Interactive robot, control method, and program

Abstract

This interactive robot acquires a voice signal indicating the voice of a user and picked up by a microphone unit, estimates the direction of the user on the basis of the voice signal, outputs, to an actuator, an operation signal for operating the interactive robot so that the direction of the directivity of the microphone unit and the estimated direction of the user match each other to thereby cause the interactive robot to execute a response operation, and continuously outputs a response sound to the user from a speaker while executing the response operation.

Description

Interactive robot, control method, and program

[0001] The present disclosure relates to an interactive robot that interacts with a user.

[0002] Patent Document 1 discloses the following technology aimed at making the movements of a robot appear natural: that is, Patent Document 1 discloses a technology in which the direction of a sound source is estimated from an input signal received by a microphone array, and the gaze or posture or both of the robot are controlled so that the robot's direction of attention coincides with the estimated direction of the sound source, thereby aligning the pointing direction of the microphone array with the direction of attention.

[0003] However, the technology of Patent Document 1 has a problem in that if a user speaks to the robot during the period when the robot's line of sight, posture, etc. are controlled so that the robot's direction of attention matches the direction of the sound source, the direction of the microphone array does not match the direction of the user, and the user's voice is misrecognized.

[0004] Patent No. 3771812

[0005] The present disclosure has been made to solve such problems, and aims to provide a technology that prevents an interactive robot from erroneously recognizing a user's voice during a response operation.

[0006] In one aspect of the present disclosure, the conversational robot comprises a microphone unit including multiple microphones, an actuator for operating the conversational robot, a speaker, and a processor, wherein the processor acquires an audio signal indicating the user's voice picked up by the microphone unit, estimates the direction of the user based on the audio signal, outputs an operation signal to the actuator for operating the conversational robot so that the directional direction of the microphone unit matches the estimated direction of the user, thereby causing the conversational robot to perform a response action, and continuously outputs a response sound to the user from the speaker while the response action is being performed.

[0007] According to the present disclosure, it is possible to prevent an interactive robot from erroneously recognizing a user's voice during a response operation.

[0008] FIG. 1 is an external view of an interactive robot according to an embodiment, as seen from the front. FIG. 2 is a diagram showing an example of a rotation mechanism of the interactive robot. FIG. 3 is a block diagram showing an example of the configuration of the interactive robot. FIG. 4 is a diagram showing the positional relationship between the interactive robot and a user. FIG. 5 is a flowchart showing an example of the processing of the interactive robot according to embodiment 1. FIG. 6 is a flowchart showing an example of the processing of the interactive robot according to embodiment 2. FIG. 7 is a block diagram showing an example of the configuration of the interactive robot according to embodiment 3. FIG. 8 is a flowchart showing an example of the processing of the interactive robot according to embodiment 3. FIG. 9 is a block diagram showing an example of the configuration of an interactive robot according to embodiment 4. FIG. 10 is a diagram showing an example of the data configuration of a response period table. FIG. 11 is a flowchart showing an example of the processing of the interactive robot according to embodiment 4. FIG. 11 is a block diagram showing an example of the configuration of an interactive robot system according to embodiment 5.

[0009] (Findings underlying the present disclosure) Research is currently underway into interactive robots that can converse with users. In response to a user's request, the interactive robot will output a response sound such as "yes" from a speaker and rotate toward the user, thereby aligning the direction of the user with the direction of the directivity of the microphone unit of the interactive robot.

[0010] For example, if a conversational robot outputs a response sound only when it starts to rotate, the user may not be sure whether to speak to the conversational robot or not because the conversational robot is rotating. As a result, a situation may arise in which the user speaks to the conversational robot while it is rotating. In this case, the direction of the microphone's directivity does not match the direction of the user, and the conversational robot may misrecognize the user's voice.

[0011] In Patent Document 1, since it is not assumed that a user will talk to the robot during an operation in which the robot's gaze and posture are controlled so that the direction of the sound source coincides with the robot's gaze direction, the user cannot be made to clearly recognize whether or not it is okay to talk to the robot during this operation. Therefore, in Patent Document 1, a situation may arise in which a user talks to the robot during such an operation. In this case, since the direction of the directivity of the microphone does not coincide with the direction of the sound source, the robot may erroneously recognize the user's voice.

