JP2003076398A - Robot apparatus, robot control method, recording medium, and program - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To provide a more suitable response in response to an utterance from a user. SOLUTION: A voice recognition unit 24 includes a microphone 11
When the feature parameter of the voice collected in step is acquired, voice recognition is performed. When detecting the predetermined keyword, the voice recognition unit 24 notifies the melody generation unit 32 and the like of the keyword. The voice pitch analysis unit 28 analyzes the voice of the keyword based on the instruction from the melody generation unit 32 and extracts the pitch frequency of the voice. The voice pitch analysis unit 28
The musical scale corresponding to the extracted pitch frequency is selected and notified to the melody generating unit 32. The melody generator 32 reads the melody data stored in the melody data storage 33 and converts the melody data based on the scale notified from the voice pitch analyzer 28. The new melody data generated by the melody generation unit 32 is reproduced by the voice synthesis unit 34 and output from the speaker 14.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a robot device, a robot control method, a recording medium, and a program, and more particularly, to a robot device and a robot device which enable a more suitable response in response to an utterance from a user. The present invention relates to a robot control method, a recording medium, and a program.

[0002]

2. Description of the Related Art In recent years, pet robots for entertainment have been realized which autonomously take various actions according to the surrounding environment and the internal state of the pet.

Such a pet robot, for example, emits a sound in response to the user's speech, or a sound indicating anger when the user's head is hit. It is done like this. Also,
When the user does not play much and the internal state becomes an emotion of "lonely", a sound indicating "lonely" (notifying that he wants to play) is output.

[0004]

However, since the sound data output from the pet robot is prepared in advance in the pet robot as described above, the sound data is stored in consideration of the built-in storage capacity. There was a problem that the types were limited. That is, the output sound is patterned.

Therefore, by observing the behavior for a while, the user can easily predict the next sound to be emitted from the pet robot, which makes communication with the pet robot uninteresting. turn into.

The present invention has been made in view of such a situation, and is to make it possible to provide a more suitable response in response to a call from a user.

[0007]

A first robot apparatus according to the present invention comprises a storage means for storing first melody data composed of a first scale and an extraction for extracting a pitch frequency of an inputted voice. Means, selection means for selecting a second scale based on the pitch frequency extracted by the extraction means, and first scale constituting the first melody data stored in the storage means by the selection means. It is characterized in that it is provided with a generating means for converting the generated second melody data by converting the generated second melody data and a reproducing means for reproducing the second melody data generated by the generating means.

The extracting means can extract the pitch frequency of the voice representing a predetermined keyword.

The extracting means may extract the pitch frequency of the voice containing the vowel.

The extracting means may extract the average value of the pitch frequencies detected during a predetermined period of the voice as the pitch frequency.

The generating means can generate the second melody data by transitioning the second scale by a predetermined number of octaves.

A first robot control method for a robot apparatus according to the present invention comprises a storage step of storing first melody data composed of a first scale, and an extraction step of extracting a pitch frequency of an input voice. , A selection step of selecting a second scale based on the pitch frequency extracted by the processing of the extraction step, and a first scale constituting the first melody data stored by the storage step,
A generation step of converting the second scale selected by the processing of the selection step to generate second melody data; and a reproduction step of reproducing the second melody data generated by the processing of the generation step Is characterized by.

The program of the first recording medium of the present invention is
A storage control step for controlling storage of the first melody data composed of the first scale, an extraction step for controlling extraction of the pitch frequency of the input voice, and a pitch frequency extracted by the processing of the extraction step. A selection step for selecting a second scale based on the first scale, and a first step constituting the first melody data stored by the processing of the storage control step.
Of the second melody data generated by the process of the generating step, and controlling the reproduction of the second melody data generated by the process of the selecting step. And a reproduction control step.

A first program of the present invention comprises a storage control step of controlling storage of first melody data composed of a first scale, and an extraction step of controlling extraction of a pitch frequency of input voice. A selection step of selecting a second scale based on the pitch frequency extracted by the processing of the extraction step, and a first scale constituting the first melody data stored by the processing of the storage control step,
A generation step of converting the second scale selected by the processing of the selection step to generate the second melody data, and a reproduction control step of controlling reproduction of the second melody data generated by the processing of the generation step. It is characterized by including.

The second robot device of the present invention is managed by the storage means for storing the first melody data composed of the first scale, the management means for managing its own internal state, and the management means. When the internal state changes, the selecting means selects the second scale corresponding to the change in the internal state, and the first scale forming the first melody data stored in the storage means is selected by the selecting means. It is characterized in that it is provided with a generating means for converting the generated second melody data by converting the generated second melody data and a reproducing means for reproducing the second melody data generated by the generating means.

The generating means can generate the second melody data by transitioning the second scale by a predetermined number of octaves.

A second robot control method for a robot apparatus according to the present invention comprises a storage step of storing first melody data composed of a first scale, a management step of managing its own internal state, and a management step. When the internal state managed by the process of 1 changes, the selection step of selecting the second scale corresponding to the change of the internal state and the first melody data which is stored by the processing of the storing step Of the second scale selected by the processing in the selection step.
And a reproducing step of reproducing the second melody data generated by the processing of the generating step.

The program of the second recording medium of the present invention is
A storage control step for controlling storage of the first melody data composed of a first scale, a management step for managing the internal state of itself, and an internal state when the internal state managed by the processing of the management step changes. The selection step of selecting the second scale corresponding to the change in the state, and the first scale constituting the first melody data stored by the processing of the memory control step are selected by the processing of the selection step. A second step generated by the process of the generating step, which generates the second melody data by converting into the second scale.
And a reproduction control step for controlling reproduction of the melody data of.

The second program of the present invention is a storage control step for controlling storage of the first melody data composed of the first scale, a management step for managing the internal state of itself, and processing of the management step. When the internal state managed by the change occurs, the selection step of selecting the second scale corresponding to the change of the internal state, and the first melody data forming the first melody data stored by the processing of the storage control step. A generation step of converting the scale into the second scale selected by the processing of the selection step to generate the second melody data, and a reproduction for controlling reproduction of the second melody data generated by the processing of the generation step. And a control step.

In the first robot apparatus, the robot control method, and the program of the present invention, the first melody data composed of the first scale is stored, and the pitch frequency of the inputted voice is extracted and extracted. A second scale is selected based on the pitch frequency of the selected pitch. Further, the first scale constituting the stored first melody data is converted into the selected second scale to generate the second melody data, and the second melody data is reproduced.

In the second robot apparatus, the robot control method, and the program of the present invention, the first melody data composed of the first scale is stored, the internal state of itself is managed, and the managed internal When the state changes, the second scale corresponding to the internal state change is selected. In addition, the first scale that constitutes the stored first melody data is converted into the selected second scale, and the second scale is converted into the second scale.
Is generated, and the generated second melody data is reproduced.

[0022]

1 is a perspective view showing an example of the external configuration of a pet robot 1 to which the present invention is applied.

As shown in the figure, for example, the pet robot 1 is in the shape of a dog with four legs, and the leg units 3A, 3 are provided on the front, rear, left and right of the body unit 2, respectively.
B, 3C, and 3D are connected, and a head unit 4 and a tail unit 5 are connected to the front end and the rear end of the body unit 2, respectively.

