotional expression stored in the storage unit 112 may be, for example, a cute laugh. Positive emotional expressions may be, for example, expressions of joy and happiness. For example, there may be multiple baby utterances that indicate positive emotional expression. In this case, there may be one or more baby utterances that indicate a weakly positive emotional expression, and one or more baby utterances that indicate a positive emotional expression that is a stronger emotional expression than the weakly positive one.

The plurality of types of baby utterances stored in the storage unit 112 may be, for example, utterances indicating at least one of normal and negative emotional expressions. The normal baby's utterances may be, for example, various utterances that the baby makes normally. The negative baby's utterance may be, for example, crying. Negative emotional expressions may be, for example, expressions of sadness and loneliness. For example, there may be a plurality of baby utterances that express negative emotions. In this case, there may be one or more baby utterances that express a weak negative emotion, and one or more baby utterances that express a negative emotion that is a stronger emotional expression than the weak negative.

The communication unit 111 and the storage unit 112 may be arranged inside at least one of the head 11 and the body 12.

The input unit 113 may be placed inside at least one of the head 11 and the body 12. The input unit 113 may include, for example, at least one selected from the group of a microphone 113a and a physical quantity sensor 113b.
The microphone 113a is, for example, a device for inputting audio.
The physical quantity sensor 113b is, for example, a device that detects a physical quantity. For example, one or more physical quantity sensors 113b may be arranged. Examples of the physical quantity sensor 113b include an acceleration sensor that detects acceleration, a contact sensor that detects contact of an object to the baby interactive robot 1, a distance sensor that detects the distance between the baby interactive robot 1 and an object, and an elderly person with dementia. A Doppler sensor that detects a heartbeat, etc., an angular acceleration sensor that detects a rate of change in angular velocity (angular acceleration), or the like may be used. The physical quantity sensor 113b is not limited to the above example, but may be a sensor capable of detecting various physical quantities.
Further, the physical quantity sensor 113b may include a camera device (not shown) capable of capturing an image of an elderly person with dementia. The camera device can, for example, take an image of a user (eg, an elderly person with dementia) who is near the baby-type interactive robot 1. A camera device can generate image information by capturing a moving image or a still image, for example. The physical quantities described above may include image information.

Note that at least one of the input units 113 may be placed outside the baby-type interactive robot 1, for example. In this case, at least one of the input units 113 transmits at least one of an audio signal based on the input audio and a detection signal based on the detected physical quantity to the baby-type interactive robot 1 via wireless or wire. Good too. For example, the control unit 101 may acquire at least one of the audio signal and the detection signal via the communication unit 111.
As an example, the input unit 113 may be provided in the above-mentioned external device (not shown) or the like.

The output unit 114 may be arranged inside at least one of the head 11 and the body 12. The output unit 114 may include, for example, a speaker 114a that outputs audio. Further, the output section 114 may include, for example, a piezoelectric element, a rotating weight (not shown), and the like. A piezoelectric element is, for example, a device that converts an electrical signal into mechanical vibration, that is, a voltage into force. The rotating weight has, for example, a rotating shaft placed eccentrically with respect to the central axis of the weight, and rotates when a signal is input. The weight generates vibrations, for example, by coming into contact with the housing during rotation.

The acquisition unit 102 acquires, for example, information input to the input unit 113. The acquisition unit 102 may include, for example, a first acquisition unit 102a, a second acquisition unit 102b, and the like. The first acquisition unit 102a may acquire, for example, the speech of an elderly person with dementia input through the microphone 113a. The second acquisition unit 102b may acquire, for example, a physical quantity detected by the physical quantity sensor 113b.

The information processing unit 103 outputs a speech signal to the speaker 114a so as to output the baby's speech according to the information acquired by the acquisition unit 102. That is, the information processing unit 103 outputs the baby's utterances according to at least one of the utterances acquired by the first acquisition unit 102a and the physical quantities (including the above-mentioned image information) acquired by the second acquisition unit 102b. A speech signal is output to the speaker 114a so as to cause the speech to occur.

