KR101159133B1 - Android robot and method for using the android robot - Google Patents

Abstract

The present invention relates to an Android robot, comprising: a plurality of microphones, a band pass filter for extracting a voice signal of the other party from a signal collected through the plurality of microphones, and a codec for converting the extracted voice signal into a digital signal, respectively And a sound source processor for tracking the location of the other party using a phase difference and magnitude difference between the voice signals converted into the digital signals, and gesture data corresponding to the analyzed voice signal after analyzing the converted voice signal. And a host processor to control the robot driver based on the inquired gesture data and the tracked position, and a robot driver driven under the control of the host processor. According to the present invention, when the Android robot recognizes the other party's voice, the Android robot can grasp the location of the other party who makes the voice and respond appropriately to the other party's voice and location.

Android robot, sound processor, sound source tracking

Description

ANDROID ROBOT AND METHOD FOR USING THE ANDROID ROBOT}

The present invention relates to an android robot. In particular, it relates to an Android robot that grasps the other party's voice and the other party's location and responds appropriately to the other party's voice and location.

For Android robots, speech recognition is a very important factor. However, the conventional Android robot recognizes the voice by using one wired microphone, which does not have a direction to the talking object and there is a problem in which the sound cannot be confirmed.

In order to solve this problem, an android robot for tracking the direction of a sound source has been developed, but the conventional android robot is configured to extract a sound source from a sound source board directly connected to an analog processing part, and thus a lot of noise components are included in a voice signal. When amplifying many voice signals, noise components are amplified as well, and thus it is difficult to distinguish the voice signals. Accordingly, a technique for minimizing noise to derive a more accurate direction value of a sound source has been requested.

In addition, the conventional Android robot has a problem that the voice degradation is severe because the analog line is connected from the microphone input to the host processor, and the host processor handles all control such as voice and motion of the robot so that the host processor can handle the robot. When processing the motion and voice at the same time, there is a problem that the output quality is degraded in the motion and the voice due to a lot of load, a technology that solves this problem has been requested.

The present invention was created to solve the above problems, and an object of the present invention is to provide an android robot that reacts by tracking the voice of the other party and the location of the other party.

Another object of the present invention is to provide a method of using an android robot that tracks and responds to the voice of the other party and the location of the other party.

It is still another object of the present invention to provide a computer readable medium recording a program for performing a method of using an android robot that tracks and reacts with a voice of a counterpart and a counterpart's location.

Android robot according to the present invention for achieving the above object is a plurality of microphones, a band pass filter for extracting the voice signal of the other party from the signals collected through the plurality of microphones, respectively, and the extracted voice signal digital signal A sound source processor for tracking the location of the other party using a codec for converting each signal to a digital signal, a phase difference and a magnitude difference between the voice signal converted into the digital signal, and analyzing the voice signal to gesture gesture data corresponding to the voice signal. And a host processor for inquiring from a DB, controlling the robot driver based on the inquired gesture data and the tracked position, and a robot driver driven under the control of the host processor. Here, the sound source processor receives the inquired gesture data from the host processor and inquires about the robot voice data mapped to the gesture data in a voice DB, and controls to output the inquired robot voice data through a speaker. do. The band pass filter extracts a signal of a predetermined frequency band centered on a frequency of 1 kHz from a signal collected through the plurality of microphones, and has a maximum gain at the frequency of 1 kHz. The sound source processor may transmit the voice signal to the host processor through a controller area network (CAN). In addition, the microphone, at least three are provided to track the three-dimensional position of the other party, may be provided on both the left and right ends of the head and the crown. In addition, the codec and the sound source processor may be provided inside the head.

According to another aspect of the present invention, there is provided a method of using Android, wherein a plurality of microphones are collected; Extracting, by a band pass filter, a voice signal of the other party from the signals collected through the plurality of microphones; Converting each of the extracted voice signals into a digital signal by a codec; Tracking, by a sound source processor, the position of the counterpart using a phase difference and a magnitude difference between the voice signals converted into the digital signals; A sound source processor transmitting the converted voice signal and the tracked position information to a host processor; Receiving, by the host processor, the voice signal and the location information; Analyzing, by the host processor, the received voice signal and querying gesture data corresponding to the analyzed voice signal in a gesture DB; And the host processor driving the robot using the inquired gesture data and the received position information, and outputting the robot voice data mapped to the inquired gesture data by the sound source processor. .