[0012] Therefore, the inventor discovered that if the response sound is continuously output from the speaker during the above-mentioned response operation, the user can be prevented from speaking during the response operation, thereby preventing misrecognition of the voice during the response operation, and came up with the present disclosure.

[0013] (1) In one aspect of the present disclosure, an interactive robot comprises a microphone unit including multiple microphones, an actuator for operating the interactive robot, a speaker, and a processor, wherein the processor acquires an audio signal indicating the user's voice picked up by the microphone unit, estimates the direction of the user based on the audio signal, outputs an operation signal to the actuator for operating the interactive robot so that the directional direction of the microphone unit matches the estimated direction of the user, thereby causing the interactive robot to perform a response action, and continuously outputs a response sound to the user from the speaker while the response action is being performed.

[0014] According to this configuration, while a response action is being performed to operate the interactive robot so that the direction of the microphone's directivity matches the estimated direction of the user, a response sound to the user is continuously output from the speaker, which prevents the user from speaking during the response action and thereby prevents the voice from being erroneously recognized during the response action.

[0015] (2) In the interactive robot described in (1) above, the processor may calculate the angle between the directional direction of the microphone unit and the direction of the user based on the audio signal, determine a response period indicating the period of the response action based on the angle, and continuously output the response sound from the speaker during the response period.

[0016] According to this configuration, the response period is determined based on the angle between the direction of the microphone unit's directivity and the direction of the user, so that the end timing of the output of the response sound can be determined without monitoring the response action to detect the end timing of the output of the response sound.

[0017] (3) In the interactive robot described in (1) or (2) above, the processor may acquire the movement speed of the interactive robot in the response movement, calculate a response period indicating the duration of the response movement based on the movement speed, and continuously output the response sound from the speaker during the response period.

[0018] According to this configuration, the response period is calculated taking into consideration the motion speed of the interactive robot, so that the end timing of the output of the response sound and the end timing of the response motion can be accurately matched.

[0019] (4) In the interactive robot described in any one of (1) to (3) above, a memory may be provided for storing response period information indicating the angle between the directional direction and the direction of the user and a response period indicating the period of the response action according to the angle, and the processor may determine the response period according to the angle calculated by referring to the response period information.

[0020] According to this configuration, since the response period information is referenced, the response period according to the angle can be easily determined.

[0021] (5) In the interactive robot described in any one of (1) to (4) above, the response action may be an action of rotating the interactive robot to align the directional direction of the microphone unit with the estimated direction of the user.

[0022] According to this configuration, the direction of the directivity of the microphone unit can be made to coincide with the direction of the user simply by rotating the robot.

[0023] (6) In the interactive robot described in any one of (1) to (5) above, the operating speed may be the rotation speed of the interactive robot, and the response period may be calculated by dividing the angle between the directional direction of the microphone unit and the estimated direction of the user by the rotation speed.

[0024] This configuration allows the response period to be calculated accurately.

[0025] (7) In the interactive robot according to any one of (1) to (6) above, the direction of the directivity may be a front direction of the interactive robot.

[0026] According to this configuration, the interactive robot turns its front direction toward the user by the response action, so that the timing when the user can talk to the interactive robot can be clearly indicated to the user.

[0027] (8) A control method in another aspect of the present disclosure is a control method for an interactive robot on a computer, which acquires an audio signal indicating a user's voice picked up by a microphone unit including multiple microphones equipped in the interactive robot, estimates the user's direction based on the audio signal, outputs an operation signal to an actuator of the interactive robot to cause the interactive robot to perform a response action that aligns the directional direction of the microphone unit with the estimated direction of the user, and continuously outputs a response sound to the user from a speaker of the interactive robot while the response action is being performed.

[0028] According to this configuration, it is possible to provide a control method for an interactive robot that can prevent erroneous recognition of the user's voice during a response operation.

[0029] (9) In yet another aspect of the present disclosure, a program causes a computer to execute a control method for an interactive robot, which program acquires an audio signal indicating a user's voice picked up by a microphone unit including multiple microphones provided in the interactive robot, estimates the user's direction based on the audio signal, outputs an operation signal to an actuator of the interactive robot to cause the interactive robot to perform a response action that aligns the directional direction of the microphone unit with the estimated direction of the user, and continuously outputs a response sound to the user from a speaker of the interactive robot while the response action is being performed.