Specifically, the information processing unit 103 may be placed inside at least one of the head 11 and the body 12. The information processing unit 103 sends a speech signal to the speaker 114a to output the baby's speech according to at least one of the speech of the elderly person with dementia inputted by the microphone 113a and the physical quantity detected by the physical quantity sensor 113b. Output. That is, the information processing unit 103 uses the speaker 114a in at least one of the following cases: when the speech of a user such as an elderly person with dementia is input through the microphone 113a, and when a physical quantity is detected by at least one physical quantity sensor 113b. output the baby's utterances. The speech signal may be, for example, an audio signal for outputting the baby's speech from the speaker 114a.
In this case, the information processing unit 103 may, for example, refer to audio information recorded in the storage unit 112 or the like and output a speech signal corresponding to the content of the baby's speech to the speaker 114a. The audio information may be, for example, information in which the contents of the baby's utterance are associated with the utterance signals corresponding to the contents of the utterance.

The information processing unit 103 may select one of a plurality of types of baby's utterances stored in the storage unit 112, and output a utterance signal regarding the selected baby's utterance to the speaker 114a. That is, the information processing unit 103 selects one of the plurality of baby's utterances according to at least one of the utterance of the elderly person with dementia and the detected physical quantity, and outputs the selected utterance from the speaker 114a. It may also be output. As an example, the information processing unit 103 may output, from the speaker 114a, utterances selected from the baby's utterances indicating emotional expressions in five categories: negative, weak negative, normal, weak positive, and positive. Further, as an example, the information processing unit 103 may output, from the speaker 114a, an utterance selected from among the baby's utterances indicating emotional expressions in three categories: normal, weakly positive, and positive.

The information processing unit 103 may output a speech signal to the speaker 114a so as to output the baby's speech when the volume of the sound input by the microphone 113a is equal to or greater than a threshold (first threshold). The threshold value (first threshold value) in this case may be set to various values depending on the size of the utterance, for example, so that the utterance of the elderly person with dementia can be detected.

Further, the information processing unit 103 may output a speech signal to the speaker 114a to output the baby's speech when the magnitude of the physical quantity detected by the physical quantity sensor 113b is equal to or larger than a threshold (second threshold). In this case, the threshold value (second threshold value) is set according to the physical quantity detected by the physical quantity sensor 113b (sensor detection content) so that the physical quantity can be detected when, for example, an elderly person with dementia is cradling the baby-type interactive robot 1. Various values may be set.

The information processing unit 103 recognizes the voice input by the microphone 113a, and based on the correspondence information (first correspondence information) that associates the voice content with the baby's utterance as a response, the information processing unit 103 converts the voice recognition result into a voice recognition result. A speech signal may be output to the speaker 114a so as to output the baby's speech in response.
The first correspondence information may be stored in the storage unit 112, for example. The first correspondence information includes, for example, a plurality of audio contents (the contents of the audio input by the microphone 113a), and the baby's utterances corresponding to each of the plurality of audio contents (the contents of the dialogue (response contents) according to the input). It may also be information that associates. As a specific example, the first correspondence information associates an utterance of "You're cute" to the baby-type conversational robot 1 as an utterance (audio content) of an elderly person with dementia, and a happy laugh as an utterance of a baby. It may be information.
The information processing unit 103 may, for example, perform voice recognition on the voice input through the microphone 113a using various methods, and identify the content of the voice input through the microphone 113a (voice content). The information processing unit 103 may, for example, refer to the first correspondence information to identify the baby's utterances according to the identified audio content. In this case, for example, if the voice recognition content includes the voice content recorded in the first correspondence information, the information processing unit 103 may identify the baby's utterance according to the voice content. Further, the information processing unit 103 may refer to the above-mentioned audio information and output a speech signal corresponding to the identified baby's speech to the speaker 114a.
Note that the audio information and the first correspondence information may be, for example, separate information, or may be information that is a combination of them.