The tail unit 5 is drawn out from the base portion 5B provided on the upper surface of the body unit 2 so as to be curved or swingable with two degrees of freedom.

As will be described later in detail, the pet robot 1 having such an external structure has melody (melody) data for responding to, for example, when a user calls "Good morning". Has been done.
Then, the pet robot 1 analyzes the voice from the user, converts the prepared melody data based on the analysis result, and reproduces the melody data obtained after the conversion.
That is, the melody data is converted and output for each user or each utterance.

Therefore, even with one kind of melody data,
Since it is converted into melody data according to the utterance from the user every time, the pet robot 1 outputs a sound corresponding to the user's singing. As a result, the degree of mutual understanding with the pet robot 1 is deepened, and it is possible to prevent the user from getting tired of communication.
More suitable communication can be achieved.

FIG. 2 is a block diagram showing an example of the internal configuration of the pet robot 1 of FIG.

The torso unit 2 includes a pet robot 1
A controller 10 for controlling the whole is stored.
The controller 10 basically includes a CPU (Central Pr
There is provided a memory 10B in which a program for the CPU 10A to control each unit is stored.

In addition to the controller 10, the body unit 2 also stores, for example, a battery (not shown) which is a power source of the pet robot 1.

As shown in FIG. 2, the head unit 4 includes:
Microphone 11, CCD (Charge C) that corresponds to the "ear" that senses sound as a sensor that senses an external stimulus
Oupled Device) and CMOS (Complementary Metal Oxide Sem)
A camera 12 corresponding to an “eye” that senses light and a touch sensor 13 that corresponds to a tactile sensation such as pressure caused by a user's touch are provided at predetermined positions. There is.
In addition, the head unit 4 has a “mouth” of the pet robot 1.
The speaker 14 corresponding to is installed at a predetermined position.

Joint parts of the leg units 3A to 3D, connecting parts of the leg units 3A to 3D and the body unit 2, connecting parts of the head unit 4 and the body unit 2, and a tail unit 5 respectively. An actuator is installed at a connecting portion between the body unit 2 and the body unit 2. The actuator operates each part based on an instruction from the controller 10.

In the example of FIG. 2, the leg unit 3A is
Is the actuator 3AA₁Through 3AA_kProvided with legs
The unit 3B includes an actuator 3BA₁Through 3B
A_kIs provided. Also, in the leg unit 3C,
Actuator 3CA₁To 3 CA_kIs installed on the leg
Actuator 3DA for knit 3D₁Through 3DA _kBut
It is provided. Furthermore, the head unit 4 is
Player 4A₁Through 4A_LIs provided and the tail uni
The actuator 5A₁And 5A ₂Is that
It is provided.

The microphone 11 installed in the head unit 4 collects the surrounding voice (sound) including the utterance from the user and outputs the obtained voice signal to the controller 10. The camera 12 captures an image of the surrounding situation and outputs the obtained image signal to the controller 10. Touch sensor 1
3 is provided on the upper part of the head unit 4, detects the pressure received by a physical action such as "stroking" or "striking" from the user, and the detection result is used as a pressure detection signal in the controller 10 Output to.

The controller 10 is the microphone 1
1. Based on the audio signal, the image signal, and the pressure detection signal provided from the camera 12, the touch sensor 13, the presence / absence of the surrounding condition, the instruction from the user, the action from the user, and the like are determined, and the determination result is used. Based on this, the next action taken by the pet robot 1 is determined. Then, the controller 10 drives the necessary actuator based on the determination, thereby swinging the head unit 4 vertically and horizontally, moving the tail unit 5, and each of the leg units 3A to 3D. To drive the pet robot 1 to walk.

In addition, the controller 10 is an LED (Light Emit) provided at the position of the "eye" of the pet robot 1.
(Ting Diode) 41 (see FIG. 3) is turned on, turned off, or blinked.

FIG. 3 is a block diagram showing a functional configuration example of the controller 10 shown in FIG. In addition, each function shown in FIG. 3 is a control program stored in the memory 10B.
It is realized by the execution of the PU 10A.

The audio signal from the microphone 11 is AD
It is supplied to the (Analog Digital) converter 21. AD converter 2
1 samples and quantizes the audio signal supplied from the microphone 11, and converts it into audio data. The voice data acquired by the AD conversion unit 21 is supplied to the voice feature amount analysis unit 22 and the voice pitch analysis unit 28.

The voice feature quantity analysis unit 22 receives, for example, MFCC (M
el Frequency Cepstrum Coefficient) analysis, and the analysis result is used as a feature parameter (feature vector).
It is output to the voice section detection unit 23. In addition, the voice feature amount analysis unit 22 can also extract, for example, a linear prediction coefficient, a cepstrum coefficient, a pair of line spectra, a power (output of a filter bank) for each predetermined frequency band, as a feature parameter. is there.

Based on the characteristic parameters supplied from the speech characteristic amount analyzing section 22, the voice section detecting section 23 determines whether or not power having a value equal to or larger than a predetermined threshold value is continuously detected for a predetermined period. Based on this, the voice section is detected.
Then, the voice section detection unit 23 outputs the detected voice section information to the voice recognition unit 24 together with the feature parameter supplied from the voice feature amount analysis unit 22. The voice section information detected by the voice section detecting unit 23 is also supplied to the voice pitch analyzing unit 28.

The speech recognition unit 24 describes an acoustic model representing acoustic features such as individual phonemes and syllables in the language of speech to be recognized, and information (phonological information) about pronunciation of each word to be recognized. Word dictionary, and
It stores grammatical rules that describe how the words registered in the word dictionary are chained (connected). Here, as the grammar rule, for example, a rule based on context-free grammar (CFG), statistical word chain probability (N-gram), or the like can be used.

Then, the voice recognition unit 24 uses the feature parameters from the voice feature amount analysis unit 22 to input to the microphone 11 while referring to such an acoustic model, word dictionary, and grammar rules as needed. The voice
For example, speech recognition is performed based on the continuous distribution HMM (Hidden Markov Model) method.

Specifically, the voice recognition unit 24 constructs an acoustic model of a word (word model) by referring to the word dictionary and connecting acoustic models. In addition, the voice recognition unit 24
Connects several word models by referring to the grammar rules, and recognizes the speech input to the microphone 11 by the continuous distribution HMM method based on the word model and the feature parameters thus connected. . That is, the voice recognition unit 24 detects the word model sequence having the highest score (likelihood) at which the time-series feature parameters output by the voice feature amount analysis unit 22 (voice section detection unit 23) are detected, and A word string corresponding to a series of word models is output as a speech recognition result.

That is, the voice recognition unit 24 accumulates the appearance probability of the feature parameter from the voice feature amount analysis unit 22 for the word string corresponding to the connected word model, and uses the accumulated value as a score to obtain the score. The highest word string is output as the voice recognition result.

For example, the voice recognition unit 24 is "walking",
Commands such as "prone" and "follow the ball"
The instinct / emotion management unit 26 and the behavior management unit 27 are notified as voice recognition information. In addition, when the voice recognition unit 24 recognizes a predetermined keyword, the voice recognition unit 24 notifies the melody generation unit 32 or the like of it.

