JP2003088686A - Nodding robot - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a nodding robot which emits sounds for nodding while emitting voices similar to those of relevant persons to users corresponding to the contents of the voices when emitted by the users. SOLUTION: There are provided a first input part, a receiving part 10, a voice recognition part 20, a memory part 26, and a control part 13. When the first input part is operated to get the voices of the users inputted into the receiving part 10, the voice recognition part 20 recognizes the voices and the control part 13 makes the memory part 26 store registration data of the voices recognized. A drive part 22 for driving the head part of the robot is provided and as nodding sounds are outputted, the control part 13 causes the drive part 22 to be driven, thereby making the head part perform nodding actions.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to an Azuma robot.

[0002]

2. Description of the Related Art Conventionally, a robot of this type is disclosed in, for example, Japanese Patent Application Laid-Open No. 8-187364. According to this publication, based on the relative position of the movable part of the user detected by the detection part 1, the control part 2 moves the driving part of the robot to a predetermined relative position corresponding to the relative position, Instruct the drive unit 3. With the above configuration, the robot performs the same operation as the operation of the movable part of the user and can imitate the user.

[0003]

The above-mentioned robot has the first drawback that it does not give a response voice when the user gives a voice. In particular, when the user is a person with dementia, the same word can be repeatedly spoken, but the robot has a second drawback in that it does not interfere.

In order to solve the above drawbacks, the present inventor has
The robot was equipped with a voice synthesizer, and after the user spoke, the robot tried to make a word that hits a hammer.
However, for the user, the voice-synthesized Azuchi has a third drawback in that the sound produced by the machine is the same as the sound produced by the machine.

In view of the above-mentioned conventional drawbacks, the present invention, when the user utters a voice, emits a voice of a hammer corresponding to the content and a person concerned with the user (eg, son).
Provide a robot that produces a voice close to the voice of.

[0006]

In order to solve the above-mentioned problems, the present invention according to claim 1 comprises a first input section, a receiving section, a voice recognition section, a storage section, and a control section. When the first input unit is operated and the user's voice is input to the earpiece, the voice recognition unit recognizes the voice, and the control unit causes the storage unit to perform voice-recognized registration data. Memorize

According to the second aspect of the present invention, when the second input section is provided and the second input section is operated, and the voice of the humming sound by the person related to the user is input to the earpiece, the control section is The azuchi voice is converted to azuchi registration data, and the azuchi registration data is stored in the storage unit.

According to the third aspect of the present invention, the storage section stores each of the entropy registration data in a one-to-one correspondence with each of the registration data recognized by voice.

According to the present invention of claim 4, a transmitter is provided, the controller recognizes the voice input through the receiver, and the controller determines the result of the voice recognition. When it is determined that the registration data matches the registration data stored in the storage unit, the control unit reads the Azuchi registration data stored in the storage unit and corresponding to the registration data, and outputs the Aizuchi voice.

According to the fifth aspect of the present invention, a voice transmitter and a voice synthesizer are provided, and the controller voice-recognizes the voice input through the receiver and determines that the voice does not match the registered data. And, and when it is determined that the result of the voice recognition requires a hammer, the control unit reads the hammer data stored in the storage unit, through the voice synthesis unit and the transmitter, The Azumi sound based on the data is output.

According to the sixth aspect of the present invention, a drive unit for driving the head is provided, and the control unit drives the drive unit to drive the drive unit, and the control unit drives the head to perform the control operation. Let

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, a hammer robot 1 according to an embodiment of the present invention will be described with reference to FIGS. 1 and 2. Figure 1
Is a block diagram of the Azuma robot 1, and FIG. 2 is an external view of the Azuma robot 1.

In these figures, the hammer robot 1 has, for example, a head 2, ears 3, 4, a body 5, legs 6,
7, 8 and 9 (the legs 8 and 9 are not visible in FIG. 2) and the like. These parts 2 to 9 are made of, for example, hollow plastic or the like.

The earpiece 10 comprises, for example, a microphone and is arranged inside the ears 3 and 4. The receiver 10 converts the input voice into an electric signal and outputs it to the CODEC 11.

The CODEC 11 comprises, for example, a baseband circuit, an AD converter, and an AF circuit (both not shown).

The transmitter 12 is composed of, for example, a speaker,
The electrical signal from the CODEC 11 is converted into voice and output.

The control unit 13 is composed of, for example, a CPU, and D
An SP (Digital Signal Processor) 14 is built in. The CODEC 11 is connected to the control unit 13.