The information processing unit 103 estimates the interaction according to the physical quantity detected by the physical quantity sensor 113b, based on the learned model that has learned the physical quantities applied to the baby-like interactive robot 1 and the interaction with the elderly person with dementia. A speech signal may be output to the speaker 114a so as to output the baby's speech according to the estimated relationship. Examples of interactions include an elderly person with dementia cradling the baby-type conversational robot 1, an elderly person with dementia sitting next to the baby-type conversational robot 1, and an elderly person with dementia looking at the baby-type conversational robot 1 with a smile. The relationship between the elderly person with dementia and the baby-type interactive robot 1 may include the elderly person with dementia and the baby-like conversation robot 1, such as the elderly person with dementia and the baby being petted.
It can be estimated that the elderly person with dementia is cradling the baby-like interactive robot 1, based on the detection results of physical quantities by, for example, an acceleration sensor, a contact sensor, an angular acceleration sensor, or the like. It is possible to estimate whether the elderly person with dementia is near the baby type, for example, based on the detection result of a physical quantity by a distance sensor or the like. The fact that the elderly person with dementia is smiling and looking at the baby-type interactive robot 1 can be estimated based on the result of imaging using, for example, a camera device. It is possible to estimate that an elderly person with dementia is fondling a baby, for example, based on the detection result of heart rate (for example, heart rate when the person is calm) by a Doppler sensor.
The trained model can be generated by, for example, a learning device (not shown). The learning device can generate a learned model by learning, for example, physical quantities and how they relate.
Note that the learned model may be, for example, a model generated by learning physical quantities, utterances of an elderly person with dementia, and how to interact.

The information processing unit 103 may acquire the learned model via the communication unit 111, for example. Further, the learned model may be stored in the storage unit 112, for example. In this case, the information processing unit 103 may refer to the trained model stored in the storage unit 112.
For example, based on the trained model and the physical quantity detected by the physical quantity sensor 113b, the information processing unit 103 estimates how the elderly person with dementia interacts with the baby-like interactive robot 1 according to the physical quantity.

For example, the information processing unit 103 identifies the baby's utterances according to the estimated manner of involvement. In this case, the information processing unit 103 may, for example, refer to correspondence information (second correspondence information) stored in the storage unit 112 or the like to identify the baby's utterances according to the estimated manner of involvement. The second correspondence information may be, for example, information that associates the contents of the plurality of ways of involvement with the baby's utterances corresponding to the contents of each of the plurality of ways of involvement. As a specific example, the second response information is information that associates a way of interacting with the elderly person with dementia, such as cradling the baby-type interactive robot 1, and a happy laugh as the baby's utterance. Good too.
The information processing unit 103 may, for example, refer to the second correspondence information and identify the baby's utterances according to the estimated baby's involvement. Further, the information processing unit 103 may refer to the above-mentioned audio information and output a speech signal corresponding to the identified baby's speech to the speaker 114a.
Note that the audio information and the second correspondence information may be, for example, separate information, or may be information that is a combination of them.
Further, the audio information, the first correspondence information, and the second correspondence information may be different pieces of information, or may be pieces of information that combine them into one piece of information, for example.
Further, the first correspondence information and the second correspondence information may be, for example, separate information, or may be information that is a combination of them.

After outputting the speech signal of the first utterance as the baby's utterance to the speaker 114a, the information processing unit 103 inputs the utterance of the elderly person with dementia through the microphone 113a or detects the physical quantity using the physical quantity sensor 113b. An utterance signal of a second utterance as an utterance indicating a more positive emotional expression than the first utterance may be output to the speaker 114a. That is, after the first utterance by the baby-like interactive robot 1 (after outputting the first utterance from the speaker 114a), when the utterance is further input through the microphone 113a, and when the physical quantity sensor 113b outputs the first utterance, the information processing unit 103 outputs the physical quantity In at least one of the cases where the second utterance is detected, the second utterance is output from the speaker 114a. The second utterance is an utterance that expresses a more positive emotion than the first utterance. In this case, the information processing unit 103 selects the first utterance from the above-mentioned five categories or a relatively low category among the three categories, outputs it from the speaker 114a, and then outputs the first utterance from the above-mentioned five categories. Alternatively, the second utterance may be selected from among the three categories that express a more positive emotion than the first utterance and output from the speaker 114a.

Each of the first utterance and the second utterance (category) may be set to the baby's utterance of the first correspondence information and the second correspondence information, for example. For example, the information processing unit 103 determines whether the elderly with dementia The speaker 114a may output a first utterance and a second utterance according to at least one of the utterance and the physical quantity detected by the physical quantity sensor 113b.