The image signal picked up by the camera 12 and the pressure detection signal detected by the touch sensor 13 are output to the sensor data processing section 25. The sensor data processing unit 25 includes an image recognition unit that performs image recognition based on the image signal supplied from the camera 12, and a touch sensor input processing unit that processes the pressure detection signal supplied from the touch sensor 13 (whichever Is also provided).

This image recognition unit performs image recognition using the image signal supplied from the camera 12, and as a result of the processing, for example, "a red round object" or "a perpendicular to the ground and a predetermined value. When detecting a "plane equal to or higher than the height of", the image recognition result such as "there is a ball" or "there is a wall" is output to the instinct / emotion management unit 26 and the behavior management unit 27 as image recognition information. To do.

The touch sensor input processing unit processes the pressure detection signal given from the touch sensor 13, and when the pressure detection signal which is equal to or higher than a predetermined threshold value for a short period of time is detected as a result of the processing, it is "stripped ( If the pressure is below a predetermined threshold for a long time, it is recognized as "stroked (praised)" and the recognition result is used as state recognition information. As instinct / emotion management unit 2
6 and the behavior management unit 27.

The instinct / emotion management unit 26 manages parameters representing the instinct or emotion of the pet robot 1, and outputs the state of instinct or the state of emotion to the behavior management unit 27 at a predetermined timing.

FIG. 4 is a diagram schematically showing an example of the functional configuration of the instinct / emotion management unit 26 shown in FIG. 3. As shown in FIG.
The instinct / emotion management unit 26 includes an emotion model 51 that expresses the emotion of the pet robot 1 and an instinct model 5 that expresses the instinct.
2 is stored and managed.

The emotion model 51 includes, for example, "joy", "sadness", "anger", "surprise", "fear",
The emotional state (degree) such as "disgust" is expressed by emotion parameters in a predetermined range (for example, 0 to 100), and the output from the voice recognition unit 24 and the sensor data processing unit 25, the passage of time, etc. Based on that, the value is changed.

In this example, the emotion model 51 is
Emotional unit 51A expressing "joy", "Sadness"
An emotional unit 51B that represents, an emotional unit 51C that represents "anger", an emotional unit 51D that represents "surprise",
It is composed of an emotion unit 51E representing "fear" and an emotion unit 51F representing "dislike".

The instinct model 52 is, for example, “appetite”,
The states (degrees) of instinct such as "sleep desire" and "exercise desire" are represented by instinct parameters in a predetermined range, and the voice recognition unit 24 and the sensor data processing unit 2
The value is changed based on the output from 5 and the passage of time. Further, the instinct model 52 enhances the parameter indicating “motility” based on the action history, or enhances the parameter indicating “appetite” based on the remaining battery level.

In this example, the instinct model 52 is
An instinct unit 52A representing "motivation", an instinct unit 52B representing "love", an instinct unit 52C representing "appetite", an instinct unit 52D representing "curiosity",
And an instinct unit 52E indicating "sleep desire".

The instinct / emotion management section 26 has the above emotion units 51A to 51F and instinct units 52A to 5F.
By changing the parameters of 2E, the emotion and instinct of the pet robot 1 are expressed and the change is modeled.

Further, the parameters of the emotion units 51A to 51F and the instinct units 52A to 52E are changed not only by external input, but also by mutual influence of the respective units as shown by arrows in the figure. To be done.

For example, an emotional unit 51A expressing "joy" and an emotional unit 51B expressing "sadness".
Instinct / emotion management unit 26 is "joyful" when the user compliments each other due to the mutual inhibition.
The emotion unit 51B that expresses "sadness" while increasing the parameter of the emotion unit 51A that expresses
Change the emotion to be expressed, for example, by reducing the parameter of.

Further, the parameters of the respective units constituting the emotion model 51 and the units constituting the instinct model 52 are changed not only between the respective units but also over both models.

For example, as shown in FIG.
Instinct unit 52B that expresses "love desire" and "appetite"
The parameters of the emotional unit 51B expressing "sadness" and the emotional unit 51C expressing "anger" of the emotion model 51 are changed according to the change of the parameter of the instinct unit 52C.

Specifically, when the parameter of the instinct unit 52C representing "appetite" becomes large, the emotion unit 51B expressing "sadness" of the emotion model 51,
The parameter of the emotion unit 51C expressing "anger" becomes large.

More specifically, not only the emotion model 51 and the instinct model 52, but also a growth model is prepared in the instinct / emotion management unit 26, and each unit of the emotion model 51 and the instinct model 52 is selected depending on the growth stage. The parameters are changed. This growth model represents, for example, growth states (degrees) such as “childhood”, “adolescence”, “mature”, “old age” by a value in a predetermined range, and the speech recognition unit 24, The value is changed based on the output from the sensor data processing unit 25, the passage of time, and the like.

The instinct / emotion management unit 26 outputs the emotions and instinct states represented by parameters such as the emotion model and the instinct model to the action management unit 27 as internal information.

The instinct / emotion management unit 26 is supplied with recognition information from the voice recognition unit 24 and the sensor data processing unit 25, and the behavior management unit 27 supplies the present or past information of the pet robot 1. Actions, specifically, for example,
Behavior information indicating the content of the behavior such as "walking for a long time" is supplied. Then, the instinct / emotion management unit 26 generates different internal information according to the action of the pet robot 1 indicated by the action information even when the same recognition information or the like is given.

For example, when the pet robot 1 greets the user and the user pats his / her head, the instinct / emotion management section provides the action information that the user greeted the user and the recognition information that the user patted the head. 26. At this time, in the instinct / emotion management unit 26, the value of the emotion unit 51A representing “joy” is increased.

Returning to the explanation of FIG. 3, the action management section 27 has information supplied from the voice recognition section 24 and the sensor data processing section 25, internal information supplied from the instinct / emotion management section 26, and The next action is determined based on the elapse of time, etc., and the command generation unit 29 is notified of information indicating the determined action.

FIG. 5 is a schematic diagram showing a functional configuration example of the behavior management unit 27 of FIG.

The behavior management unit 27 is composed of a behavior model library 61 and a behavior selection unit 62, and the behavior model library 61 responds to preset conditions (triggers) as shown in FIG. Have a behavior model of.

In the example shown in the figure, the behavior model library 61 includes a ball-corresponding behavior model 61A indicating an action to be taken when a ball is detected, an autonomous retrieval behavior model 61B indicating an action to be taken when the ball is lost, and the like. An emotion expression action model 61C indicating an action to be taken when a change in the emotion model 51 is detected is prepared. In addition, an obstacle avoidance behavior model 61D that shows an action to be taken when an obstacle is detected, a fall recovery behavior model 61E that shows an action to be taken when a fall is detected, and an action to be taken when the battery level is low. A battery management behavior model 61F indicating is shown.

Then, the action selecting section 62 is connected to the voice recognizing section 2
4. Information supplied from the sensor data processing unit 25 and the like,
By referring to the internal information supplied from the instinct / emotion management unit 26, the passage of time, etc., the next action to be taken by the pet robot 1 is selected from the action models prepared in the action model library 61.

The action selection unit 62 determines which state the node representing the current state of the pet robot 1 transits to, based on the transition probability set in the arc connecting the respective nodes. The finite probability automaton as shown in FIG.