A computer-readable medium having recorded thereon a program for executing a method of using an android robot according to the present invention for achieving the another object comprises the steps of: collecting a plurality of microphones; Extracting, by a band pass filter, a voice signal of the other party from the signals collected through the plurality of microphones; Converting each of the extracted voice signals into a digital signal by a codec; Tracking, by a sound source processor, the direction and position of the counterpart using a phase difference and a magnitude difference between the voice signals converted into the digital signals, respectively; A sound source processor transmitting the converted voice signal and information about the tracked direction and position to a host processor; Receiving, by the host processor, the voice signal and information about the direction and location; Analyzing, by the host processor, the received voice signal, and querying gesture data corresponding to the analyzed voice signal in a gesture DB; A program for the host processor to drive the robot using the inquired gesture data and the received position information, and the sound source processor to output a robot voice data corresponding to the inquired gesture data to be recorded. Can be.

According to the Android robot according to the present invention, when recognizing the voice of the other party, it is possible to grasp the location of the other party who makes a voice instead of simply recognizing the voice and to perform an appropriate response corresponding to the voice and the location of the other party. As a result, Android robots can communicate more naturally with humans.

In addition, according to the Android robot according to the present invention, the microphone, the codec, and the sound processor are all embedded in the head of the robot, thereby minimizing the analog line, and as a result, it is possible to minimize the noise portion in the analog signal collected from the microphone.

In addition, according to the Android robot according to the present invention, by converting the voice signal collected from the microphone into a digital signal to be transmitted to the sound source processor, it is possible to reduce the analog communication distance by improving the voice degradation phenomenon compared to the prior art.

In addition, according to the Android robot according to the present invention, the motion and gesture corresponding to the other party's voice is controlled by the host processor, and the robot voice corresponding to the other party's voice is separated to be controlled by the sound source processor, thereby controlling all the host processors. Compared with the prior art, it is possible to improve the motion and voice quality of the robot.

Since the present invention can be variously modified and can have various embodiments, specific embodiments will be illustrated in the drawings and described in detail. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all changes, equivalents, and substitutes included in the spirit and scope of the present invention. In describing the drawings, the same reference numerals are used for the same components.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. The term “and / or” includes any combination of a plurality of related description items or any of a plurality of related description items.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, the terms "comprise" or "have" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art. Terms, such as those defined in the commonly used dictionaries, should be construed as having meanings consistent with the meanings in the context of the related art, and shall not be construed in ideal or excessively formal meanings unless expressly defined in this application. .

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram showing the configuration of an Android robot according to the present invention.

2 is a view showing the inside of the head of the Android robot according to the present invention.

3 to 5 are views showing the positions of a plurality of microphones installed in the android robot according to the present invention.

Referring to FIG. 1, an android robot according to the present invention extracts a voice signal of a counterpart from a plurality of microphones 101a, 101b, and 101c and signals collected through the plurality of microphones 101a, 101b, and 101c, respectively. Tracking the position of the other party using a phase difference and magnitude difference between a band pass filter 102, a codec 103 for converting the extracted speech signal into a digital signal, and a speech signal converted into the digital signal, respectively. A host that analyzes the voice signal and analyzes the voice signal to query gesture data corresponding to the voice signal in a gesture DB (not shown), and controls the robot driver based on the inquired gesture data and the tracked position. A processor 105 and a robot driver 106 driven under the control of the host processor.

In addition, the sound source processor 104 receives the inquired gesture data from the host processor 105 and inquires robot voice data mapped to the gesture data in a robot voice DB (not shown), and the inquiry The robot voice data may be controlled to be output through the speaker 107.

Here, the band pass filter 103 extracts a signal of a predetermined frequency band centered on a frequency of 1 kHz from a signal collected through the plurality of microphones, and has a maximum gain at the frequency of 1 kHz.

Also, the sound source processor 104 may transmit the converted voice signal and the tracked location information to the host processor 105 through a controller area network (CAN).

Here, at least three microphones 101a, 101b, and 101c are provided to track the three-dimensional position of the other party, and as shown in FIG. 5, the microphones 101a, 101b, and 101c may be provided at both ends of the head and the parietal.

The codec and the sound source processor may be provided inside the head to minimize an analog communication distance.

1 to 5, each component of the Android robot according to the present invention will be described in detail.