[0030] According to this configuration, it is possible to provide a program that can prevent erroneous recognition of the user's voice during a response operation.

[0031] The present disclosure can also be realized as an interactive robot system operated by such a computer program. Needless to say, such a computer program can be distributed on a non-transitory computer-readable recording medium such as a CD-ROM or via a communication network such as the Internet.

[0032] Note that each of the embodiments described below represents a specific example of the present disclosure. The numerical values, shapes, components, steps, and step orders shown in the following embodiments are merely examples and are not intended to limit the present disclosure. Furthermore, among the components in the following embodiments, components that are not described in the independent claims that represent the highest concept are described as optional components. Furthermore, in all of the embodiments, the respective contents can be combined.

[0033] (Embodiment 1) Fig. 1 is an external view of a conversational robot 1 in embodiment 1 as seen from the front. In Fig. 1, the Y direction indicates the front direction of the conversational robot 1, the X direction indicates the left-right direction of the conversational robot 1, and the Z direction indicates the up-down direction of the conversational robot 1. The conversational robot 1 includes a main body 10, cameras 11 and 12, and a base 13. The main body 10 is configured to be rotatable around the Z axis, i.e., in the yaw direction, relative to the base 13.

[0034] The main body 10 includes a bottom surface 10a. The bottom surface 10a has a ring shape when viewed from above and includes a protrusion 10b protruding downward. The top surface of the base 13 has a ring-shaped groove that fits into the protrusion 10b. The main body 10 includes a dome-shaped housing 10c. The housing 10c houses the microphone 30, speaker 40, processor 20, and actuator 50 shown in FIG. 3 inside. The main body 10 rotates in the yaw direction by power from the actuator 50, guided by the groove in the base 13. As the main body 10 rotates relative to the base 13, the microphone 30, speaker 40, processor 20, and actuator 50 housed in the main body 10 also rotate integrally with the main body 10. The base 13 has a circular shape when viewed from above and is placed on the ground. This allows the interactive robot 1 to rest on the ground.

[0035] The cameras 11 and 12 are disposed at positions corresponding to the eyes of the interactive robot 1. More specifically, the cameras 11 and 12 are disposed symmetrically with respect to a central front position 2. The cameras 11 and 12 are image sensors that capture images of the surrounding environment.

[0036] 2 is a diagram showing an example of the rotation mechanism of the interactive robot 1. A gear shaft 131 is attached to the center of the base 13 and stands upright. A first gear 132 is attached to the upper end of the gear shaft 131. The actuator 50 is attached to a gear shaft 51 that extends downward from the bottom surface. A second gear 52 is attached to the lower end of the gear shaft 51. The first gear 132 and the second gear 52 are meshed with each other. When the actuator 50 is driven, the power of the actuator 50 causes the second gear 52 to rotate around the first gear 132. As a result, the main body 10 rotates around the Z axis relative to the base 13.

[0037] 3 is a block diagram showing an example of the configuration of the interactive robot 1. The interactive robot 1 includes a processor 20, a microphone unit 30, a speaker 40, and an actuator 50. The processor 20 is configured as a central processing unit (CPU). The processor 20 includes an acquisition unit 21, an estimation unit 22, a movement control unit 23, and a voice synthesis unit 24. The acquisition unit 21 to the voice synthesis unit 24 are realized by the processor 20 executing a program. However, this is just an example, and the acquisition unit 21 to the voice synthesis unit 24 may also be configured as dedicated hardware circuits.

[0038] The acquisition unit 21 acquires an audio signal indicating the user's voice picked up by the microphone unit 30. The audio signal acquired by the acquisition unit 21 includes multiple microphone-specific audio signals picked up by the multiple microphones constituting the microphone unit 30. When the acquisition unit 21 does not detect any voice, it sets the interactive robot 1 to a voice standby state. In the standby state, the acquisition unit 21 sets the direction of directivity of the microphone unit 30 to omnidirectional, i.e., 360 degrees. This allows the microphone unit 30 to pick up the voice of a user located at any position relative to the interactive robot 1 in the standby state. On the other hand, when the acquisition unit 21 acquires the user's voice signal, it sets the direction of directivity of the microphone unit 30 to the reference direction of the microphone unit 30. The reference direction of the microphone unit 30 is set to the front direction of the interactive robot 1. Therefore, the direction of directivity of the microphone unit 30 is directed toward the front direction of the interactive robot 1. The acquisition unit 21 may set the direction of directivity of the microphone unit 30 to the front direction using beamforming technology.

