JP2008197381A - Loudspeaker controller, robot, loudspeaker control method and loudspeaker control program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a loudspeaker controller etc. capable of automatically controlling sound output by a loudspeaker so that sound may be suitable for both normal person and hearing-impaired person. <P>SOLUTION: The loudspeaker controller of the embodiment is equipped with: a person recognition means 201 for detecting a position of a first person and a position of a second person; an arrangement calculation means 301 for calculating a range of a position of a second loudspeaker, where the first person is not within the range, and the second person is within the range of directivity of the second loudspeaker 102, based on the positions of the first person and the second person; an operation means 401 which changes a position of the second loudspeaker 102 so that it may be within the range; and a sound volume compensation means 501 for adjusting a sound volume of the first loudspeaker 101 and the second loudspeaker 102 so that the sound volume of the first loudspeaker reaching the first person may be within a predetermined range, and the sound volume of the second loudspeaker 102 reaching the second person may be within a predetermined range. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a speaker control device, a robot including the same, a speaker control method, and a speaker control program.

  In recent years, many partner-type robots have been developed that communicate with people by voice. Among them, there are robots that have been developed for the purpose of talking to the hearing-impaired person or their caregiver or responding to the hearing from the hearing-impaired person or their caregiver. The purpose of this is for the robot to support communication between the hearing impaired and the caregiver, or for the caregiver to support communication between the robot and the hearing impaired.

  When a robot performs such voice communication, the loudness of the sound produced by the robot is important. This is because if a person with a normal activity under middle age and a robot communicate with each other by voice, if the human thinks that the voice produced by the robot is not appropriate, the robot can change the loudness of the sound, If the sound is low, humans may approach the robot, but in the case of the hard of hearing, the degree of activity and interest are decreasing, so even if the sound emitted by the robot is loud or low, it suits itself I don't try to fix it, and the hearing-impaired people rarely move.

  Even for a caregiver, it is difficult to correct the volume because it is not clear whether the volume of the sound produced by the robot is favorable for the hearing impaired. As a result, voice communication between the hard of hearing and the robot often fails. Further, in the case of a hearing impaired person, the hearing ability is generally weakened, and as shown in FIG. 5 of Patent Document 3, the high frequency level is reduced by 10 to 20 dB on average. We must also consider to compensate.

  As a method for controlling the volume of the robot, there is a method of measuring the distance from the robot to the other person and changing the volume according to the distance, as described in Patent Document 1 and Patent Document 2. In Patent Document 1, the other person is identified, and the reference volume is changed according to the other person. In Patent Document 2, the loudness is changed according to the distance from the robot to the human.

  Moreover, although it is not a robot, there exist patent document 3 and patent document 4. In patent document 3, an omnidirectional speaker and a directional speaker are used simultaneously. The omnidirectional speaker emits a normal sound for a normal hearing person, and the directional speaker emits a sound that emphasizes a high frequency level for a hearing impaired person. According to Patent Document 3, the directional speaker is attenuated by about 15 dB in the high frequency range of 2 KHz in a direction deviated by 30 degrees from the front. Therefore, by adjusting the direction of the directional speaker, it is suitable for both the hearing person and the deaf. You will hear the sound.

  Patent Document 4 is a speaker device that can adjust the directivity of a directional speaker. As an object, it is assumed that the directivity is narrowed when both the normal hearing person and the deaf person are present, and the directivity is widely adjusted when only the deaf person is present.

Non-Patent Document 1 discloses a robot capable of detecting and identifying a human face.
JP 2005-202076 A JP 2005-335001 A Japanese Patent Laid-Open No. 10-234094 JP 2000-184488 A NEC Media Information Laboratory Robot Development Center, Internet <http://www.incx.nec.co.jp/robot/robotcenter.html>

  In Patent Document 1, since the other person is identified and the reference volume is changed according to the other person, it is possible to deal with both normal hearing persons and hearing-impaired persons. Personal adaptation to is possible. However, Patent Document 1 does not consider the case where the communication partners of the robot are a plurality of people with different reference volumes. Therefore, when a hearing-impaired person and a caregiver communicate with a robot by voice, both sounds generated by the robot are not appropriate, and voice communication is not successful.