As an example, the information processing unit 103 outputs, from the speaker 114a, a first utterance selected from the baby's utterances indicating emotional expressions in five categories: negative, weak negative, normal, weak positive, and positive, and then After a lapse of time, a second utterance selected from among the baby's utterances indicating emotional expressions in the five categories described above is output from the speaker 114a, depending on at least one of the utterance of the elderly person with dementia or the detected physical quantity. It's okay. In this case, the information processing unit 103 may make the emotional expression of the second utterance more positive than the emotional expression of the first utterance. In addition, in the five categories mentioned above, the positive degree of emotional expression increases in the order of negative, weak negative, normal, weak positive, and positive, for example.
Note that the emotional expression is not limited to the emotional expression classified into the five categories described above, but may be an emotional expression classified into three categories: normal, weak positive, and positive, for example.

Alternatively, the first utterance and the second utterance may be different in at least one of the frequency of the baby's utterance and the size of the baby's utterance, for example. In this case, the first utterance may be, for example, a baby's utterance with a relatively low frequency or a baby's utterance with a relatively small volume. In addition, the second utterance may have a relatively high frequency of the baby's utterance (frequency of the first utterance <frequency of the second utterance), or a relatively large size of the baby's utterance (the first utterance The size of the second utterance may be smaller than the size of the second utterance.
In this case, the frequency and loudness of the baby's speech may be set in the speech signal, for example, depending on the difference between the first speech and the second speech.

Note that the output section 114 described above may include a vibrating section, a light emitting section (not shown), etc. in addition to the speaker 114a. The vibrating section includes, for example, a piezoelectric element or a rotating weight, and vibrates when a signal is input. The light emitting section includes, for example, a light emitting element that emits light, and emits light when a signal is input. In addition to outputting the speech signal to the speaker 114a, the information processing section 103 may also output a vibration signal to the vibration section so that the baby-like interactive robot 1 vibrates in response to the baby's speech. Furthermore, in addition to outputting the speech signal to the speaker 114a, the information processing section 103 may output a light emission signal to the light emitting section that emits light in response to the baby's speech. The baby-type interactive robot 1 may perform baby utterances and at least one of vibration and light emission in response to at least one of the utterances of the elderly person with dementia and the detected physical quantity.

[Baby-style dialogue method]
Next, an information processing method according to an embodiment will be described.
FIG. 3 is a flowchart for explaining an information processing method according to an embodiment.

In step ST101, the input unit 113 receives input of at least one of the utterance of the elderly person with dementia input through the microphone 113a and the physical quantity detected by the physical quantity sensor 113b.

In step ST102, the information processing unit 103 selects the utterance of the baby's utterance in accordance with at least one of the utterance of the elderly person with dementia and the physical quantity input in step ST101.

Specifically, the acquisition unit 102 first acquires, for example, the information input in step ST101. That is, the first acquisition unit 102a may acquire, for example, the utterance of an elderly person with dementia input through the microphone 113a. The second acquisition unit 102b may acquire, for example, a physical quantity detected by the physical quantity sensor 113b.
The information processing unit 103 selects the baby's utterance according to the information acquired by the acquisition unit 102. That is, the information processing unit 103 selects the baby's utterance according to at least one of the utterance acquired by the first acquisition unit 102a and the physical quantity acquired by the second acquisition unit 102b.

In this case, the information processing unit 103 may select one of a plurality of types of baby's utterances stored in the storage unit 112. An example of the plurality of types may be five categories of emotional expressions, such as negative, weak negative, normal, weak positive, and positive. Alternatively, an example of the plurality of types may be three categories of emotional expressions, such as normal, weakly positive, and positive.

The information processing unit 103 may select the baby's utterance when the loudness of the voice input to the microphone 113a in step ST101 is equal to or greater than a threshold (first threshold). Further, the information processing unit 103 may select the baby's utterance when the magnitude of the physical quantity detected by the physical quantity sensor 113b in step ST101 is equal to or greater than a threshold (second threshold).

The information processing unit 103 may recognize the voice input to the microphone 113a, and select the baby's utterance according to the result of voice recognition based on the correspondence information (first correspondence information). The correspondence information (first correspondence information) may be, for example, information that associates input audio content with the baby's utterance as a response.