[0039] The estimation unit 22 estimates the direction of the user who spoke the voice, i.e., the direction of the sound source, based on the voice signal acquired by the acquisition unit 21. Specifically, the estimation unit 22 detects the time difference (phase difference) between the multiple microphone-specific voice signals collected by the multiple microphones constituting the microphone unit 30, and estimates the direction of the user using the detected phase difference. For example, if the microphone unit 30 is composed of a first microphone and a second microphone, the pitch between the first microphone and the second microphone is d, the speed of sound is c, the time difference is τ, and the angle between the reference direction of the microphone unit 30 and the direction of the user is θ, then the angle θ is expressed by the following equation (1).

[0040] θ=sin－１ (c×τ / d) (1)

[0041] The operation control unit 23 outputs an operation signal to the actuator 50 to operate the interactive robot 1 so that the direction of the directivity of the microphone unit 30 coincides with the estimated direction of the user. As a result, the interactive robot 1 performs a response operation.

[0042] The response action is an action of aligning the front direction of the interactive robot 1 with the direction of the user. In the following description, the response action is an action of rotating the main body 10 so that the front direction of the interactive robot 1 matches the direction of the user. Matching does not necessarily mean perfect match, but some deviation may be allowed. The allowable range of deviation is, for example, plus or minus 1 to 5 degrees.

[0043] The operation control unit 23 outputs an instruction signal to the voice synthesis unit 24 to continuously output a response sound to the user from the speaker 40 while the interactive robot 1 is performing a response operation. The response sound is a sound that responds to the user's voice. An example of the response sound is the sound "Yes." However, this is just an example, and any other appropriate sound such as "Okay" may be used as the response sound. Furthermore, sounds other than voice, such as a beep or a melody, may also be used as the response sound.

[0044] The voice synthesis unit 24 performs voice synthesis processing in accordance with the instruction signal output by the operation control unit 23 to generate a voice signal indicating a response sound, and outputs the generated voice signal indicating the response sound to the speaker 40. As a result, the response sound is output from the speaker 40.

[0045] The microphone unit 30 includes a plurality of microphones, picks up sound, and outputs an audio signal indicating the picked up sound to the acquisition unit 21 .

[0046] The speaker 40 converts the audio signal into sound.

[0047] The actuator 50 is configured by, for example, an electric motor. The actuator 50 receives an operation signal output from the operation control unit 23 and rotates the second gear 52 in accordance with the received operation signal. This causes the interactive robot 1 to rotate relative to the base 13.

[0048] Figure 4 is a diagram showing the positional relationship between the interactive robot 1 and the user 100. Figure 4 is a diagram showing the interactive robot 1 and the user 100 viewed from above. As shown in Figure 4, the interactive robot 1 has a circular shape when viewed from above. D0 indicates the front direction of the interactive robot 1. D1 indicates the direction of the user 100 (hereinafter referred to as the user direction). The front direction D0 faces in a direction connecting the center O of the interactive robot 1 and the front central position 2.

[0049] Since the center O and the front central position 2 are at the same height, the front direction D0 is parallel to the surface on which the interactive robot 1 is installed. The angle θ is the angle between the front direction D0 and the user direction D1. In other words, the angle θ is the angle of the yaw direction in which the user 100 is located when the front direction D0 is used as the reference. In this embodiment, the angle θ takes a value between 0 degrees and 180 degrees, with the clockwise direction being positive and the counterclockwise direction being negative, for example.

[0050] 5 is a flowchart showing an example of the processing of the interactive robot 1 in embodiment 1. Before the processing of this flowchart starts, the acquisition unit 21 sets the directivity direction of the microphone unit 30 to 360 degrees and puts the interactive robot 1 into a standby state.