  In Patent Document 2, it is possible to change the volume of the sound according to the distance from the robot to the human, but since the amount of change does not change depending on the person, it is assumed whether it is a normal hearing person or a deaf person. Change the volume. However, as described above, even if it is a hearing-impaired person, the way of hearing is different for each person, and it is desirable to change the reference volume for each individual. Therefore, the invention according to Patent Document 2 is insufficient when targeting a hearing impaired person.

  Patent Document 3 and Patent Document 4 describe using a directional speaker, but do not mention at all how to direct the directional speaker toward a hearing-impaired person. It may be assumed that the person changes his / her own direction, but changing the position of the speaker manually is quite cumbersome (there is no habit of doing that in normal life) and effective It is thought that there are many cases that are not used for

  Among robots that are currently being developed, there are robots that can detect and identify human faces, as in Non-Patent Document 1, for example. Therefore, in combination with the face detection / identification function, a method of directing the directional speaker toward the deaf (deaf) can be considered. However, the problem is not so simple. Even if the directional speaker is directed toward the hearing impaired person, if the caregiver (normal hearing person) is not as far as possible from the directional range of the directional speaker, the sound becomes difficult for the caregiver to hear. Therefore, it recognizes the positional relationship between the hearing-impaired person and the hearing-impaired person, and determines the appropriate position and direction of the speaker (robot) so that the hearing-impaired person does not enter the range while directing the directional speaker toward the hearing-impaired person. Or change the direction of the speaker.

  The speaker control device according to the present invention is a device for controlling the first and second speakers, and includes a person recognition unit that detects the position of the first person and the position of the second person, and the person recognition unit. Based on the detected positions of the first and second persons, the second person enters the second speaker and the first person does not enter the directivity range of the second speaker. An arrangement calculating means for calculating the range of the position of the second speaker, and an operating means for changing the position of the second speaker so as to fall within the range of the position of the second speaker calculated by the arrangement calculating means. And the volume of the first speaker reaching the first person is within a predetermined range, and the volume of the second speaker reaching the second person is within a predetermined range. Adjust the volume of the first and second speakers And volume compensation means that, characterized in that it comprises a.

  In this speaker control device, the positions of the first and second people are detected by the person recognition means. Based on the detected position, the position range (possible area) of the second speaker is calculated by the arrangement calculating means. The existence possible area is an area where the second person falls within the pointing range of the second speaker and the first person does not fall within the range. Then, the position of the second speaker is changed by the operating means so as to fall within the possible area. Therefore, the sound emitted from the speaker can be automatically controlled so as to be appropriate for both the normal hearing person (first person) and the hard of hearing person (second person).

  The speaker control method according to the present invention is a method for controlling the first and second speakers, the person recognition step for detecting the position of the first person and the position of the second person, and the person recognition Based on the positions of the first and second persons detected in the step, the second person enters the directivity range of the second speaker and the first person does not enter, An arrangement calculation step for calculating the range of the position of the second speaker, and the position of the second speaker is changed so as to fall within the range of the position of the second speaker calculated in the arrangement calculation step. The volume of the first speaker reaching the first person is within a predetermined range, and the volume of the second speaker reaching the second person is within a predetermined range. As above And volume compensation step of adjusting the volume of the second speaker beauty, characterized in that it comprises a.

  The speaker control program according to the present invention is a program for controlling the first and second speakers, the person recognition step for detecting the position of the first person and the position of the second person, and the person recognition Based on the positions of the first and second persons detected in the step, the second person enters the directivity range of the second speaker and the first person does not enter, An arrangement calculation step for calculating the range of the position of the second speaker, and the position of the second speaker is changed so as to fall within the range of the position of the second speaker calculated in the arrangement calculation step. The volume of the first speaker reaching the first person is within a predetermined range, and the volume of the second speaker reaching the second person is within a predetermined range. like Wherein the executing and volume compensation step of adjusting the volume of said first and second speakers, to the computer.