The information processing unit 103 may estimate the manner of involvement according to the physical quantity detected by the physical quantity sensor 113b based on the learned model, and select the baby's utterance according to the estimated manner of involvement. The learned model may be, for example, a model that has learned the physical quantities applied to the baby-like interactive robot 1 and how an elderly person with dementia interacts with the robot.

When the information processing unit 103 outputs the baby's utterance (first utterance) from the speaker 114a in step ST103, which will be described later, and then returns to step ST101 to receive input of at least one of the utterance of the elderly person with dementia and the physical quantity. , an utterance (second utterance) indicating a more positive emotional expression than the first utterance may be selected.

In step ST103, the information processing unit 103 outputs the baby's utterance selected in step ST102. That is, the information processing unit 103 may control the speaker 114a to output the baby's utterance by outputting the utterance signal corresponding to the baby's utterance selected in step ST102 to the speaker 114a, for example.
In other words, the information processing unit 103 controls the speaker 114a to output a speech signal to output the baby's speech, depending on at least one of the speech and the physical quantity acquired by the acquisition unit 102 in step ST102.

Each part of the baby-type interactive robot 1 described above may be realized as a function of a computer processing unit or the like. That is, the acquisition unit 102 (the first acquisition unit 102a and the second acquisition unit 102b) and the information processing unit 103 (control unit 101) of the baby-type interactive robot 1 perform an acquisition function (first acquisition function) by an arithmetic processing unit of a computer, etc. and a second acquisition function) and an information processing function (control function).
The baby-type dialogue program can cause a computer to realize each of the above-mentioned functions. The baby-type interactive program may be recorded on a computer-readable non-transitory recording medium, such as a memory, a solid state drive, a hard disk drive, or an optical disk. A recording medium may also be referred to as a non-transitory computer-readable medium, for example.
Further, as described above, each part of the baby-type interactive robot 1 may be realized by a computer processing unit or the like. The arithmetic processing device and the like are constituted by, for example, an integrated circuit or the like. Therefore, each part of the baby-type interactive robot 1 may be realized as a circuit that constitutes an arithmetic processing device or the like. That is, the acquisition unit 102 (the first acquisition unit 102a and the second acquisition unit 102b) and the information processing unit 103 (control unit 101) of the baby-type interactive robot 1 are connected to the acquisition circuit (the first It may be realized as an information processing circuit (an acquisition circuit and a second acquisition circuit) and an information processing circuit (a control circuit).
Furthermore, the communication unit 111, storage unit 112, input unit 113, and output unit 114 of the baby-type interactive robot 1 may be realized as a communication function, a storage function, an input function, and an output function, including functions such as an arithmetic processing unit, for example. good. Furthermore, the communication section 111, storage section 112, input section 113, and output section 114 of the baby-type interactive robot 1 are realized as a communication circuit, a storage circuit, an input circuit, and an output circuit by being configured with an integrated circuit, for example. It's okay. Furthermore, the communication unit 111, storage unit 112, input unit 113, and output unit 114 of the baby-type interactive robot 1 are configured as a communication device, a storage device, an input device, and an output device by being configured with a plurality of devices, for example. It's okay.

The baby-type interactive robot 1 can combine one or any plurality of the plurality of parts described above.
Although the wording of "information" is used in this disclosure, the wording of "information" can be rephrased as "data" and the wording of "data" can be rephrased as "information."

[Aspects and effects of this embodiment]
Next, one aspect of this embodiment and the effects of each aspect will be described. Note that each aspect described below is an example at the time of filing, and the present embodiment is not limited to the aspects described below. That is, this embodiment is not limited to each aspect described below, and may be realized by appropriately combining each of the above-mentioned parts. Furthermore, a lower-level aspect may be cited by any higher-level aspect.
Further, the effects described below are just examples, and the effects achieved by each aspect are not limited to those described below. Moreover, each aspect may exhibit at least one effect described below, for example.