[0051] In step S1, the acquisition unit 21 determines whether or not a voice signal has been acquired. Here, if the sound pressure of the acquired voice signal is equal to or greater than a predetermined level, the acquisition unit 21 determines YES in step S1. If the sound pressure is less than the predetermined level, the acquisition unit 21 determines NO in step S1. For example, if the average sound pressure of multiple microphone-specific voice signals acquired from multiple microphones is equal to or greater than a predetermined level, the acquisition unit 21 may determine that a voice signal has been acquired. The predetermined level may be, for example, a sound pressure level that is considered to be a person's speech. Note that if multiple users are speaking and there are multiple sound sources, the acquisition unit 21 may determine the direction of the sound source with the largest sound pressure level as the user direction D1. If the determination is YES in step S1, the process proceeds to step S2. If the determination is NO in step S1, the process waits in step S1.

[0052] Next, in step S2, the estimation unit 22 estimates the user direction D1 from the voice signal acquired by the acquisition unit 21. For example, the estimation unit 22 calculates the angle θ by applying Equation (1) to the voice signal acquired by the acquisition unit 21, and estimates the angle θ as the user direction D1.

[0053] Next, in step S3, the operation control unit 23 sets the direction of directivity of the microphone unit 30 to the front direction D0 by beamforming.

[0054] Next, in step S4, the operation control unit 23 starts a response operation by outputting an operation signal to the actuator 50 to rotate the main body 10 so that the front direction D0 coincides with the user direction D1. As a result, the interactive robot 1 starts a rotation operation to turn the front direction D0 of the main body 10 toward the user direction D1.

[0055] Next, in step S5, the operation control unit 23 outputs an instruction signal to the voice synthesis unit 24 to instruct the output of a response sound, thereby causing the speaker 40 to start outputting the response sound.

[0056] Next, in step S6, operation control unit 23 determines whether the direction of the directivity of microphone unit 30 coincides with user direction D1. Specifically, operation control unit 23 coincides the front direction D0 with user direction D1, thereby aligning the direction of the directivity of microphone unit 30 with user direction D1.

[0057] The operation control unit 23 may output an operation signal to the actuator 50 to rotate the main body unit 10 by the angle θ. In this case, the operation control unit 23 may use a current sensor to determine whether the supply of current to the actuator 50 has ended, and if it is determined that the supply of current has ended, determine that the direction of the directivity of the microphone unit 30 matches the user direction D1. Alternatively, the operation control unit 23 may use an angle sensor to determine whether the rotation angle of the main body unit 10 has reached the angle θ, and if the rotation angle of the main body unit 10 has reached the angle θ, determine that the direction of the directivity of the microphone unit 30 matches the user direction D1.

[0058] If the directional direction of the microphone unit 30 matches the user direction D1 (YES in step S6), the process proceeds to step S7, and if the directional direction of the microphone unit 30 does not match the user direction D1 (NO in step S6), the process waits in step S6.

[0059] Next, in step S7, the operation control unit 23 ends the response operation and the output of the response sound. For example, when a current sensor is used, the operation control unit 23 ends the output of the response sound by outputting an instruction signal to end the output of the response sound to the voice synthesis unit 24. When an angle sensor is used, the operation control unit 23 outputs an operation signal to the actuator 50 instructing the actuator 50 to end the response operation in addition to the instruction signal to end the output of the response sound.

[0060] As described above, according to the interactive robot 1 of the first embodiment, while the interactive robot 1 is performing a response action to operate the interactive robot 1 so that the direction of the directivity of the microphone unit 30 coincides with the user direction D1, the response sound to the user 100 is continuously output from the speaker. This prevents the user 100 from speaking during the response action, thereby preventing erroneous recognition of the voice during the response action.

[0061] (Embodiment 2) The interactive robot 1 in embodiment 2 is characterized in that it calculates a response period and performs a response action using the response period. In embodiment 2, the same components as in embodiment 1 are given the same reference numerals and their explanations will be omitted. The block diagram in embodiment 2 is the same as that in FIG. 3.