  In these speaker control method and speaker control program, the positions of the first and second persons are detected in the person recognition step. Based on the detected position, an area where the second speaker can exist is calculated in the arrangement calculating step. Then, the position of the second speaker is changed in the operation step so as to fall within the possible area. Therefore, the sound emitted from the speaker can be automatically controlled so as to be appropriate for both the normal hearing person (first person) and the hard of hearing person (second person).

  ADVANTAGE OF THE INVENTION According to this invention, the speaker control apparatus which can control automatically the sound which a speaker emits so that it may become suitable for both a normal hearing person and a hard-of-hearing person, a robot provided with the same, a speaker control method, and a speaker control program Is realized.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same reference numerals are assigned to the same elements, and duplicate descriptions are omitted.

  FIG. 1 is a block diagram showing an embodiment of a speaker control apparatus according to the present invention. This speaker control apparatus is an apparatus that controls the first speaker 101 and the second speaker 102, and includes a person recognition unit 201 that detects the position of the first person and the position of the second person, and person recognition. Based on the positions of the first and second persons detected by the means 201, the second person enters the second speaker 102 within the directivity range and the second person does not enter the second person 102. The position calculation means 301 for calculating the position range (possible area) of the second speaker 102 and the second speaker 102 so as to fall within the position range of the second speaker 102 calculated by the position calculation means 301. The volume of the operation means 401 for changing the position and the volume of the first speaker 101 reaching the first person is within a predetermined range, and the volume of the second speaker 102 reaching the second person is within the predetermined range. To be inside , And a volume compensating means 501 for adjusting the volume of the first speaker 101 and the second speaker 102, a. In this embodiment, the second person is a hearing impaired person, and the first person is the caregiver (normal hearing person).

  FIG. 2 is a front view illustrating an appearance of a robot including the speaker control device of the present embodiment. FIG. 3 is a block diagram showing an electrical configuration of the robot shown in FIG. In the present embodiment, the robot is configured by connecting, for example, the body 1 and the head 2. The body portion 1 has a cylindrical shape, and a plane is arranged vertically. Wheels 3A and 3B are attached to the lower part of the body part 1 on the left and right, respectively, and these wheels can rotate back and forth independently. The head 2 can rotate within a predetermined range with respect to a vertical axis that is vertically attached to the body 1 and a horizontal axis that is installed at an angle of 90 degrees with respect to the vertical axis. The vertical axis is installed so as to pass through the center of the head 2, and the horizontal axis is installed horizontally through the center of the head 2 with the body 1 and the head 2 facing the front and horizontally in the left-right direction. Yes. That is, the head 2 can rotate within a predetermined range with two degrees of freedom, left and right and up and down.

  The body unit 1 houses a controller 10 for controlling the entire robot, a battery 11 serving as a power source for the robot, a speaker 12, actuators 14A and 14B for moving two wheels, and a touch sensor 15.

  The head 2 houses a microphone 13, a speaker 16, CCD cameras 21A and 21B, actuators 22A and 22B for rotating the head 2, and the like. The microphone 13 in the head 2 collects surrounding sounds including utterances from the user, and sends the obtained sound signals to the controller 10. The speaker 16 is a directional speaker. In the present embodiment, the directivity range of the speaker 16 is fixed. The CCD cameras 21 </ b> A and 21 </ b> B take an image of the surrounding situation and send the obtained image signal to the controller 10.

  The touch sensor 15 in the body part 1 detects that a human has touched the touch sensor, and sends a contact detection signal to the controller 10 when it is detected.

  The controller 10 includes a CPU 10A and a memory 10B. The CPU 10A executes various processes by executing a control program stored in the memory 10B.

  That is, the controller 10 determines a surrounding situation and a command from the user based on the sound signal, the image signal, and the contact detection signal given from the microphone 13, the CCD cameras 21A and 21B, and the touch sensor 15.