(Aspect 1)
One aspect of a baby-type conversational robot (for example, a baby-type conversational robot for elderly people with dementia) has a head without a face design, a body without a movable part connected to the head, an input unit including a microphone for inputting audio and at least one of one or more physical quantity sensors for detecting a physical quantity, disposed inside at least one of the head and the body; An output section including a speaker that outputs sound is arranged inside the device, and the baby's voice is output according to at least one of the utterances of the elderly person with dementia inputted by the microphone, and the physical quantity detected by the physical quantity sensor. The device includes an information processing unit that outputs a speech signal to a speaker so as to output speech, and clothing that is detachably attached to the torso.
By having a baby-shaped interactive robot, it can be more easily accepted by elderly people with dementia, and, for example, it can be more actively engaged with by elderly people with dementia than animal-type robots. . In addition, since the baby-type conversational robot does not have a face design, it is difficult for elderly people with dementia to estimate the baby's facial expressions based on their own ideas, such as "laughing" or "happy." You can actively participate in this process. In addition, since the baby-type conversational robot emits baby-like speech, it can act on the limbic system of the brain, which controls the emotions of elderly people with dementia, and can help them become attached to elderly people with dementia. This allows the baby-type interactive robot to continue to be used for a relatively long period of time for elderly people with dementia.
That is, the baby-like conversational robot can make elderly people with dementia feel positive emotions by, for example, making utterances that express positive emotions. Therefore, the baby-type interactive robot can continue to be used for a relatively long period of time for elderly people with dementia.
By making the elderly with dementia feel positive emotions, the baby-like conversational robot can help prevent problem behaviors such as verbal abuse, assault, and wandering, as well as behavioral and psychological symptoms of BPSD (BPSD) such as delusions and decreased motivation. Dementia) can be reduced.

(Aspect 2)
In one aspect of the baby-like interactive robot, the information processing unit may output a speech signal to the speaker so as to output the baby's speech when the volume of the sound input by the microphone is equal to or greater than a threshold value.
This allows the baby-like conversation robot to distinguish between the speech of an elderly person with dementia and the surrounding environmental sounds based on the threshold value.

(Aspect 3)
In one embodiment of the baby-like interactive robot, the information processing unit may output a speech signal to the speaker so as to output the baby's speech when the magnitude of the physical quantity detected by the physical quantity sensor is equal to or greater than a threshold value.
Thereby, the baby-like interactive robot can distinguish, for example, between the baby-like interactive robot being cradled by an elderly person with dementia and physical quantities based on the surrounding environment (such as noise), based on the threshold value.

(Aspect 4)
In one embodiment of the baby-type conversational robot, the information processing unit recognizes the voice input through the microphone, and based on the correspondence information that associates the voice content with the baby's utterance as a response, the information processing unit determines the voice recognition result. A speech signal may be output to a speaker so as to output the baby's speech according to the baby's speech.
That is, based on the correspondence information, the baby-like conversational robot outputs, from the speaker, the baby's utterance that corresponds to the utterance of the elderly person with dementia among the utterances of the plurality of babies. As a result, the baby-type conversational robot responds (baby's utterances) according to the emotional expression of the elderly person with dementia to the baby-type conversational robot, so it is possible to make the elderly person with dementia feel positive emotions and recognize It can be used for a relatively long period of time for elderly people.

(Aspect 5)
In one aspect of the baby-type conversational robot, the information processing unit is configured to perform processing according to the physical quantities detected by the physical quantity sensor, based on a trained model that has learned the physical quantities applied to the baby-type conversational robot and how an elderly person with dementia interacts with the baby-type conversational robot. The manner of involvement may be estimated and a speech signal may be output to a speaker so as to output the baby's utterances in accordance with the estimated manner of involvement.
That is, based on the learned model, the baby-like conversational robot outputs, from the speaker, a baby's utterance that corresponds to the emotional expression toward the baby by the elderly person with dementia, among the utterances of a plurality of babies. As a result, the baby-type conversational robot makes utterances that correspond to the emotional expressions expressed to the baby-type conversational robot by the elderly with dementia, making it possible to make the elderly with dementia feel positive emotions. It can be used for a relatively long period of time.