The finite probability automaton shown in FIG. 6 has a probability P when the current state is node 0 (NODE ₀ ).
Transition to node 1 (NODE ₁ ) at ₁ and node 2 (NODE at probability P ₂
DE ₂ ), with probability P _n , to node n (NODE _n ). Further, this finite probability automaton shows that it transits to the node 0 (NODE ₀ ) with the probability P _n−1 , that is, it does not transit to any node.

Each action model defined in the action model library 61 is composed of nodes representing a plurality of states, and each node has a state transition in which the probability of transition to another node is described. The table is set.

FIG. 7 shows a node 100 (NODE) belonging to a predetermined behavior model defined in the behavior model library 61.
It is a figure which shows the example of the state transition table of ₁₀₀ ).

In FIG. 7, the input event as a condition for transition to another node (or its own node) is described in the column of "input event name", and the "input event name" defined as the transition condition is specified. Are described in the columns of “data name” and “data range”.

In the example of the figure, the information having the ID "1" is "ball detected" (input event name "BA
LL ”) information about that is notified to the behavior management unit 27,
And the size of the detected ball is "0 to 1000".
In the range of, the transition from the node 100 to the node 120 is shown with a probability of “30%”.

Further, the information to which the ID "2" is set is
When the action management unit 27 is notified of information regarding "slapped on the head" (input event name "PAT"),
Node 40 to node 1 with a probability of “40%”
The transition to 50 is shown.

Furthermore, the information for which the ID "3" is set is "Head hit hard" (input event name "HIT").
When the behavior management unit 27 is notified of the information regarding the thing, it indicates that the transition from the node 100 to the node 150 is made with a probability of “20%”.

As for the information having the ID "4", the information about "obstacle detected" (input event name "OBSTACLE") is notified to the action management unit 27, and the distance to the obstacle is detected. Indicates that the node 100 is transitioned from the node 100 to the node 1000 with a probability of “100%” when it is notified that the value is in the range of “0 to 100”.

Further, in this example, the voice recognition unit 24
Even when there is no input from the sensor data processing unit 25 or the like, the transition is made to another node according to the parameters of the emotion model 51 or the instinct model 52.

For example, when the parameter of the emotion unit 51A representing the "joy (JOY)" of the emotion model 51 is in the range of "50 to 100", the information in which ID5 is set has a probability of "5%". Indicates that a transition is made to the node 120 and a transition is made to the node 150 with a probability of “30%”.

The information to which the ID 6 is set is the emotion unit 51 representing “SURPRISE” of the emotion model 51.
When the parameter of D is in the range of “50 to 100”, it indicates that the node 1000 is transited to the node 1000 with a probability of “15%”.
When the parameter of the emotion unit 51B representing “SUDNESS” of the emotion model 51 is in the range of “50 to 100”, the node 12 has a probability of “50%”.
This indicates that the transition to 0 is made.

Further, in the state transition table of FIG. 7, the action output at each node when transitioning to each node is described in the column of “output action”. For example, when transitioning to node 120, The action that is output is "ACTION1", the action that is output when transitioning to the node 150 is "ACTION2", and the action that is output when transitioning to the node 1000 is "MOVE BACK" (back)
Is said to be.

The action selecting section 62 makes a state transition by referring to such a state transition table, and outputs information indicating an action set in the node to the instinct / emotion managing section 2
6, output to the command generation unit 29 and the like.

[0083] Then, for example, is managed by the behavior management unit 27, if the action is defined in the transition node is a "speak Hello", the command generating unit 29, based on an instruction from the behavior management unit 27 A speech command is generated and output to the voice synthesis unit 34.

If the action defined for the transitioned node is "backward", the command generation unit 29
Based on the instruction from the behavior management unit 27, a command to execute it is generated and output to the control unit 31. Further, when the LED 41 provided at the position of the “eye” of the pet robot 1 is turned on, turned off, or blinks, the command generation unit 29 generates a command instructing the LED 41 and outputs the command to the LED control unit 30.

The LED control section 30 controls the LED 41 based on the command supplied from the command generation section 29.

The controller 31 controls the actuator 3AA ₁ based on the command supplied from the command generator 29.
To 5A ₂ are controlled to change the posture of the pet robot 1 from the current posture to the next posture.

Here, the posture to which the current posture can be changed next is, for example, the physical shape of the pet robot 1 such as the shape and weight of the body, hands and feet, and the connection state of the respective parts, and the joint bends. It is determined by the mechanism of the actuators 3AA _{1 to} 5A ₂ such as direction and angle.

The postures include a posture that can directly make a transition from the current posture and a posture that cannot make a direct transition. For example,
The four-legged pet robot 1 can make a direct transition to a prone state from a state of lying down with its limbs largely thrown out, but cannot make a direct transition to a standing state.
Once I pulled my limbs close to my torso and fell down,
Then, a two-step operation of starting up is required.
There are also postures that cannot be safely executed. For example, the pet robot 1 easily falls when attempting to banzai with both front legs raised from a standing posture with four legs.

Therefore, the control unit 31 registers in advance a posture in which direct transition is possible, and when the command supplied from the command generation unit 29 indicates a posture in which direct transition is possible, the control unit 31 responds based on that. Drive the actuator.

On the other hand, when the command supplied from the command generation unit 29 indicates a posture in which direct transition is not possible, the control unit 31 makes a transition to another posture in which transition is possible and then
The corresponding actuator is driven so as to transition to the desired posture. As a result, it is possible to avoid a situation in which the pet robot 1 is forced to execute an untransitable posture or a situation in which the pet robot 1 falls.

When the melody generating section 32 is notified by the voice recognizing section 24 that the keyword has been recognized, the melody generating section 32 instructs the voice pitch analyzing section 28 to analyze the pitch frequency of the keyword. Then, the melody generating unit 32 reads the predetermined melody data from the melody data storage unit 33 when the voice of the keyword is analyzed and the scale selected based on the pitch frequency is notified from the voice pitch analyzing unit 28. The melody data is converted based on the notified scale.

That is, in the melody data storage unit 33, a plurality of melody data of sounds (melody) output from the pet robot 1 are prepared, for example, in the MIDI format.

The melody generating section 32 performs processing such as replacing each scale of the melody data with the scale supplied from the voice pitch analyzing section 28 to generate new melody data. The melody data generated by the melody generator 32 is output to the voice synthesizer 34.

The voice pitch analysis section 28 includes a melody generation section 32.
When instructed by, the input voice of the keyword supplied from the AD conversion unit 21 is analyzed, and the pitch frequency at each time is calculated. For example, the voice pitch analysis unit 28 uses the time shift amount corresponding to the pitch frequency as a variable, and calculates the pitch frequency at each time of the input voice based on the time shift amount having the maximum value of the autocorrelation coefficient. . For example, when the sampling frequency of the input voice by the voice pitch analysis unit 28 is 16 kHz and the time shift amount at which the autocorrelation coefficient has the maximum value is 118, the voice pitch analysis unit 28 determines the pitch frequency f (Hz). F _PITCH = 160
Calculated as 00/118 = 135.59 Hz.