  Further, the controller 10 determines a subsequent action based on the determination result and drives the necessary actuators 14A, 14B, 22A, and 22B based on the determination result. As a result, actions such as rotating the head 2 up and down, left and right, and moving or rotating the robot are performed. Moreover, the controller 10 produces | generates a synthetic | combination sound as needed, supplies it to the speaker 12 and the speaker 16, and outputs them.

  As described above, the robot takes action based on the surrounding situation.

  Next, FIG. 4 is a block diagram illustrating a functional configuration example of the controller 10 of FIG. The functional configuration shown in FIG. 4 is realized by the CPU 10A executing a control program stored in the memory 10B.

  The controller 10 includes a person recognition unit 51A, a voice recognition unit 51B, a contact recognition unit 51C, an action determination unit 52, a control unit 53, a voice synthesis unit 54, an output unit 55, and an arrangement calculation unit 56.

  The person recognizing unit 51A performs image processing on images input from the CCD cameras 21A and 21B, identifies a person by collating with a pre-registered feature, and identifies the identified person as a normal hearing person and a deaf person. The information is classified and sent to the arrangement calculation unit 56. Thus, in the present embodiment, it is assumed that the registered person is stored as a normal hearing person or a hearing impaired person. This is done by a person specifying at the time of registration. The person recognizing unit 51A uses the characteristic that the input is a stereo camera, calculates the position information of the person, and sends it to the arrangement calculating unit 56.

  The voice recognition unit 51 </ b> B recognizes the human voice input from the microphone 13 and sends the recognized result to the action determination unit 52. As a result of recognition and is, such as "Good morning", "Hello", the person is a word that spoke to the robot. When the behavior determination unit 52 receives the recognition result from the speech recognition unit 51B, the behavior determination unit 52 sends an instruction to the control unit 53 and the speech synthesis unit 54 to make a response according to the recognition result. For example, if the recognition result is “good morning”, the control unit 53 is instructed to shake the head 2 up and down, and the voice synthesis unit 54 is instructed to utter “good morning”. . As a result, when a person speaks “good morning” to the robot and the robot recognizes the word, the robot can utter “good morning” while shaking the head 2 up and down.

  When the person touches the touch sensor, the touch sensor 15 sends the detection information to the contact recognition unit 51C. The contact recognition unit 51C sends the contact recognition information to the action determination unit 52. When the action determination unit 52 receives the contact recognition information, the behavior determination unit 52 sends an instruction to the control unit 53 and the voice synthesis unit 54 to make a response according to the recognition result. For example, to express joy for being touched, an instruction to shake the head 2 from side to side is sent to the control unit 53, and the word “Wai” is given to the speech synthesis unit 54. The content is such that an instruction to speak is sent.

  The control unit 53 generates a control signal for driving the actuators 14A, 14B, 22A, and 22B based on the operation data given from the behavior determining unit 52, and sends this to the actuators 14A, 14B, 22A, and 22B. . As a result, the actuators 14A, 14B, 22A, 22B are driven according to the control signal, and the robot operates.

  When the position information of the normal hearing person and the deaf person is sent from the person recognizing part 51A, the arrangement calculating unit 56 enters the range of directivity of the speaker 16, and the normal person hears the directivity of the speaker 16. Calculate and determine the position of the robot itself that deviates from the range. In the present embodiment, the directivity range of the speaker 16 is outside the directivity range outside the range of 30 degrees left and right from the front of the speaker. As described in Patent Document 3, since the high frequency sound of 2 KHz or more is attenuated by about 15 dB or more at 30 degrees from the front, the outside of the range of 30 degrees from the front of the speaker is outside the directivity range. This is considered reasonable. The angle indicating the outside of the directivity range is generalized and expressed as θ0. In this example, θ0 = 30 (deg).

  The arrangement of the robot, the hearing impaired person, and the normal hearing person is shown in FIG. In FIG. 5, the point R indicates the position of the robot, the point A (the origin of the XY coordinates) indicates the position of the hearing impaired person, and the point B (X = d, Y = 0) indicates the position of the normal hearing person. The distance between the hearing impaired person and the normal hearing person is d (m). Here, the angle between the X axis and RA is φ, and the angle between RA and RB is θ. Actually, there are cases where the position of the hearing impaired person and the normal hearing person are reversed when viewed from the robot. In that case, the X-axis is reversed and the following discussion can be applied similarly. .