The plurality of microphones 101a, 101b, and 101c collect signals (sounds) from sound sources (relatives) in arbitrary directions. Here, at least three microphones are provided to track the three-dimensional position of the other party. When estimating the position of a sound source by measuring the sound source arrival delay time using the geometric shape of the microphone, it is necessary to use at least three microphones to track the three-dimensional position, that is, to prevent front-back confusion. This is because it is preferable. In addition, when three microphones are provided, the microphones may be provided at both left and right ends and the crown of the head of the Android robot, as shown in FIG. 4. The distance (d1) between the microphone 101a installed at the head parietal of the android robot and the microphone 101b installed at the left end of the head of the android robot, the microphone 101a installed at the head parietal of the android robot and the head of the android robot The distance d2 at which the microphone 101c installed at the right end is separated from each other and the distance d3 at which the microphone 101b installed at the left end of the head of the Android robot and the microphone 101c installed at the right end of the Android robot are different may be different. It is preferable to install the microphones so that the distances d1, d2, d3 are equal.

The band pass filter 102 extracts a signal of a predetermined frequency band centered on a frequency of 1 kHz from a signal collected through the plurality of microphones 101a, 101b, and 101c, but at a maximum of the frequency of 1 kHz. Have a gain. Since the signals collected by the plurality of microphones 101a, 101b, and 101c include not only the voice of the other party, but also various noises, the signals collected by the plurality of microphones 101a, 101b, and 101c to minimize noise components. Are each filtered by the band pass filter 102. In addition, since 1 kHz is the frequency band closest to a human voice, the band pass filter 102 extracts a signal of a predetermined frequency band centered on a frequency of 1 kHz, but obtains a maximum gain at the frequency of 1 kHz. In the remaining frequencies, the gain is minimized to optimally recognize the other party's voice.

The codec 103 converts a plurality of audio signals (three or more) extracted by the band pass filter 102 into digital signals, respectively. In addition, the codec 103 transmits the converted voice signal to the sound source processor 104. The codec 103 converts the voice signal collected from the microphone into a digital signal and transmits the voice signal to the sound source processor, thereby reducing the analog communication distance and improving voice degradation compared to the prior art.

The sound source processor 104 tracks the position of the counterpart by using a phase difference and a magnitude difference between the voice signals converted into the digital signals, respectively. Specifically, the digitally converted voice signal is converted into a frequency domain, and a delay time between the voice signals converted into the frequency domain is obtained. Specifically, between the voice signal of the microphone 101a installed on the head of the Android robot (voice signal of the first microphone) and the voice signal of the microphone 101b installed on the left end of the head of the Android robot (voice signal of the microphone 2) Between the voice signal of the microphone 101a installed on the head of the Android robot (voice signal of the microphone 1) and the voice signal of the microphone 101c installed on the right end of the head of the Android robot (voice signal of the microphone 3), Generalization between the voice signal of the microphone 101b installed at the left end of the head of the android robot (voice signal of the microphone # 2) and the voice signal of the microphone 101c installed at the right end of the head of the android robot (voice signal of the third microphone). The peak point is found by using the cross correlation function. At this time, if the delay time value is positive, the signal arrives before the reference microphone, and if negative, the reference microphone arrives first. Subsequently, the candidate tracking angles are obtained after obtaining several tracking angles which are candidates based on a geometric method directly without the conventional empirical section selection method using the delay time calculation values calculated by the delay time calculation unit 420. Two tracking angles which are calculated to be the closest to the other's direction are calculated and averaged, and the resulting tracking angle is estimated to finally estimate the opponent's position.

In addition, the sound source processor 104 transmits the converted voice signal and the tracked position information to the host processor 105 through a controller area network (CAN). Here, a controller area network (CAN) refers to a serial bus network for a microcontroller that connects peripheral devices such as sensors or starters in a real-time control application system. In CAN, addressing concepts such as those used in Ethernet are not used, and messages are simultaneously sprayed on all nodes in the network using unique identifiers in that network. Individual nodes decide whether or not to process the message based on this identifier. The bus access order also prioritizes messages according to the competition principle. Unlike Ethernet, which stops transmission when a collision is detected, the advantage of using a CAN-like approach is that it allows for uninterrupted data transmission.