The display unit 15 is composed of, for example, a liquid crystal display device and is connected to the control unit 13. The input unit 16 includes, for example, a first input unit 17, a second input unit 18, and a power button 19.
Etc., and is connected to the control unit 13.

The input voice is converted into an electric signal by the receiving unit 10, and the voice recognition unit 20 voice-recognizes the electric signal via the CODEC 11 and the control unit 13.

The voice synthesizer 21 is connected to the controller 13,
Audio is output via the control unit 13, the CODEC 11, and the transmitter unit 12.

The drive unit 22 is, for example, a swinging motor, and is connected to the control unit 13 via a motor driver (drive circuit) 23.

The head portion 2 is connected to the body portion 5 via a hinge or the like (not shown), and is configured, for example, so that it can move back and forth (bow). The drive unit 22 is connected to the head 2 via, for example, a link mechanism. Drive unit 22
When a predetermined voltage is applied to the drive unit 22, the drive unit 22 rotates, the link mechanism operates, and the head 2 moves back and forth so that the heads 2 cooperate with each other. In this way, the drive unit 22 drives the head 2.

The ROM 24 stores the control program of the control unit 13 and is connected to the control unit 13. RA
The M25 stores data necessary for the operation of the Azuma robot 1, and is connected to the control unit 13.

The storage unit 26 is, for example, a hard disk driver and is connected to the control unit 13. Storage unit 26
Stores, as will be described later, the Azuma data, the registration data, the Azumi registration data, and the like.

The power supply unit 27 is composed of, for example, a secondary battery, and supplies a predetermined voltage to the control unit 13, the motor driver 23, and the like. The above-mentioned components make up the Azuma robot 1.

Next, referring to FIGS. 1 to 4, the operation of the hammer robot 1 will be described. FIG. 3 is a flow chart for explaining the main operation of the Azuma robot 1. Figure 4
FIG. 4A shows an example of the matching table 28 stored in the storage unit 26, and FIG. 4B shows an example of the voice registration table 29 and the matching registration table 30 stored in the storage unit 26.

As shown in FIG. 4 (a), the summation table 28 of the storage unit 26 has, for example, 28a and 28.
b, yeah so 28c, then 28d, good 28
The summation data (digital value) such as e is registered in advance.

First, it is assumed that a person B (for example, an old man with dementia) and a person B (for example, the son of the old man) of the user A (for example, a demented old man) continue to press the first input unit 17 provided in the Aizuchi robot 1 (FIG. reference). At this time, the control unit 13 of the Aizuchi robot 1
Affirms the determination as to whether or not the voice registration button (first input unit 17) is ON (see S1 in FIG. 3).

Next, the user A speaks his / her favorite words (for example, "good old days") to the receiver 10.
Then, the control unit 13 causes the voice recognition unit 20 to recognize the voice of the user A by voice. The DSP 14 of the control unit 13 compresses the voice-recognized electric signal, digitizes it, and stores it in the storage unit 26 as registration data 29a (see FIG. 4B) (see S2 in FIG. 3).

In other words, the above operation is performed by the first input section 17
When is operated and the voice of the user A is input to the earpiece 10, the voice recognition unit 20 recognizes the voice. Then, the control unit 13 causes the storage unit 26 to store the voice-recognized registration data 29a.

Next, for example, the related person B is the first input unit 1
It is assumed that pressing 7 is stopped and the second input unit 18 is continuously pressed (see FIG. 2). At this time, the control unit 13 affirms the determination as to whether or not the sumobu registration button (second input unit 18) is ON (see S3 in FIG. 3).

Next, the party B heads to the receiver 10.
Speaking the voice of Aizuchi for the voice of the user A (for example, "yes"). Then, the DSP 14 compresses and digitizes the electric signal obtained by converting the above-mentioned azure voice by the transmitter 12 or the like, and outputs the azuchi registration data 30a (FIG. 4 (b)).
Reference). The control unit 13 causes the storage unit 26 to store the Atsuuchi registration data 30a (see S4 in FIG. 3).

In other words, the above operation is performed by the second input section 18
Is operated, and the voice of a hammer (that's right) by the person B related to the user A is input to the earpiece 10. The control unit 13 converts the azuchi voice into the azuchi registration data 30a, and the storage unit 26
On the other hand, the Atsuuchi registration data 30a is stored.

In this way, the voice registration table 29 and the Atsui registration table 30 are completed by repeating the operations of S1 to S4 a predetermined number of times.