  In this embodiment, the robot points the front of the speaker 16 toward the hearing impaired person. In FIG. 5, the direction of the speaker 16 is indicated by an arrow. This is because, although it is a directional speaker, the sound intensity becomes weak when it is displaced from the front, and therefore it is most desirable to face the front.

  Then, at this time, if the angle θ is larger than θ 0, in the arrangement shown in FIG. 5, the hearing impaired person is located in front of the directional speaker 16, and the normal hearing person is outside the directivity range of the directional speaker 16. Will do.

Consider the following. θ = θ0 and a constant value. At this time, since the length of RA is a function of φ, it is set as r (φ). Then, the function form of r (φ) is expressed by the following equation.
r (φ) = d × sin (θ0 + φ) / sin (θ0) (1)

  It can be easily seen that r (φ) represented by the equation (1) is the maximum distance between the robot and the deaf person. Therefore, in order for the normal hearing person (B) to be out of the directivity range, the robot (R) must be located at a position closer to the distance r (φ) from the hearing impaired person (A).

  Here are some specific numbers. When d = 0.5 (m), φ = 90 (deg), that is, r (φ) = 0.866 (m) when the robot is located in front of the hearing impaired person. Further, when looking at a place shifted by 30 degrees therefrom, r (φ) = 0.5 (m) when φ = 120 (deg), r (φ) = 1. 0 (m).

  When performing voice communication with two humans separated by a distance d (m), it is considered that 60 (deg) ≦ φ ≦ 120 (deg) is appropriate as the positioning of the robot. Therefore, when the placement calculation unit 56 finds a hearing impaired person and a normal hearing person, a virtual line that connects the two people is considered, and when the robot faces the hearing impaired person, The robot is moved to a position where the angle with the virtual straight line is in the range of 60 degrees to 120 degrees. Then, the robot may be moved to a position where the distance between the robot and the deaf person is equal to or less than the value represented by the expression (1).

  FIG. 6 shows the position of the robot. It is desirable that a robot exists inside the area indicated by “Acde” in FIG. 6 (a hearing person B is outside the directional area of the directional speaker 16 while capturing the deaf person A in front of the directional speaker 16). It is a range. However, the robot (directional speaker 16) faces the hearing impaired person (A).

  The distance between the robot and the deaf person is specific. In the present embodiment, when the robot detects a hearing-impaired person and a normal hearing person, the placement calculation unit 56 instructs the action determination unit 52 so that the robot moves to the position d when the robot is outside the region Acde. The action determination unit 52 sends an instruction to the control unit 53 to operate the actuators 14A and 14B and move the robot to the position d. Further, when the robot detects the deaf and hard-of-hearing people and the robot is present in the area Acde or on the boundary, the robot is rotated on the spot so that the front of the robot and the directional speaker 16 faces the deaf person. Rotate the robot.

  In addition, the arrangement calculation unit 56 sends the distance between the robot and the hearing impaired person and the distance between the robot and the normal hearing person to the action determination unit 52, and the action determination unit sends the distance information to the speech synthesis unit 54.

  The voice synthesizer 54 generates a synthesized voice from the text given from the action determination unit 52 and supplies the synthesized voice to the output unit 55. At that time, the distance to the hearing impaired person and the distance to the normal hearing person are referred to. That is, there are a reference volume and a reference distance, and an appropriate volume is calculated from the reference volume according to the distance to the actual person and output. Here, in this embodiment, when the reference volume is D0, the reference distance is L0, and the distance to the person is R, the volume is calculated as D0 × R × R / (L0 × L0). In this way, appropriate sound volumes are calculated with reference to the distance to the hearing impaired person and the distance to the normal hearing person, and each synthesized sound is generated and sent to the output unit 55. However, the sound output from the speaker 16 is a sound obtained by emphasizing a high range of 2 kHz or higher from 10 dB to 15 dB.