Then, the voice pitch analysis section 28 extracts a predetermined pitch frequency for selecting a scale based on the calculated pitch frequency, and selects a scale corresponding to the extracted pitch frequency. The voice pitch analysis unit 28 extracts, for example, a plurality of pitch frequencies composed of an average value of a predetermined period, and selects a plurality of scales based on the extracted pitch frequencies. The scale selected by the voice pitch analysis unit 28 is the melody generation unit 32.
Is used to convert melody data.

When the predetermined utterance command and the text data indicating the content of the utterance are supplied from the command generating unit 29, the voice synthesizing unit 34 refers to the internal voice synthesizing dictionary to convert the text data into the text data. For example, WAV format corresponding audio data is generated.

More specifically, in this voice synthesis dictionary,
A word dictionary in which part-of-speech information of each word, reading, accent, etc. is described, grammatical rules for generation of restrictions on word chains, and phoneme piece data as speech information for the words described in the word dictionary. The stored voice information dictionary is stored.

The voice synthesizer 34 performs morphological analysis (syntactic analysis) and other text analysis (language analysis) of the input text based on the word dictionary and the grammatical rules for generation,
Extract the information required for speech synthesis. As information necessary for voice synthesis, for example, prosodic information for controlling pose position, accent, intonation, power, etc.,
There is phonological information indicating the pronunciation of each word.

The voice synthesizer 34 connects the necessary phoneme piece data while referring to the voice information dictionary based on the extracted various information, and further processes the waveform of the phoneme piece data to make a pause, Accents, intonations, etc. are appropriately added, and voice data corresponding to the supplied text data is generated.

The voice data generated by the voice synthesizer 34 is digital-analog converted in the DA converter 35 and output from the speaker 14.

The voice synthesizer 34 also includes a melody generator 32.
When the melody data is supplied from, play it,
It is output from the speaker 14.

Next, the operation of the pet robot 1 having the above configuration will be described.

First, the processing of the pet robot 1 for generating melody data using the scale selected based on the input voice and reproducing it will be described with reference to the flowchart of FIG.

In step S1, the voice recognition unit 24
Determines whether or not a keyword is detected as a result of voice recognition, and waits until it is determined that a keyword is detected.
For example, the voice uttered by the user is collected by the microphone 11 as described above, subjected to each processing, and then supplied to the voice recognition unit 24.

The voice recognition unit 24 stores a plurality of keywords, and based on this, determines whether or not the input voice is a keyword. This keyword is
For example, it may be composed of a vowel (for example, a sound of "A") whose frequency change is relatively small and whose pitch frequency is easily extracted.

Then, the voice recognition section 24 operates in step S1.
In step S2, if it is determined that the keyword is detected, the process proceeds to step S2, and the action management unit 27 and the melody generation unit 32 are notified of the fact.

In step S3, the melody generator 32
Instructs the voice pitch analysis unit 28 to analyze the detected keyword.

In step S4, the voice pitch analysis section 28 analyzes the waveform of the keyword supplied from the AD conversion section 21 according to the instruction from the melody generation section 32, and extracts the pitch frequency.

FIG. 9 is a diagram for explaining the processing of the voice pitch analysis unit 28 which extracts the pitch frequency. It should be noted that FIG. 9 shows an example in the case where “LaLaLa” is detected as the keyword.

The upper part of FIG. 9 shows the waveform of the keyword "la la la" with the vertical axis representing the amplitude of the voice and the horizontal axis representing the time. Further, the vertical axis of the diagram shown in the lower part of FIG. 9 represents the pitch frequency of the keyword “La la la”,
The horizontal axis represents time.

When the waveform of the keyword "La la la" as shown in the upper part of FIG. 9 is supplied, the voice pitch analysis unit 28 sets the pitch frequency at each time to the time shift amount and the sampling frequency as described above. Calculate based on

Then, the voice pitch analysis section 28, for example, uses the voice intervals (t ₁ , t ₂ ,
t ₃ ) is equally divided into three, and the average pitch frequency of the central section is extracted from each of the three equally divided speech sections. The information regarding the voice section has been detected and supplied by the voice section detection unit 23.

In the example of the figure, the voice section t ₁ is divided into three voice sections t ₁₁ , t ₁₂ , and t ₁₃ , respectively, and similarly, the voice section t ₂ is divided into voice sections t ₂₁ , t ₂₂ , and t ₂₃ . The voice section t ₃ is set as voice sections t ₃₁ , t ₃₂ , and t ₃₃ .

Then, the voice section t₁₂In the section
The average value of the H frequency is set to 131, and the voice section t_{twenty two}Section of
The average value of the pitch frequency in
Interval t ₃₂The average value of the pitch frequency in the section is
Has been done.

As described above, not only the average value of the pitch frequencies in the central section of each sound of the keyword is calculated and extracted, but also the pitch frequency can be extracted by various methods. As a method of extracting the pitch frequency, for example, the average pitch frequency of each sound of the keyword may be extracted, or the maximum pitch frequency in a predetermined section may be extracted. Alternatively, the pitch frequency at the center time of each sound may be extracted.

Then, in step S5, the voice pitch analysis section 28 selects a predetermined scale based on the pitch frequency thus extracted.

FIG. 10 is a diagram showing an example of correspondence between pitch frequencies and scales managed by the voice pitch analysis section 28. Figure 1
A correspondence example of 0 is that "A4" is a sound of "440 Hz", and the scale from "C1" to "B7" and the respective frequencies are shown.

Therefore, for example, when the pitch frequencies of "131 Hz, 148 Hz, 163 Hz" are extracted from the keyword "Rarara" as described with reference to FIG. The scale of "C3, D3, E3" corresponding to is selected. That is, the voice pitch analysis unit 28 selects the scale having the frequency closest to the extracted pitch frequency.

The pet robot 1 extracts pitch frequencies in the range of 80 Hz to 400 Hz from the frequencies shown in FIG. 10 for easy recognition.

The scale selected by the voice pitch analysis section 28 in this manner is notified to the melody generation section 32.

On the other hand, the action management section 27 instructs the melody generation section 32 to generate melody data when the voice recognition section 24 notifies that the keyword is detected in step S6. In addition to the instruction, the melody generator 32 is also notified of the identification information of the melody data to be converted.

In step S7, the melody generator 32
Reads the melody data corresponding to the identification information notified from the behavior management unit 27 from the melody data storage unit 33,
New melody data is generated based on the scale supplied from the voice pitch analysis unit 28.

FIG. 11 is a diagram showing an example in which the melody data stored in the melody data storage unit 33 is represented on a musical score, and the scales 1 to 9 of this melody data 1-1 are "C3, C3," respectively. , E3, G3, C3, C3, E3, G
3, C3 (+12) ". Note that scale 9 is scale 1
Based on (“C3”), for example, 1 on a piano
It indicates that the sound is only two keys above. In this way, when the types of scales forming the melody data are equal to or larger than the number of scales extracted by the keyword, the predetermined scale forming the melody data is expressed with another scale as a reference.

Then, the melody generator 32 first generates, as shown in FIG.
Replace each scale with a given letter. It should be noted that this processing is used for convenience of explanation, and actually, the melody data 1
-1 may be directly converted into the melody data 1-3 (Fig. 13).

For example, the melody generator 32 uses the melody data 1
-1 "C3" to "X", "D3" to "Y", "E"
3 ”is replaced with“ Z ”. As a result, the melody data 1-1 becomes "X,
X, Y, Z, X, X, Y, Z, X (+12) ".