At this time, the storage unit 26 stores the corresponding entropy registration data 30a to 30c in a one-to-one correspondence with the respective speech-recognized registration data 29a to 29c. In other words, the storage unit 26 stores the Azuchi registration data 30a corresponding to the registration data 29a, stores the Azuchi registration data 30b corresponding to the registration data 29b, and registers the registration data 29c.
Corresponding to, the Atsuuchi registration data 30c is stored.

The control unit 13 determines whether or not there is an input voice (see S5 in FIG. 3). If the user A is speaking something, the control unit 13 affirms the determination in S5.

Then, the control unit 13 causes the voice recognition unit 20 to voice-recognize the input voice (see S6 in FIG. 3).

Next, the control unit 13 controls the registration data 2 in which the voice-recognized electric signal is stored in the storage unit 26.
It is determined whether or not it matches any of 9a to 29c (see S7 of FIG. 3).

If the voice of the user A input to the earpiece 10 is "I wish I used to be good", the controller 13
Affirms the determination at S7.

Then, the control unit 13 causes the storage unit 26 to execute the matching registration data 3 corresponding to the registration data 29a.
Read 0a. Then, the DSP 14 of the control unit 13 reproduces the Azuchi registration data 30a, converts it into an electric signal, and outputs the Aizuchi voice "Soudane" via the CODEC 11 and the transmitter 12 (see S8 in FIG. 3). ).

In other words, the control section 13 puts the voice "I wish it was good" into the voice input through the earpiece section 10.
Voice recognition. When the control unit 13 determines that the result of the voice recognition matches the registration data 29a stored in the storage unit 26, the control unit 13 is stored in the storage unit 26 and corresponds to the registration data 29a. The registration data 30a is read. Then, the control unit 13 converts the Azuchi registration data 30a and outputs the Aizuchi voice "Yes, it is".

Next, the control unit 13 causes the summing operation (see S9 in FIG. 3). That is, the control unit 13 outputs the above-mentioned hammer sound, drives the driving unit 22, and causes the head 2 to perform the hammer motion (for example, move the head 2 up and down).

Further, in relation to S7 of FIG. 3, if the voice of the user A input to the receiving section 10 is different from the registered data 29a to 29c, the control section 13 will be described.
Denies S7.

Next, the control unit 13 determines whether or not the input voice of the user A requires the Azumi voice (see S10 in FIG. 3). That is, the control unit 13 determines whether or not the input voice of the user A is a word that asks for the consent of the other party, or the ending of the word ...

If the input voice is a word requesting the consent of the other party, the control unit 13 affirms S10. Then, the control unit 13 obtains the interaction data 28a to 28e in the interaction table 28 stored in the storage unit 26.

For example, a random number table may be provided, and the storage unit 26 may randomly select the ensemble data 28a to 28e. For example, it is assumed that the auction data 28e is selected.

Then, the control unit 13 outputs the voice of "Waitane" which is based on the voice data 28e through the voice synthesizer 21, the CODEC 11 and the transmitter 12 (see S11 of FIG. 3).

In other words, the control unit 13
Causes the voice input through the receiver 10 to be recognized. When the control unit 13 determines that the result of the voice recognition does not match the registration data 29a to 29c, and determines that the result requires a match, the control unit 13 stores the match data stored in the storage unit 26. (Eg, 28e) is read. Then, the control unit 13 controls the voice synthesis unit 21 and the COD.
Via the EC 11 and the transmitter 12, the above-mentioned summation data 28e
The Aizuchi sound based on is output.

Next, the control unit 13 performs the summing operation (see S9 in FIG. 3) and ends the series of operations. In addition, the control part 13 will complete | finish operation | movement, if the determination of S10 is denied.

[0050]

According to the present invention of claim 1, a first input section,
A receiver section, a voice recognition section, a storage section, and a control section,
When the first input unit is operated and the voice of the user is input to the earpiece, the voice recognition unit recognizes the voice, and the control unit registers the voice-recognized registration in the storage unit. It is configured to store data. Thus, when the first input section is operated, the voice-recognized registration data is stored, so that only the voice of the user can be accurately registered. Furthermore, since voice recognition is performed, correct registration data can be stored even if the user's voice is somewhat vague.

According to the present invention of claim 2, when the second input section is provided and the second input section is operated and the competing voice by the person concerned of the user is inputted to the earpiece section, the control section is It is configured such that the azuchi voice is converted into azuchi registration data and the azuchi registration data is stored in the storage unit.
In this way, when the second input unit is operated, the Atsuuchi registration data is stored, so that only the Atsuuchi voice of the person concerned can be accurately registered. In addition, since the data is stored without voice recognition, it is possible to store the enrollment registration data based on the live voice of the person concerned.