The host processor 105 receives the converted voice signal and the tracked position information from the sound source processor 104, and analyzes the voice signal to obtain gesture data corresponding to the voice signal in a gesture DB (not shown). And query the robot driver based on the inquired gesture data and the tracked position. Specifically, the host processor 105 includes a predicted speech DB (not shown) that stores predictable speech data of a plurality of counterparts and a gesture DB (not shown) which stores gesture data mapped to the plurality of predictable speech data. The host processor 105 analyzes the voice signal received from the sound source processor 104 for each word, extracts a keyword, and extracts the keyword, and includes the most of the keywords included in the voice signal. The predictable speech data is searched in the predicted speech DB (not shown), and the gesture data mapped with the searched predictable speech data is searched in the gesture DB (not shown) and output.

The robot driver 106 is driven under the control of the host processor 105. Specifically, the robot driver 106 performs an operation corresponding to the inquired gesture data under the control of the host processor 105, and drives the robot driver to rotate toward the tracked position of the counterpart.

In addition, the Android robot according to the present invention in order to output the voice of the robot in response to the voice of the other party, the sound source processor 104 (specifically, the robot voice processing unit 320 of the sound source processor 104), the host The controller 105 receives the inquired gesture data and inquires about the robot voice data mapped to the gesture data in a robot voice DB (not shown), and controls the robot voice data to be output through a speaker. In detail, the sound source processor 104 receives the inquired gesture data from the host processor 105 and inquires about the robot voice data mapped to the gesture data in a robot voice DB (not shown), and the inquired robot Voice data is transmitted to the codec 103. The codec 103 receives the inquired robot voice data and converts the received robot voice data into an analog signal. The voice data of the robot converted into an analog signal is output through the speaker 107. Here, the inquired gesture data is transmitted from the host processor 105 to the sound source processor 104 through a controller area network (CAN).

As shown in FIG. 2, the codec 103 and the sound source processor 104 are preferably provided inside the head. This is to minimize the noise by minimizing the distance between the microphone (101a, 101b, 101c) and the speaker 107 installed in the head of the robot.

6 is a flowchart illustrating a method of using an android robot according to the present invention.

Referring to FIG. 6, in the method of using an android robot according to the present invention, first, a plurality of microphones 101a, 101b, and 101c are collected at step S601. Here, the plurality of microphones 101a, 101b, and 101c are embedded in an Android robot, and it is preferable to collect at least three microphones to track the three-dimensional position of the opponent. When the sound is collected by three microphones, the sound may be collected by microphones provided at both left and right ends of the head and the crown.

Next, the band pass filter 102 extracts the voice signal of the other party from the signals collected through the plurality of microphones 101a, 101b, and 101c, respectively (S602). Specifically, the band pass filter 102 extracts a signal of a predetermined frequency band centered on a frequency of 1 kHz from a signal collected through the plurality of microphones 101a, 101b, and 101c, but at a frequency of 1 kHz. Try to get the maximum gain.

Next, the codec 103 converts the extracted voice signals into digital signals, respectively (S603).

Next, the sound source processor 104 tracks the position of the counterpart using the phase difference and the magnitude difference between the voice signals converted into the digital signals (S604). In detail, the sound source processor 104 converts the digitally converted voice signal into a frequency domain, calculates a delay time between voice signals converted into the frequency domain received by each microphone, and uses the calculated delay time values. After obtaining several tracking angles which are candidates based on the geometric method directly without the conventional empirical section selection method, two tracking angles which are calculated as the closest to the opponent's direction among these candidate tracking angles are calculated and averaged. By taking the resulting tracking angle and finally estimating the opponent's position.

Next, the sound source processor 104 transmits the converted voice signal and the tracked position information to the host processor 105 (S605). In detail, the sound source processor 104 transmits the converted voice signal to the host processor 105 through a controller area network (CAN).

Next, the host processor 105 receives the voice signal and the location information (S606).

Next, the host processor 105 analyzes the received voice signal and inquires gesture data corresponding to the analyzed voice signal in a gesture DB (not shown) (S607). Specifically, the embedded voice DB (not shown) storing predictable voice data of a plurality of counterparts and a gesture DB (not shown) storing gesture data mapped to the plurality of predictable voice data are embedded. The host processor 105 analyzes the voice signal received from the sound source processor 104 for each word, extracts a keyword, and predicts the predictable voice data of the other party that includes the keyword most included in the voice signal. Searched in the voice DB (not shown), the gesture data mapped to the searchable predictable voice data is searched in the gesture DB (not shown) and output.