  The output unit 55 is supplied with digital data of two synthesized sounds from the voice synthesis unit 54. The output unit 55 performs D / A conversion of the digital data into an analog voice signal, and The signals are supplied to and output from the speakers 12 and 16.

  In this embodiment, image recognition is used as means for recognizing a person and calculating a distance to the person. However, other methods such as an ultrasonic tag technique may be used. A laser range finder may be used as a method for calculating the distance to a person.

  The angle indicating the directivity range is 30 degrees, but this should be changed in accordance with the actual performance of the directional speaker 16.

  Further, although the directivity range of the speaker 16 is fixed, this is not limited thereto. That is, the directivity range of the speaker 16 may be variable. In this case, since the directivity range can be changed, the position where the speaker 16 can be present is wider than in the present embodiment.

  In the present embodiment, it is assumed that the directional speaker 16 is directed toward the hearing impaired person. However, this may be such that the hearing impaired person comes within the directivity range of the directional speaker.

  Although 60 (deg) ≦ φ ≦ 120 (deg) is set as a preferable region where the robot is present, the present invention is not limited to this.

  Further, when the robot detects the hearing impaired person and the normal hearing person, the robot moves to the position d when the robot is outside the area Acde. However, this is not limited to this. You may move anywhere as long as it is inside the area.

  A person may move after the robot has moved to an appropriate location. The arrangement calculation unit 56 always grasps the position of the person, and when the robot arrangement is inappropriate, sends an instruction to the action determination unit 52 so that the robot arrangement is appropriate.

  In the speech synthesizer 54, the high frequency of 2 kHz or higher is emphasized from 10 dB to 15 dB for the sound emitted from the directional speaker, but the range to be emphasized is not limited to 2 kHz or higher. Further, the degree of emphasis is not limited to 10 dB to 15 dB, and it is possible to make the degree of emphasis different for each frequency band, and the degree of emphasis differs for each individual deaf person (deaf person). You can also

  Further, since the normal sound that reaches from the speaker 12 reaches the hearing-impaired person, the sound synthesizer 54 may emit a sound obtained by subtracting the amount that reaches from the speaker 12 as the sound emitted from the speaker 16.

  Further, the speech synthesizer 54 may not only determine the volume according to the distance to the person but also move the robot and determine the volume according to the distance from the moved position to the person. In that case, the speech synthesizer 54 sends a movement command to the controller 53. However, the range to be moved is the range calculated by the arrangement calculation unit 56. For example, when it is desirable to approach the position determined by the arrangement calculation unit 56, such a method can be considered. When the maximum limit volume of the speaker is small, the robot must approach the person in this way.

  Furthermore, in the present embodiment, the form in which the speaker is mounted on the robot has been described. However, the present invention can also be applied to the case where the speaker 12 is placed in a room and the speaker 16 is mounted on the robot. Furthermore, the present invention can be applied not only to a robot but also to a system that can change the position of a speaker.

  In the present embodiment, each of the speaker 12 and the speaker 16 is a single speaker. However, in actuality, the speaker 12 and the speaker 16 may each be composed of a plurality of speakers.

  Next, the overall operation of the present embodiment will be described in detail with reference to the flowchart of FIG.

  In step S1, the robot detects a hearing impaired person and a normal hearing person. In step S2, it is determined whether or not the robot is properly arranged for the hearing impaired person and the normal hearing person. If it is determined in step S2 that the robot is not properly arranged for the hearing impaired person and the normal hearing person, the robot is moved to an appropriate position in step S3. If it is determined in step S2 that the robot is properly positioned with respect to the hearing impaired person and the normal hearing person, in step S4, it is waited whether the touch sensor is touched or voice recognition is performed. When the touch sensor is touched or voice recognition is performed, the robot responds in step S5.

  Next, the effect of this embodiment will be described. In the present embodiment, it is possible to output a sound that emphasizes high frequencies for the hearing impaired to the hearing impaired and to output a normal sound for the hearing impaired. Voice communication can be suitably performed.