Further, the melody generating section 32 uses the melody data 1-
The scale selected by the voice pitch analysis unit 28 is associated with each character of 2. FIG. 13 is a diagram illustrating a process in which the scale selected by the voice pitch analysis unit 28 is associated and new melody data is generated.

For example, as described above, when the scale of "C3, D3, E3" is selected by the voice pitch analysis unit 28 based on the keyword "La la la", the melody generation unit 32.
Is "C3, D" in "X, Y, Z" of melody data 1-2.
3, E3 ”are associated with each other. Then, the melody generation unit 32 displays “C3, C3, D3, E3, C
3, C3, D3, E3, C3 (+12) ”is generated to generate melody data 1-3. That is, at this point, the melody data 1-1 prepared in advance is converted into the melody data 1-3 based on the pitch frequency extracted from the keyword “La la la” (scales 3 and 4, and scale 7). And 8 are converted to other scales respectively).

Further, when the melody data generated as described above is a relatively low tone, the melody generating section 32 transitions each tone of the generated melody data to a high pitch by a predetermined number of octaves to generate a new melody. Generate data. Of course, if the generated melody data is a relatively high-pitched sound, it may be lowered by a few octaves.

FIG. 14 is a diagram for explaining the processing of the melody generating section 32 which generates new melody data by shifting each scale of the melody data by, for example, 3 octaves.

As shown in the figure, each scale of the melody data 1-3 has transitioned to the high pitch by 3 octaves,
"C6, C6, D6, E6, C6, C6, D6, E6
C6 (+12) ”melody data consisting of a sequence of scales 1-
4 has been generated. That is, the melody data 1-4 is newly generated from the melody data 1-1 prepared in advance.

Then, the melody data generated by the melody generating section 32 is output to the voice synthesizing section 34.

The voice synthesizing unit 34 reproduces the melody data supplied from the melody generating unit 32 and outputs it from the speaker 14 in step S8. The length of the output sound may be the same as the melody data stored in advance in the melody data storage unit 33, or may be the pitch frequency extracted as described above or the time when the pitch frequency is detected, Alternatively, the conversion may be performed based on the range of the detection section or the like.

FIG. 15 is a diagram for explaining other melody data stored in the melody data storage unit 33 and the conversion processing thereof.

In FIG. 15, the arrangement of notes in the melody data 2-1 is "C3, E3, G3, E3, G3, C3 (-
5), E3, G3, E3, G3 ". Then, as described above, from the keyword “LaLaLa” to “C
When the scale of "3, D3, E3" is selected, "C3, E3, G3" of the melody data 2-1 is changed to "C3, D", respectively.
"3, E3" (arrow processing), "C3, D
3, E3, D3, E3, C3 (-5), D3, E3, D
3, E3 ”is converted into melody data 2-2 consisting of a sequence of sounds. That is, at this point, the melody data 2-
The scales 2 to 5 and the scales 7 to 10 which form 1 are converted into other scales, respectively.

Further, each note of the melody data 2-2 is transitioned to a high pitch by 3 octaves by the melody generating section 32 (processed by an arrow), and "C6, D6, E6, D6, E"
6, C6 (−5), D6, E6, D6, E6 ”, the melody data 2-3 is generated. Then, the melody data 2-3 is reproduced, and the corresponding sound is output.

FIG. 16 is a diagram for explaining still another melody data stored in the melody data storage unit 33 and the conversion processing thereof.

In FIG. 16, the arrangement of notes in the melody data 3-1 is "C3, E3, G3, C3 (+12), C3, E.
3, G3, C3 (+12), C3 (-5), C3 (-
1) ”. Then, as described above, when the scale of "C3, D3, E3" is selected from the keyword "Rarara", "C3, E3, G3" of the melody data 3-1.
Are converted to "C3, D3, E3" respectively (arrows

[Outer 1] Processing), “C3, D3, E3, C3 (+12), C
3, D3, E3, C3 (+12), C3 (-5), C3
(-1) "is converted into melody data 2-2 consisting of a sequence of sounds. That is, at this point, the melody data 2-
The scales 2 and 3 constituting 1 and the scales 6 and 7 are converted into other scales, respectively.

Further, each note of the melody data 3-2 is transitioned to a high note by 3 octaves by the melody generating section 32 (arrows).

[Outside 2] Processing), “C6, D6, E6, C6 (+12), C
6, D6, E6, C6 (+12), C6 (-5), C6
Melody data 3-3 composed of a sequence of sounds of (-1) is generated. Then, the melody data 3-3 is reproduced and the corresponding sound is output.

As described above, by converting the prepared melody data based on the pitch frequency extracted from the input voice of the keyword, even if a plurality of users sequentially utter the same keyword. , Different sounds are output from the same melody data.

Therefore, since the sound output from the pet robot 1 becomes unpredictable, the pet robot 1
Communication with will be more interesting.

In the above, the scale corresponding to the extracted pitch frequency is selected, and the melody data obtained after conversion based on the scale is transitioned by a predetermined number of octaves, but of course, conversely, It is also possible to select a scale corresponding to the extracted pitch frequency, and use the transition of that scale by a predetermined number of octaves to convert the melody data prepared in advance.

In this case, for example, the keyword "La la la"
"C3, D3, E selected based on the pitch frequency of
The scale of "3" is transitioned to the high pitch by 3 octaves, and the scale of "C6, D6, E6" is obtained. Then, for example, "C3, E3, G" of the melody data 1-1 of FIG.
3 ”are converted by“ C6, D6, E6 ”, respectively, and“ C6, C6, D6, E6, C6, C6, D6, E
6, C6 (+12) ”sequence of notes 1
-4 is generated.

Further, the scale selected based on the pitch frequency may not only be transited by a predetermined number of octaves, but may be transited to a sound within the frequency range of the utterance by the pet robot 1.

FIG. 17 is a diagram for explaining the processing for transitioning the scale selected based on the pitch frequency to the range of a predetermined frequency.

FIG. 17A is a diagram showing an example of the scale selected based on the pitch frequency. As described above, the scale of "C3, D3, E3" is selected based on the keyword "La la la". There is. Then, the selected scale is always transited as shown in FIG. 17B so that it is within the standard speech range of the pet robot 1.

For example, in this example, the central note "D3" of the selected scale is changed to the note "C6".

Therefore, when the melody data is converted on the basis of the scale "B5, C6, D6" which has been changed to the standard utterance range of the pet robot 1 as described above, the melody data 1-1
Is "B5, B5, C6, D6, B5, B5, C6, D
6, B5 (+12) ”is converted into melody data consisting of a sequence of sounds. That is, each scale of the converted melody data is within the range of the standard speech areas “F5” to “F6” of the pet robot 1 shown in FIG.

As described above, the sound output from the pet robot 1 can be controlled by transitioning the scale selected based on the pitch frequency to various regions. As a result, it is possible to prevent the pitch of the input voice from appearing directly in the sound output from the pet robot 1. That is, if such a transition in the frequency domain is not performed, and if a high voice is input, a high tone is output from the pet robot 1 accordingly.

In the above description, the melody data is converted based on the scale corresponding to the pitch frequency of the voice.
It is also possible to convert the melody data based on various information and output it.