[0062] See Fig. 3. The operation control unit 23 determines a response period indicating the period of the response operation based on the angle θ calculated by the estimation unit 22, and causes the speaker 40 to continuously output a response sound during the response period. For example, the operation control unit 23 may calculate the response period by dividing the angle θ by a predetermined rotation speed. The predetermined rotation speed is the rotation angle per unit time in the rotation operation of the main body unit 10, i.e., the angular velocity. This rotation speed is stored in advance in a memory (not shown). The rotation speed is an example of an operation speed.

[0063] 6 is a flowchart showing an example of the processing of the interactive robot 1 in embodiment 2. The processing of step S11 is the same as step S1 in FIG. 5. In step S12, the estimation unit 22 calculates the angle θ between the front direction D0 and the user direction D1 from the voice signal acquired by the acquisition unit 21. Here, the estimation unit 22 may calculate the angle θ using the above-mentioned formula (1). The processing of step S13 is the same as step S3.

[0064] Next, in step S14, the operation control unit 23 determines the response period by dividing the angle θ calculated in step S12 by the rotation speed.

[0065] Next, in step S15, the operation control unit 23 outputs an operation signal to the actuator 50 to rotate the main body 10. As a result, the interactive robot 1 starts a rotation operation to orient the front direction D0 of the main body 10 toward the user direction D1. When rotating the main body 10 clockwise, the operation control unit 23 outputs an operation signal to the actuator 50 to rotate the actuator 50 clockwise. On the other hand, when rotating the main body 10 counterclockwise, the operation control unit 23 outputs an operation signal to the actuator 50 to rotate the actuator 50 counterclockwise.

[0066] The process in step S16 is the same as step S5 in FIG.

[0067] Next, in step S17, the operation control unit 23 determines whether or not the response period has elapsed. For example, the operation control unit 23 may measure the elapsed time from the start of the response operation using a timer, and determine whether or not the response period has elapsed from the measured elapsed time.

[0068] If the response period has elapsed (YES in step S17), the process proceeds to step S18, and if the response period has not elapsed (NO in step S17), the process waits in step S17. The process in step S18 is the same as step S7 in FIG. 5.

[0069] As described above, according to the interactive robot 1 in embodiment 2, the response period is determined based on the angle θ between the directional direction of the microphone unit 30 and the user direction D1, so that the end timing of the output of the response sound can be determined without monitoring the response behavior in order to detect the end timing of the output of the response sound.

[0070] (Embodiment 3) Embodiment 3 is characterized in that the response period is calculated using a rotation speed obtained from a speed table. In Embodiment 3, the same components as those in Embodiments 1 and 2 are given the same reference numerals, and their description will be omitted. Figure 7 is a block diagram showing an example of the configuration of the interactive robot 1 in Embodiment 3.

[0071] In the third embodiment, the interactive robot 1 further includes a memory 60. The memory 60 is configured by a rewritable nonvolatile storage device such as a flash memory. The processor 20 further includes a voice recognition unit 25.

[0072] The voice recognition unit 25 applies voice recognition processing to the voice signal acquired by the acquisition unit 21 and recognizes the content of the voice spoken by the user 100 .

[0073] The memory 60 stores a speed table T1. The speed table T1 stores a plurality of rotation speeds that can be set in the rotation operation of the main body 10. Specifically, the speed table T1 stores a plurality of rotation speeds corresponding to the utterance content recognized by the voice recognition unit 25. For example, if the utterance content of the user 100 is "Good morning," the speed table T1 stores a rotation speed corresponding to "Good morning," and if the utterance content of the user 100 is the name of the interactive robot 1, the speed table T1 stores a rotation speed corresponding to the name.

[0074] The operation control unit 23 obtains a rotation speed corresponding to the speech content recognized by the voice recognition unit 25 from the speed table T1, and sets the rotation speed of the actuator 50 so that the main body unit 10 rotates at the obtained rotation speed.

[0075] 8 is a flowchart showing an example of the processing of the interactive robot 1 in the embodiment 3. The processing of steps S21 to S23 is the same as steps S11 to S13 in FIG.

[0076] Next, in step S24, the voice recognition unit 25 applies voice recognition processing to the voice signal acquired by the acquisition unit 21, and recognizes the content of the utterance of the user 100. Here, when multiple users speak, the content of the utterance of the user 100 may be recognized from the voice signal from the direction of the sound source with the greatest sound pressure.