(Aspect 6)
In one embodiment, a baby-type conversational robot includes a storage unit that stores a plurality of types of baby's utterances, and the information processing unit selects one of the plurality of types of baby's utterances to be stored in the storage unit. , a speech signal regarding the selected baby's speech may be output to the speaker.
As a result, the baby-like interactive robot can respond (baby's utterances) according to the emotional expression of the elderly person with dementia.

(Aspect 7)
In one embodiment of the baby-like interactive robot, the plurality of types of baby utterances stored in the storage unit may be utterances indicating positive emotional expressions.
This allows the baby-like conversational robot to evoke positive emotions in elderly people with dementia.

(Aspect 8)
In one aspect of the baby-type conversational robot, the information processing unit outputs the speech signal of the first utterance as the baby's utterance to the speaker, and then inputs the utterance of the elderly person with dementia through the microphone, or inputs the utterance of the elderly person with dementia using the physical quantity sensor. When detected, a speech signal of a second utterance as a utterance indicating a more positive emotional expression than the first utterance may be output to the speaker.
This allows the baby-like conversational robot to evoke more positive emotions in elderly people with dementia.

(Aspect 9)
In one aspect of the baby-like interactive robot, the baby's utterance that indicates a positive emotional expression and is stored in the storage unit may be a cute laugh.
This allows the baby-like conversational robot to evoke positive emotions in elderly people with dementia.

(Aspect 10)
In one aspect of the baby-like interactive robot, the plurality of types of baby utterances stored in the storage unit may be utterances indicating at least one of normal and negative emotional expressions.
As a result, the baby-type conversational robot can perform baby utterances in accordance with the emotional expressions expressed to the baby-type conversational robot by the elderly person with dementia.

(Aspect 11)
In one aspect of the baby-type interactive robot, the physical quantity sensor may include a camera device capable of capturing an image of an elderly person with dementia.
Thereby, the baby-type conversational robot can acquire emotional expressions to the baby-type conversational robot by the elderly person with dementia based on the image information generated by the camera device.

(Aspect 12)
In one aspect of a baby-type interaction method (for example, a baby-type interaction method for elderly people with dementia), a head without a face design, a torso without a movable part connected to the head, An input unit including at least one of one or more physical quantity sensors for detecting a physical quantity and a microphone for inputting audio, which is disposed inside at least one of the head and the body, and a speaker. A baby-type conversational robot for elderly people with dementia, which is equipped with an output unit and clothes that can be attached to the torso in a removable manner.
A first acquisition step of acquiring the utterance of an elderly person with dementia inputted by a microphone, a second acquisition step of acquiring a physical quantity detected by a physical quantity sensor, the utterance acquired in the first acquisition step, and the second acquisition and an output control step of outputting a speech signal to a speaker so as to output the baby's speech according to at least one of the physical quantities acquired in the step.
Thereby, the baby-type conversation method may produce the same effects as the baby-type conversation robot of one embodiment described above.

(Aspect 13)
One aspect of a baby-shaped dialogue program (for example, a baby-shaped dialogue program for elderly people with dementia) includes a head without a face design, a torso with no movable parts connected to the head, An input unit including at least one of one or more physical quantity sensors for detecting a physical quantity and a microphone for inputting audio, which is disposed inside at least one of the head and the body, and a speaker. A first acquisition function that acquires the utterances of an elderly person with dementia input through a microphone to a baby-type conversation robot for elderly people with dementia that includes an output unit and clothing that is removably attached to the body. and a second acquisition function that acquires the physical quantity detected by the physical quantity sensor, and outputs the baby's utterance according to at least one of the utterance acquired by the first acquisition function, and the physical quantity acquired by the second acquisition function. and an output control function that outputs a speech signal to a speaker so as to achieve the desired result.
Thereby, the baby-type conversation program may produce the same effects as the baby-type conversation robot of one embodiment described above.