Next, the host processor 105 drives the robot using the inquired gesture data and the received position information, and the sound source processor 104 outputs the robot voice data corresponding to the inquired gesture data. (S608).

In operation S608, in detail, the host processor uses the inquired gesture data and the received position information to enable the Android robot to properly operate the gesture corresponding to the voice of the other party and the location of the other party. And controlling the robot driver to be driven under the control of the host processor.

In operation S608, the host processor transmits the inquired gesture data to the sound source processor so that the android robot outputs a voice corresponding to the other party's voice, and the sound source processor receives the gesture data. And querying the robot voice data mapped to the received gesture data in a voice DB (not shown), and outputting the retrieved robot voice data through a speaker. Specifically, the host processor transmits the inquired gesture data to the sound source processor through a controller area network (CAN), and the sound source processor 104 receives the inquired gesture data from the host processor 105. The robot voice data mapped to the gesture data is inquired from a robot voice DB (not shown), and the robot voice data is transmitted to the codec 103. The codec 103 receives the inquired robot voice data, converts the received robot voice data into an analog signal, and the voice data of the robot converted into the analog signal is output through the speaker 107.

The computer-readable medium which records the program which performs the method of using the Android robot which concerns on this invention is the step (S601) which the several microphone 101a, 101b, 101c collects, and the band pass filter 102 makes the said multiple Extracting the other party's voice signal from the signals collected through the microphones 101a, 101b, and 101c (S602), and the codec 103 converts the extracted voice signals into digital signals, respectively (S603). (S604), and the sound source processor 104 tracks the position of the other party using the phase difference and magnitude difference between the voice signals converted into the digital signal (S604), and the sound source processor 104 converts the converted voice signal. And transmitting the tracked location information to the host processor 105 (S605), receiving, by the host processor 105, the voice signal and the location information (S606), and the host program. The parser 105 analyzes the meaning of the received speech signal and inquires gesture data corresponding to the analyzed speech signal in a gesture DB (not shown) (S607), and the host processor 105 inquires the query. A program for driving the robot using the received gesture data and the received position information, and executing the step (S608) of the sound source processor 104 outputting the robot voice data corresponding to the inquired gesture data. .

Here, the step (S601) of the plurality of microphones (101a, 101b, 101c) is collected by the at least three microphones to track the three-dimensional position of the other party, when the three microphones are collected, It can be picked up through the microphone provided at both ends and the crown.

Here, the step of extracting the voice signal of the other party from the signal collected by the plurality of microphones 101a, 101b, 101c by the band pass filter 102, the band pass filter 102 is the plurality of microphones A signal of a predetermined frequency band centered on the frequency of 1 kHz is extracted from the signal collected through the signal, and has a maximum gain at the frequency of 1 kHz.

The transmitting of the converted voice signal and the tracked position information to the host processor 105 may include transmitting the converted voice signal to the controller area network by the sound source processor 104. ) To the host processor 105.

Here, the host processor 105 drives the robot using the inquired gesture data and the received position information, and the sound source processor 105 outputs the robot voice data corresponding to the inquired gesture data. In operation S608, the host processor 105 uses the inquired gesture data and the received location information in order for the android robot to properly operate the gesture corresponding to the voice of the other party and the location of the other party. Controlling the driver 106, and driving the robot driver 106 under the control of the host processor 105.

In addition, the host processor 105 drives the robot using the inquired gesture data and the received position information, and the sound source processor 104 outputs the robot voice data corresponding to the inquired gesture data. In operation S608, the host processor 105 transmits the inquired gesture data to the sound source processor 104 so that the android robot outputs a voice corresponding to the counterpart voice. 104, the gesture data is received, and the robot voice data corresponding to the received gesture data inquiries in a voice DB (not shown), the sound source processor 104 is the speaker 107 to retrieve the inquired robot voice data And outputting through).

As described above, according to the present invention, when recognizing the other party's voice, it is possible to grasp the location of the other party making the voice and not to simply recognize the voice, and to respond appropriately to the other party's voice and the location. As a result, Android robots can communicate more naturally with humans.

In addition, according to the Android robot according to the present invention as described above, by minimizing the analog line by embedding the microphone, codec, and sound processor in the head of the robot, the noise portion in the analog signal collected from the microphone can be minimized as a result. have.