Next, with reference to the flowchart of FIG. 18, the processing of the pet robot 1 for generating melody data using the scale selected based on the change in emotion and reproducing it will be described.

In step S21, the behavior management unit 27 determines whether or not a change in emotion of the pet robot 1 is detected, based on the internal information notified from the instinct / emotion management unit 26, and it is detected. Wait until the judgment.

The action management section 27 then proceeds to step S2.
When it is determined in 1 that a change in emotion has been detected, the process proceeds to step S22, and information indicating that the emotion of the pet robot 1 has changed is notified to the melody generation unit 32 together with the identification information of the melody data to be converted.

In step S23, the melody generator 32 selects the scale used for converting the melody data based on the changed emotion when the behavior manager 27 is notified that the emotion has changed.

FIG. 19 is a diagram showing an example of correspondence between the changed emotion of the pet robot 1 and the musical scale, and such correspondence is prepared in advance in the melody generation section 32.

In this example, when the emotion changes to "JOY (joy)", the melody generator 32 outputs "C, E,
When the "G" scale is selected and the emotion changes to "SAD", when the "C, D #, G" scale is selected and the emotion changes to "ANGRY", " It is said that a scale of "C, C, F" is selected. In addition, the melody generator 32
When the emotion changes to "SURPRISE",
Select the scale of "C, D #, F #" and the emotion is "DISGUST
(Aversion) ”, the scale of“ C, F #, C ”is selected, and when the emotion changes to“ FEAR ”,
It is said that the scale of "C, C #, C" is selected.

For example, when "JOY" is associated with a major sound and "SAD" is associated with a minor sound, when the pet robot 1 is happy, it generally feels bright. When the person is sad, a sound that feels dark and sad is generally output. Therefore, the emotion of the pet robot 1 can be expressed.

Further, when the emotion value is large, it is natural to raise the pitch, so when the emotion value is 10, the pitch is +1.
It is possible to shift 0.

Then, in step S24, the melody generating section 32 reads out the melody data instructed by the behavior management section 27 from the melody data storage section 33, and converts the melody data based on the selected scale.

FIG. 20 is a diagram for explaining conversion processing of melody data.

FIG. 20 shows an example in which the melody data 1-1 shown in FIG. 11 is converted, and the emotion is changed to “ANGRY” with respect to the melody generating section 32. It is an example of when notification is given.

Therefore, the melody generator 32 selects the scale of "C, C, F" from the correspondence table as shown in FIG. 19 and converts the melody data 1-1 based on the selected scale. For example, the melody generation unit 32 sets the selected scale “C, C, F” to “C6, C6, F6” so that the output sound is within the standard utterance range of the pet robot 1, and the melody data 1 -1 "C3, E3, G3" is changed to "C6, E3, G3", respectively.
Convert with "C6, F6".

Then, by the above conversion, "C6
C6, C6, F6, C6, C6, C6, F6, C6 (+
12) ”is formed, and melody data 1-5 is generated.

Further, for example, when it is notified that the emotion of the pet robot 1 has changed to "SAD (sadness)", the melody generating section 32 selects the scale of "C, D #, G". , So that the output sound is within the standard utterance range of the pet robot 1, the scale is “C6, D6 #, G6”.
And Then, the melody generator 32 displays “C6, D6 #,
The melody data 1-1 is converted using the scale of "G6", and "C
6, C6, D6 #, G6, C6, C6, D6 #, G6
New melody data composed of a sequence of sounds of "C6 (+12)" is generated.

The melody data generated in this way is
It is supplied to the voice synthesizer 34, and in step S25,
It is reproduced by the voice synthesizer 34.

As described above, the melody data prepared in advance can be converted based on the various information detected by the pet robot 1, and new melody data can be generated. For example, the proportion of a predetermined color (light) included in the image captured by the camera 12 may be extracted, and the scale may be selected based on the extracted proportion.
It is also possible to select a scale based on the level of the external pressure detected by 3 and use it to generate new melody data. Also, the microphone 11
It is also possible to generate new melody data based on the volume of a voice or the like collected by.

As a result, the pet robot 1 can be made to output numerous patterns of sounds (melody) that cannot be predicted by the user. Further, it is possible to reduce the number of melody data prepared in advance.

The above-mentioned various processes are not only executed by the animal type robot as shown in FIG.
It may be executed by a humanoid robot capable of bipedal walking, a virtual robot that operates in a computer, or the like.

The series of processes described above can be executed by hardware, but can also be executed by software. In this case, the information processing device that executes the software is configured by a personal computer as shown in FIG. 21, for example.

In FIG. 21, the CPU 71 uses the ROM (Read O
nly Memory) 72 or a program stored in the storage unit 78 to RAM (Random Access Memory) 73
Various processes are executed according to the program loaded in. The RAM 73 also appropriately stores data necessary for the CPU 71 to execute various processes.

The CPU 71, ROM 72, and RAM 73 are connected to each other via a bus 74. An input / output interface 75 is also connected to the bus 74.

The input / output interface 75 includes an input unit 76 including a keyboard and a mouse and a CRT (Cathode Ray).
Tube), an LCD (Liquid Crystal Display) display, etc., and an output unit 77 including a speaker, a storage unit 78 including a hard disk, a modem,
A communication unit 79 composed of a terminal adapter or the like is connected. The communication unit 79 performs communication processing via the network.

A drive 80 is connected to the input / output interface 75 if necessary, and a magnetic disk 81,
The optical disc 82, the magneto-optical disc 83, the semiconductor memory 84, or the like is mounted as appropriate.

When a series of processes is executed by software, a program forming the software is installed in a general-purpose personal computer as shown in FIG. 21 from a network or a recording medium.

As shown in FIG. 21, this recording medium is a magnetic disk 81 (including a floppy disk) on which a program is recorded, which is distributed in order to provide the program to the user, separately from the main body of the apparatus, an optical disk. 82 (CD
-ROM (Compact Disk-Read Only Memory), DVD (Digital V
ersatile disk), a magneto-optical disk 83 (including MD (registered trademark) (Mini-Disk)), or a package medium including a semiconductor memory 84, etc. ROM 72 in which the program is recorded, which is provided to the user in the state
Or a hard disk included in the storage unit 78.

In the present specification, the steps for writing the program recorded on the recording medium are not limited to the processing performed in time series according to the order described, but are not necessarily performed in time series, but may be performed in parallel. Alternatively, it also includes processes that are individually executed.

[0176]

According to the first robot apparatus, the robot control method, and the program of the present invention, the first melody data composed of the first scale is stored, and the pitch frequency of the inputted voice is extracted. Then, the second scale is selected based on the extracted pitch frequency. In addition, the first scale constituting the stored first melody data is converted into the selected second scale to generate the second melody data, and the second melody data is reproduced. Therefore, it is possible to provide a more suitable response according to the utterance from the user.

According to the second robot device, the robot control method, and the program of the present invention, the first melody data composed of the first scale is stored, and the internal state of itself is managed and managed internally. When the state changes, the second scale corresponding to the change in the internal state is selected. In addition, the first scale constituting the stored first melody data is converted into the selected second scale to generate the second melody data, and the generated second melody data is reproduced. As a result, the user can generate a sound that cannot be predicted.