  The robot may have a function of recording a person's video and sending the person's video to an external terminal when requested via the Internet or the like. In the present embodiment, the robot is equipped with a microphone, but a form using a wireless microphone is also conceivable. The robot may include an infrared sensor and a function of checking the body temperature of the hearing impaired person. The robot may have a map generation function and a self-position identification function. The robot may include a plurality of touch sensors.

  As described above, in the speaker control apparatus according to the present embodiment, the volume compensation unit 501 (see FIG. 1) calculates the position of the first speaker 101, outputs the position to the operation unit 401, and calculates by the arrangement calculation unit 301. The position of the second speaker 102 is calculated within the range of the position of the second speaker 102 and is output to the operating means 401, and the volume of the first speaker 101 and the volume of the second speaker 102 are The operation unit 401 may move the first speaker 101 and the second speaker 102 according to the information regarding the positions of the first speaker 101 and the second speaker 102 given from the volume compensation unit 501. .

  When calculating the range of the position of the second speaker 102 from the position of the first person and the position of the second person detected by the person recognizing means 201, the arrangement calculating unit 301 On the assumption that the front faces the direction of the second person and the second speaker 102 exists in a predetermined angle range from the front direction of the second person to the left and right, the first person is the second speaker. A position range of the second speaker 102 that does not fall within the directivity range of 102 may be calculated.

  The directionality of the directivity of the second speaker 102 is variable and the range of directivity is variable, and the arrangement calculation unit 301 includes a second person within the directivity range of the second speaker 102 and the first person. The position of the second speaker 102, the direction of directivity, and the range of directivity may be calculated such that no other person enters.

  The speaker control method of the present embodiment is a method for controlling the first speaker 101 and the second speaker 102, and a person recognition step for detecting the position of the first person and the position of the second person. Based on the positions of the first and second people detected in the person recognition step, the second person enters the directivity range of the second speaker 102 and the first person does not enter. A position calculating step for calculating the range of the position of the second speaker 102, and the position of the second speaker 102 is changed so as to fall within the range of the position of the second speaker 102 calculated in the position calculating step. The operation step and the volume of the first speaker 101 reaching the first person are within a predetermined range, and the volume of the second speaker 102 reaching the second person is within the predetermined range. ,first And volume compensation step of adjusting the volume of the speaker 101 and the second speaker 102, is intended to include.

  The speaker control program according to the present embodiment causes the computer to execute the person recognition step, the arrangement calculation step, the operation step, and the volume compensation step.

  The present invention is not limited to the above embodiment, and various modifications can be made.

  The present invention can be widely applied to a robot that includes a voice input device such as a microphone and has a function of detecting a person and that interacts with the person. Further, in the present embodiment, the above-described series of processing is performed by causing the CPU 10A (FIG. 3) to execute a program, but the series of processing can also be performed by dedicated hardware. is there.

  The program is stored in advance in the memory 10B (FIG. 3), or temporarily or temporarily stored on a removable recording medium such as a floppy (registered trademark) disk, CD-ROM, MO disk, DVD, magnetic disk, or semiconductor memory. It can be stored permanently (recorded). Such a removable recording medium can be provided as so-called package software and installed in the robot (memory 10B).

  Further, the program can be transferred from a download site wirelessly via a digital satellite broadcasting artificial satellite or wiredly via a network such as a LAN or the Internet, and can be installed in the memory 10B. In this case, when the program is upgraded, the upgraded program can be easily installed in the memory 10B.

  Here, in the present specification, the processing steps for describing a program for causing the CPU 10A to perform various types of processing do not necessarily have to be processed in time series in the order described in the flowchart, but in parallel or individually. This includes processing to be executed. Further, the program may be processed by one CPU, or may be processed in a distributed manner by a plurality of CPUs.