In addition, according to the present invention as described above, by reducing the analog communication distance by converting the voice signal collected from the microphone into a digital signal to the sound source processor it is possible to improve the voice degradation phenomenon compared to the prior art.

In addition, as described above, according to the present invention, the motion and gesture corresponding to the other party's voice are controlled by the host processor, and the robot voice corresponding to the other party's voice is separated to be controlled by the sound source processor. Compared with the prior art, it is possible to improve the motion and voice quality of the robot.

1 is a block diagram showing the configuration of an Android robot according to the present invention.

2 is a view showing the inside of the head of the Android robot according to the present invention.

3 to 5 are views showing the position of a plurality of microphones installed in the android robot according to the present invention, Figure 3 is a view showing the front of the android robot according to the invention, Figure 4 is an android according to the present invention It is a figure which shows the right side of a robot, and FIG. 5 is a figure which shows the upper side of the android robot concerning this invention.

6 is a flowchart illustrating a method of using an android robot according to the present invention.

[Description of Drawing Symbol]

101a, 101b, 101c Microphone 102 Bandpass Filter 103 Codec

104 Sound Processor 105 Host Processor 106 Robot Drive

107 speakers

Claims (9)

Three microphones arranged to be directional with respect to the sound source, A band pass filter for extracting a voice signal of the other party from the signals collected through the plurality of microphones; A codec for converting the extracted voice signal into a digital signal, respectively; A sound source processor for tracking the position of the counterpart using a phase difference and a magnitude difference between the voice signals converted into the digital signals, respectively; A host processor that analyzes the converted voice signal, inquires gesture data corresponding to the analyzed voice signal in a gesture DB, and controls a robot driver based on the retrieved gesture data and the tracked position; A robot driver driven under the control of the host processor, The band pass filter extracts a signal of a predetermined frequency band centered on a frequency of 1 kHz from a signal collected through the three microphones, and has an maximum gain at the frequency of 1 kHz. The method of claim 1, The sound source processor receives the inquired gesture data from the host processor, inquires about the robot voice data mapped to the gesture data in a robot voice DB, and controls the robot voice data to be output through a speaker. Android robot. delete The method of claim 1, And the sound source processor transmits the converted voice signal and the tracked position information to the host processor through a controller area network (CAN). delete The method of claim 1, The microphone is an android robot, characterized in that provided on both the left and right ends of the head and the crown. The method of claim 1, The codec and the sound source processor, Android robot, characterized in that provided inside the head. Collecting three microphones arranged to be directional with respect to the sound source, A band pass filter extracts a voice signal of the other party from the signals collected through the plurality of microphones, and extracts a signal of a predetermined frequency band centered on a frequency of 1 kHz from the signals collected through the three microphones, Extracting a voice signal to have a maximum gain at the frequency of 1 kHz; A codec converting the extracted speech signals into digital signals, respectively; A sound source processor tracking a position of the counterpart using a phase difference and a magnitude difference between the voice signals converted into the digital signals, respectively; A sound source processor transmitting the converted voice signal and the tracked location information to a host processor; Receiving, by the host processor, the voice signal and the location information; Analyzing, by the host processor, the received voice signal and querying gesture data corresponding to the analyzed voice signal in a gesture DB; And the host processor driving the robot using the inquired gesture data and the received location information, and outputting the robot voice data corresponding to the inquired gesture data by the sound source processor. . In a computer-readable medium that records a program for using the Android robot, Collecting three microphones arranged to be directional with respect to the sound source, A band pass filter extracts a voice signal of the other party from the signals collected through the plurality of microphones, and extracts a signal of a predetermined frequency band centered on a frequency of 1 kHz from the signals collected through the three microphones, Extracting a voice signal to have a maximum gain at the frequency of 1 kHz; A codec converting the extracted speech signals into digital signals, respectively; A sound source processor tracking a position of the counterpart using a phase difference and a magnitude difference between the voice signals converted into the digital signals, respectively; A sound source processor transmitting the converted voice signal and the tracked location information to a host processor; Receiving, by the host processor, the voice signal and the location information; Analyzing, by the host processor, the received voice signal and querying gesture data corresponding to the analyzed voice signal in a gesture DB; A computer recording a program for executing the step of the host processor driving the robot using the inquired gesture data and the received position information, and outputting the robot voice data corresponding to the inquired gesture data by the sound source processor; Readable media.

Images