[0077] Next, in step S25, the operation control unit 23 acquires from the speed table T1 a rotation speed corresponding to the voice recognized by the voice recognition unit 25. If the speed table T1 does not store a rotation speed corresponding to the voice recognized by the voice recognition unit 25, the operation control unit 23 may acquire a default rotation speed from the speed table T1.

[0078] Next, in step S26, the operation control unit 23 determines the response period by dividing the angle θ calculated in step S22 by the rotation speed acquired in step S25.

[0079] Next, in step S27, similarly to step S15, the operation control unit 23 outputs an operation signal to the actuator 50. At this time, the operation control unit 23 sets the rotation speed of the actuator 50 to the rotation speed acquired in step S25.

[0080] The processing in steps S28 to S30 is the same as that in steps S16 to S18 in FIG.

[0081] In this way, according to the interactive robot 1 of embodiment 3, the response period is calculated taking into account the rotation speed according to the content of the utterance, so that the end timing of the output of the response sound and the end timing of the response action can be accurately matched.

[0082] (Fourth embodiment) The interactive robot 1 of the fourth embodiment is characterized in that it determines the response period by referring to a response period table. Fig. 9 is a block diagram showing an example of the configuration of the interactive robot 1 of the fourth embodiment.

[0083] In the fourth embodiment, the memory 60 of the interactive robot 1 stores a response period table T2. The response period table T2 stores the angle θ between the directivity direction of the microphone unit 30 and the user direction D1, and a response period indicating the period of a response action corresponding to the angle θ. The response period table T2 is an example of response period information.

[0084] FIG. 10 is a diagram showing an example of the data configuration of the response period table T2. The response period table T2 stores a response period of 0.5 seconds when the angle θ is 0 degrees or greater and less than 45 degrees, a response period of 0.75 seconds when the angle θ is 45 degrees or greater and less than 90 degrees, and a response period of 1 second when the angle θ is 90 degrees or greater and less than 180 degrees. The response period table T2 stores the absolute value of the angle θ. Therefore, the response period table T2 defines an angle range of 0 degrees or greater and 180 degrees or less. The response period table T2 defines a relationship between the angle θ and the response period such that the response period increases as the angle θ increases. The response period values ​​stored in the response period table T2 are merely examples, and other values ​​may be used as appropriate. In FIG. 10, the response period is defined according to three angle ranges, but the angle ranges are not limited to three and may be two, four, or more.

[0085] Returning to Fig. 9, the operation control unit 23 determines the response period according to the angle θ by referring to the response period table T2.

[0086] 11 is a flowchart showing an example of the processing of the interactive robot 1 in the embodiment 4. The processing of steps S31 to S33 is the same as steps S11 to S13 in FIG.

[0087] In step S34, the operation control unit 23 refers to the response period table T2 and determines the response period according to the angle θ calculated in step S32. The processes in steps S35 to S38 are the same as steps S15 to S18 in FIG.

[0088] As described above, the interactive robot 1 according to the fourth embodiment refers to the response period table T2, so that the response period according to the angle θ can be easily determined.

[0089] (Embodiment 5) Embodiment 5 is characterized in that the interactive robot 1 is controlled using a cloud server. Fig. 12 is a block diagram showing an example of the configuration of an interactive robot system in embodiment 5. The interactive robot system includes an interactive robot 1 and a cloud server 200. The interactive robot 1 and the cloud server 200 are connected to each other via a network NT so as to be able to communicate with each other.

[0090] The cloud server 200 includes a communication unit 210 and a processor 220. The communication unit 210 is configured with a communication circuit that connects the cloud server 200 to the network NT. The communication unit 210 transmits the movement signal generated by the movement control unit 23 and the audio signal of the response sound generated by the voice synthesis unit 24 to the interactive robot 1. The processor 220 is configured with, for example, a central processing unit (CPU), and includes the acquisition unit 21, the estimation unit 22, the movement control unit 23, and the voice synthesis unit 24 shown in FIG. 1 .

[0091] The interactive robot 1 includes a communication unit 110 and a processor 120 in addition to the microphone unit 30, the speaker 40, and the actuator 50 shown in Fig. 3. The communication unit 110 is a communication circuit that connects the interactive robot 1 to the network NT. The processor 120 controls the interactive robot 1. The communication unit 110 transmits an audio signal indicating the sound picked up by the microphone unit 30 to the cloud server 200.