According to the third aspect of the present invention, the storage section is configured to store each of the entropy registration data in a one-to-one correspondence with each of the registration data recognized by voice.
In this way, since the registration data and the Azuma registration data are stored in a one-to-one correspondence, by storing the Atsuuchi registration data that is most suitable for the specific registration data, the most suitable (according to the content issued by the user You can hit a hammer.

According to the present invention of claim 4, a transmitter is provided, the controller recognizes the voice input through the receiver, and the controller determines the result of the voice recognition. When it is determined that the registration data matches the registration data stored in the storage unit, the control unit reads the Azuchi registration data stored in the storage unit and corresponding to the registration data, and outputs the Aizuchi voice. To do. With the above configuration, when the user utters a voice, a person concerned with the user (eg son)
The return of the Azuma sound, which is similar to the real voice, will increase user satisfaction.

According to the fifth aspect of the present invention, a voice transmitter and a voice synthesizer are provided, and the controller voice-recognizes the voice input via the receiver, and determines that the voice does not match the registered data. And, and when it is determined that the result of the voice recognition requires a hammer, the control unit reads the hammer data stored in the storage unit, through the voice synthesis unit and the transmitter, The configuration is such that the Azumi sound based on the data is output. In this way, even if the voice of the user is not registered, the response is automatically returned depending on the state of the voice of the user, so that the satisfaction of the user is further enhanced.

According to the sixth aspect of the present invention, a drive unit for driving the head is provided, and while the sound of the hammer is output, the control unit drives the driver to move the head to move the hammer. It is configured so that In this way, the Aizu robot outputs the Aizu sound and performs the Aizu operation, so that the user gives the user an atmosphere and familiarity as if he were an acquaintance.

[Brief description of drawings]

FIG. 1 is a block diagram of a hammering robot 1 according to an embodiment of the present invention.

FIG. 2 is an external view of the Atsuuchi robot 1.

FIG. 3 is a flow chart showing the main operation of the above-mentioned Atsuuchi robot 1.

FIG. 4 (a) shows a play table in a storage unit 26 used in the play robot 1, and FIG. 4 (b) shows a voice registration table and a play registration table stored in the storage unit 26.

[Explanation of symbols]

10 Earpiece 13 Control unit 20 Speech recognition unit 26 Memory

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁷ Identification code FI Theme Coat (reference) G10L 15/22 G10L 3/00 521C (72) Inventor Miwa Yoshishige 3-201 Minamiyoshikata, Tottori City, Tottori Prefecture Tottori Sanyo Electric Co., Ltd. F term (reference) 2C150 BA11 CA01 CA02 CA04 DA24 DA26 DA27 DA28 DF03 DF04 DF06 DF33 ED42 ED47 ED52 EF16 EF17 EF23 EF29 EF33 EF36 5D015 KK02 KK04 LL06 5D045 AB11

Claims (6)

[Claims] 1. A first input unit, a receiving unit, a voice recognition unit, a storage unit, and a control unit, wherein the first input unit is operated and the voice of the user is input to the receiving unit. Then, the voice recognition unit voice-recognizes the voice, and the control unit causes the storage unit to store the voice-recognized registration data. 2. A second input unit is provided, and when the second input unit is operated and an auction voice by a person related to the user is input to the earpiece unit, the control unit registers the auction voice in the auction. The auspices robot according to claim 1, wherein the auspices registration data is converted into data and stored in the storage unit. 3. The summing robot according to claim 2, wherein the storage unit stores the respective engraving registration data on a one-to-one basis corresponding to each of the registration data recognized by voice. 4. A voice transmitter is provided, the controller recognizes voice input through the receiver, and the controller stores the result of the voice recognition in the storage. 4. The control unit, when determining that the registration data matches the registration data, reads the Azuchi registration data stored in the storage unit and corresponding to the registration data, and outputs the Aizuchi voice. Robot of Azuma. 5. A voice transmitting unit and a voice synthesizing unit are provided, and the control unit causes the voice input through the voice receiving unit to be voice-recognized, determines that the voice does not match the registered data, and outputs the voice. If you determine that the recognized result requires a hammer,
2. The robot according to claim 1, wherein the control unit reads the read data stored in the storage unit, and outputs a read voice based on the read data via the voice synthesis unit and the transmission unit. 6. A drive unit for driving a head is provided, and the control unit drives the drive unit so that the head is driven to perform the reciprocating motion while the drive sound is output. The Atsuchi robot according to claim 4 or claim 5.