1 Baby type interactive robot 11 Head 12 Body part 13 Clothes 101 Control part 102 Acquisition part 102a First acquisition part 102b Second acquisition part 103 Information processing part 111 Communication part 112 Storage part 113 Input part 113a Microphone 113b Physical quantity sensor 114 Output part 114a speaker

Claims (13)

A head without a face design,
a body part without a movable part connected to the head;
an input unit that is disposed inside at least one of the head and the body and includes a microphone that inputs audio, and at least one of one or more physical quantity sensors that detect a physical quantity;
an output unit disposed inside the interior and including a speaker that outputs audio;
A speech signal is sent to the speaker so as to output the baby's speech in response to at least one of the speech of the elderly person with dementia inputted by the microphone arranged inside the interior and the physical quantity detected by the physical quantity sensor. an information processing unit that outputs;
Clothing that is removably attached to the torso;
A baby-shaped conversational robot for elderly people with dementia. The baby-type interactive robot according to claim 1, wherein the information processing unit outputs a speech signal to the speaker so as to output the baby's speech when the volume of the sound input by the microphone is equal to or greater than a threshold value. The baby-type interactive robot according to claim 2, wherein the information processing section outputs a speech signal to the speaker so as to output the baby's speech when the magnitude of the physical quantity detected by the physical quantity sensor is equal to or larger than a threshold value. The information processing unit recognizes the voice input by the microphone, and based on correspondence information that associates the voice content with the baby's utterance as a response, generates the baby's utterance according to the result of voice recognition. The baby-type interactive robot according to claim 1, wherein a speech signal is output to the speaker for output. The information processing unit estimates, based on a trained model that has learned the physical quantities applied to the baby-like interactive robot and the manner in which the elderly person with dementia interacts, the manner in which the elderly person with dementia interacts according to the physical quantity detected by the physical quantity sensor. 5. The baby-type interactive robot according to claim 4, wherein a speech signal is output to the speaker so as to output speech of the baby according to the baby's interaction. Equipped with a memory unit that stores multiple types of baby utterances,
The information processing unit selects one of a plurality of types of baby's utterances stored in the storage unit, and outputs a utterance signal regarding the selected baby's utterance to the speaker. The baby-type interactive robot according to any one of the items. The baby-like interactive robot according to claim 6, wherein the plurality of types of baby utterances stored in the storage unit are utterances indicating positive emotional expressions. The information processing unit outputs a speech signal of a first utterance as the baby's utterance to the speaker, and then inputs the speech of an elderly person with dementia through the microphone or detects a physical quantity using the physical quantity sensor. The baby-type interactive robot according to claim 7, wherein a speech signal of a second utterance, which is an utterance indicating a more positive emotional expression than the first utterance, is output to the speaker. 9. The baby-type interactive robot according to claim 8, wherein the baby's utterance indicating a positive emotional expression stored in the storage unit is a cute laugh. The baby-like interactive robot according to claim 9, wherein the plurality of types of baby utterances stored in the storage unit are utterances indicating at least one of normal and negative emotional expressions. The baby-type interactive robot according to claim 10, wherein the physical quantity sensor includes a camera device capable of capturing an image of an elderly person with dementia. A head without a face design,
a body part without a movable part connected to the head;
an input unit that includes at least one of one or more physical quantity sensors that detect a physical quantity and a microphone that inputs audio, the input unit being disposed inside at least one of the head and the body;
an output section including a speaker;
Clothing that is removably attached to the torso;
A baby-shaped conversational robot for elderly people with dementia that is equipped with
a first acquisition step of acquiring the utterances of the elderly person with dementia input by the microphone;
a second acquisition step of acquiring a physical quantity detected by the physical quantity sensor;
an output control step of outputting a speech signal to the speaker so as to output the baby's speech according to at least one of the speech obtained in the first acquisition step and the physical quantity obtained in the second acquisition step;
Dialogue methods for elderly people with dementia. A head without a face design,
a body part without a movable part connected to the head;
an input unit that includes at least one of one or more physical quantity sensors that detect a physical quantity and a microphone that inputs audio, the input unit being disposed inside at least one of the head and the body;
an output section including a speaker;
Clothing that is removably attached to the torso;
A baby-shaped conversational robot for elderly people with dementia, equipped with
a first acquisition function that acquires the utterances of an elderly person with dementia input by the microphone;
a second acquisition function that acquires a physical quantity detected by the physical quantity sensor;
an output control function that outputs a speech signal to the speaker so as to output the baby's speech according to at least one of the speech acquired by the first acquisition function and the physical quantity acquired by the second acquisition function;
A dialogue program for elderly people with dementia that realizes this.

Images