[0092] According to the interactive robot 1 of the fifth embodiment, the interactive robot 1 can be controlled by the cloud server 200. Note that the cloud server 200 may be equipped with each block of the processor 220 shown in the second to fourth embodiments.

[0093] The present disclosure can employ the following modifications.

[0094] In the above embodiment, the interactive robot 1 is a stationary robot, but it may also be a mobile robot. In this case, the interactive robot 1 has one or more pairs of wheels on the base 13. The actuator 50 drives the wheels to move the interactive robot 1 in any direction. The operation control unit 23 outputs an operation signal to the actuator 50 so that the user direction D1 coincides with the directional direction of the microphone unit 30, thereby moving the interactive robot 1. As a result, the interactive robot 1 can align the user direction D1 with the directional direction of the microphone unit 30 by moving.

[0095] 5, if the sound pressure level of the audio signal is equal to or higher than a predetermined level, the acquisition unit 21 may determine whether the audio signal represents a voice uttered by a person from the feature of the audio signal, and may determine YES in step S1 if it is determined that the audio signal represents a voice uttered by a person, or may determine NO in step S1 if it is determined that the audio signal does not represent a voice uttered by a person. This is also applicable to steps S11, S21, and S31.

[0096] The present disclosure is useful in the field of pet robots that interact with users.

Claims

1. A conversational robot comprising: a microphone unit including a plurality of microphones; an actuator for operating the conversational robot; a speaker; and a processor, wherein the processor acquires an audio signal indicating a user's voice picked up by the microphone unit; estimates the direction of the user based on the audio signal; outputs an operation signal to the actuator for operating the conversational robot so that the direction of the directivity of the microphone unit coincides with the estimated direction of the user, thereby causing the conversational robot to perform a response operation; and continuously outputs a response sound to the user from the speaker while the response operation is being performed.

2. The interactive robot of claim 1, wherein the processor calculates an angle between the directional direction of the microphone unit and the direction of the user based on the audio signal, determines a response period indicating the duration of the response action based on the angle, and continuously outputs the response sound from the speaker during the response period.

3. The interactive robot according to claim 1, wherein the processor: acquires a movement speed of the interactive robot in the response movement; calculates a response period indicating a period of the response movement based on the movement speed; and continuously outputs the response sound from the speaker during the response period.

4. The interactive robot according to claim 2, further comprising a memory for storing response period information indicating an angle between the direction of the directivity and the direction of the user, and a response period indicating the duration of the response action corresponding to the angle, wherein the processor refers to the response period information and determines the response period corresponding to the calculated angle.

5. The conversational robot according to claim 1 or 2, wherein the response action is an action of rotating the conversational robot so as to match the direction of the directivity of the microphone unit with the estimated direction of the user.

6. The interactive robot according to claim 3, wherein the operating speed is a rotation speed of the interactive robot, and the response period is calculated by dividing the angle between the directional direction of the microphone unit and the estimated direction of the user by the rotation speed.

7. The conversational robot according to claim 1 or 2, wherein the direction of the directivity is a front direction of the conversational robot.

8. A control method for a conversational robot in a computer, comprising the steps of: acquiring an audio signal indicating a user's voice picked up by a microphone unit including multiple microphones equipped in the conversational robot; estimating the direction of the user based on the audio signal; outputting an operation signal to an actuator of the conversational robot to cause the conversational robot to perform a response action that matches the directional direction of the microphone unit with the estimated direction of the user; and continuously outputting a response sound to the user from a speaker of the conversational robot while the response action is being performed.

9. A program for causing a computer to execute a control method for a conversational robot, the program comprising: acquiring an audio signal indicating a user's voice picked up by a microphone unit including multiple microphones equipped in the conversational robot; estimating the direction of the user based on the audio signal; outputting an operation signal to an actuator of the conversational robot to cause the conversational robot to perform a response action that matches the directional direction of the microphone unit with the estimated direction of the user; and continuously outputting a response sound to the user from a speaker of the conversational robot while the response action is being performed.

Images