It is a block diagram which shows one Embodiment of the speaker control apparatus by this invention. It is a front view which shows the external appearance of a robot provided with the speaker control apparatus of this embodiment. FIG. 3 is a block diagram illustrating an electrical configuration of the robot of FIG. 2. FIG. 4 is a block diagram illustrating a functional configuration example of a controller in FIG. 3. It is a figure for demonstrating the operation | movement of embodiment. It is a figure for demonstrating the operation | movement of embodiment. It is a flowchart for demonstrating operation | movement of embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 1st speaker 102 2nd speaker 201 Person recognition means 301 Arrangement calculation means 401 Operation means 501 Volume compensation means 1 Robot body part 2 Robot head part 3A, 3B Wheel 10 attached to robot Controller 10A CPU
10B Memory 11 Battery 12 Speaker 13 Microphone 14A, 14B Actuator 15 Touch sensor 16 Speaker 21A, 21B CCD camera 22A, 22B Actuator 51A Person recognition unit 51B Speech recognition unit 51C Contact recognition unit 52 Action determination unit 53 Control unit 54 Speech synthesis unit 55 Output unit 56 Arrangement calculation unit

Claims (7)

A device for controlling the first and second speakers,
Person recognition means for detecting the position of the first person and the position of the second person;
Based on the positions of the first and second persons detected by the person recognition means, the second person enters the directivity range of the second speaker and the first person does not enter. An arrangement calculating means for calculating the range of the position of the second speaker,
An operation means for changing the position of the second speaker so as to fall within the range of the position of the second speaker calculated by the arrangement calculating means;
The volume of the first speaker reaching the first person is within a predetermined range, and the volume of the second speaker reaching the second person is within a predetermined range. Volume compensation means for adjusting the volume of the first and second speakers;
A speaker control device comprising: The speaker control device according to claim 1,
The volume compensation means calculates the position of the first speaker and outputs the position to the operation means, and determines the position of the second speaker within the range of the position of the second speaker calculated by the arrangement calculation means. Calculate and output to the operating means, and change the volume of the first speaker and the volume of the second speaker,
The operating means moves the first and second speakers in accordance with information about the positions of the first and second speakers given from the volume compensation means. The speaker control device according to claim 1 or 2,
The arrangement calculating means calculates the second speaker position range from the position of the first person and the position of the second person detected by the person recognition means. Assuming that the front of the speaker faces the direction of the second person and that the second speaker exists within a predetermined angular range from the front direction of the second person to the left and right, the first person A speaker control device that calculates a range of the position of the second speaker such that does not fall within the directivity range of the second speaker. The speaker control device according to any one of claims 1 to 3,
The direction of directivity of the second speaker is variable and the range of directivity is variable,
The arrangement calculating means is arranged such that the position of the second speaker, the direction of the directivity and the directivity so that the second person enters the directivity range of the second speaker and the first person does not enter. Speaker control device for calculating the range of the sex. A speaker control device according to any one of claims 1 to 4;
The first and second speakers controlled by the speaker control device;
A robot characterized by comprising: A method for controlling first and second speakers, comprising:
A person recognition step for detecting the position of the first person and the position of the second person;
Based on the positions of the first and second persons detected in the person recognition step, the second person enters the directivity range of the second speaker and the first person does not enter. An arrangement calculating step for calculating the range of the position of the second speaker,
An operation step of changing the position of the second speaker so as to fall within the range of the position of the second speaker calculated in the arrangement calculating step;
The volume of the first speaker reaching the first person is within a predetermined range, and the volume of the second speaker reaching the second person is within a predetermined range. A volume compensation step for adjusting the volume of the first and second speakers;
The speaker control method characterized by including. A program for controlling the first and second speakers,
A person recognition step for detecting the position of the first person and the position of the second person;
Based on the positions of the first and second persons detected in the person recognition step, the second person enters the directivity range of the second speaker and the first person does not enter. An arrangement calculating step for calculating the range of the position of the second speaker,
An operation step of changing the position of the second speaker so as to fall within the range of the position of the second speaker calculated in the arrangement calculating step;
The volume of the first speaker reaching the first person is within a predetermined range, and the volume of the second speaker reaching the second person is within a predetermined range. A volume compensation step for adjusting the volume of the first and second speakers;
Is executed by a computer.

Images