[Brief description of drawings]

FIG. 1 is a perspective view showing an example of the appearance of a pet robot to which the present invention is applied.

FIG. 2 is a block diagram showing an example of an internal configuration of the pet robot of FIG.

FIG. 3 is a block diagram showing an example of a functional configuration of the pet robot of FIG.

4 is a diagram schematically showing an example of functions of an instinct / emotion management unit in FIG.

5 is a diagram schematically showing an example of functions of an action management unit shown in FIG.

FIG. 6 is a diagram showing an example of a finite state automaton.

FIG. 7 is a diagram showing an example of a state transition probability.

8 is a flowchart illustrating a process of the pet robot of FIG.

FIG. 9 is a diagram illustrating a process of extracting a pitch frequency.

FIG. 10 is a diagram showing an example of correspondence between musical scales and pitch frequencies.

FIG. 11 is a diagram showing an example of melody data.

FIG. 12 is a diagram illustrating a process of converting melody data.

FIG. 13 is a diagram illustrating another process of converting melody data.

FIG. 14 is a diagram illustrating still another process of converting melody data.

FIG. 15 is a diagram illustrating a process of converting another piece of melody data.

FIG. 16 is a diagram illustrating a process of converting still another melody data.

FIG. 17 is a diagram for explaining transition of musical scales.

FIG. 18 is a flowchart illustrating another process of the pet robot of FIG. 1.

FIG. 19 is a diagram showing an example of correspondence between emotions and musical scales.

FIG. 20 is a diagram illustrating a process of converting melody data.

FIG. 21 is a block diagram showing an example of a personal computer.

[Explanation of symbols]

10 controller, 11 microphone, 14
Speaker, 21 AD conversion unit, 22 voice feature amount analysis unit, 23 voice section detection unit, 24 voice recognition unit,
26 Instinct / Emotion Management Department, 27 Behavior Management Department, 28
Voice pitch analysis unit, 32 melody generation unit, 33 melody data storage unit, 34 voice synthesis unit, 35 DA conversion unit, 81 magnetic disk, 82 optical disk, 8
3 magneto-optical disk, 84 semiconductor memory

─────────────────────────────────────────────────── ─── Continued Front Page (51) Int.Cl. ⁷ Identification Code FI Theme Coat (Reference) G10L 13/00 G10L 3/00 531W 13/06 537G 15/00 J 15/08 R 15/10 9/08 B 15/18 9/00 B 5/04 F 3/00 531N (72) Inventor Ryo Hanya 6-735 Kita-Shinagawa, Shinagawa-ku, Tokyo Sonomi Co., Ltd. F-term (reference) 2C150 BA11 CA02 DA05 DA24 DA27 DA28 DF03 DF04 DF06 DF08 DF33 ED42 ED47 ED52 EF07 EF16 EF17 EF23 EF28 EF29 EF33 EF36 3C007 AS36 CS08 KS11 KS31 KS39 MT14 WA04 WA14 WB19 WB27 WC07 WC30 5D015 AFF09 5H003 KK

Claims (13)

[Claims] 1. A storage means for storing first melody data composed of a first scale, an extraction means for extracting a pitch frequency of an input voice, and the pitch frequency extracted by the extraction means. Selecting means for selecting a second scale based on the second scale selected by the selecting means, and the first scale constituting the first melody data stored in the storage means. A robot apparatus comprising: a generation unit that converts and generates second melody data; and a reproduction unit that reproduces the second melody data generated by the generation unit. 2. The robot apparatus according to claim 1, wherein the extraction means extracts a pitch frequency of the voice that represents a predetermined keyword. 3. The robot apparatus according to claim 1, wherein the extraction unit extracts a pitch frequency of the voice including a vowel. 4. The robot apparatus according to claim 1, wherein the extracting unit extracts an average value of pitch frequencies detected in a predetermined period of the voice as the pitch frequency. 5. The robot apparatus according to claim 1, wherein the generating means further transits the second scale by a predetermined number of octaves to generate the second melody data. 6. A storage step of storing first melody data composed of a first scale, an extraction step of extracting a pitch frequency of an input voice, and the pitch extracted by the processing of the extraction step. A selection step of selecting a second scale based on a frequency, and the first step stored by the processing of the storage step.
Generating the second melody data by converting the first scale constituting the melody data of the above into the second scale selected by the processing of the selecting step; and generating by the processing of the generating step. A reproducing step of reproducing the generated second melody data, the robot controlling method. 7. A storage control step of controlling storage of first melody data composed of a first scale, an extraction step of extracting a pitch frequency of input voice, and a processing of the extraction step. A selection step of selecting a second scale based on the pitch frequency, and a process of the selection step of selecting the first scale constituting the first melody data stored by the processing of the storage control step. And a reproduction control step of controlling reproduction of the second melody data generated by the processing of the generation step. A recording medium having a computer-readable program recorded thereon. 8. A storage control step of controlling storage of first melody data composed of a first scale, an extraction step of extracting a pitch frequency of an input voice, and a extraction step performed by the processing of the extraction step. A selection step of selecting a second scale based on the pitch frequency, and a process of the selection step of selecting the first scale constituting the first melody data stored by the processing of the storage control step. And a reproduction control step of controlling reproduction of the second melody data generated by the processing of the generation step. A program that causes a computer to execute. 9. A storage means for storing first melody data composed of a first scale, a management means for managing its own internal state, and a change in the internal state managed by the management means. , Selecting means for selecting a second scale corresponding to the change in the internal state, and selecting the first scale constituting the first melody data stored by the storage means by the selecting means. A robot apparatus comprising: a generation unit that converts the second melody data into the second scale to generate second melody data; and a reproduction unit that reproduces the second melody data generated by the generation unit. . 10. The generating means sets the second scale as:
The robot apparatus according to claim 9, wherein the second melody data is generated by further transitioning by a predetermined number of octaves. 11. A storage step of storing first melody data composed of a first scale, a management step of managing its own internal state, and a change of the internal state managed by the processing of said management step. The selection step of selecting a second scale corresponding to the change in the internal state, and the first step stored by the processing of the storage step.
Generating the second melody data by converting the first scale constituting the melody data of the above into the second scale selected by the processing of the selecting step; and generating by the processing of the generating step. A reproducing step of reproducing the generated second melody data, the robot controlling method. 12. A storage control step of controlling storage of first melody data composed of a first scale, a management step of managing an internal state of itself, the internal section managed by processing of the management step. A selection step of selecting a second scale corresponding to the change of the internal state when the state changes, and the first scale that constitutes the first melody data stored by the processing of the storage control step. To a second scale selected by the process of the selecting step to generate second melody data; and reproduction of the second melody data generated by the process of the generating step. A recording medium having a computer-readable program recorded thereon, which comprises a reproduction control step for controlling. 13. A storage control step of controlling storage of first melody data composed of a first scale, a management step of managing an internal state of itself, the internal section managed by processing of the management step. A selection step of selecting a second scale corresponding to the change of the internal state when the state changes, and the first scale that constitutes the first melody data stored by the processing of the storage control step. To a second scale selected by the process of the selecting step to generate second melody data; and reproduction of the second melody data generated by the process of the generating step. A program that causes a computer to execute a reproduction control step for controlling.